COLUMBIA LIBRARIES OFFSITE HEALTH ScYeNCES STANDAHD HX641 37660 RC857 RS7 The pancreas: Its s Columbia Zinitiersfitp College of ^fjpgiciang anb ^urgeonjJ i^eference Hibvaxp 2>^ wl 3.ir Digitized by the Internet Archive in 2010 with funding from Open Knowledge Commons http://www.archive.org/details/pancreasitssurgeOOrobs 'A-/;^.^^ THE PANCREAS ITS SURGERY AND PATHOLOGY A. W. MAYO ROBSON, D.Sc. (Leeds), F.R.C.S. (Eng.) LONDON P. J. CAMMIDGE, M.B. (Lond.), D.RH. (Camb.) LONDON ILLUSTRATED PHILADELPHIA AND LONDON W. B. SAUNDERS COMPANY 1907 Copyright, 1907, by W. B. Saunders Company Registered at Stationers' Hall, London, England PRINTED IN PHILADELPHIA PREFACE Our present knowledge of the physiology, pathology, and surgery of the pancreas, like so many other advances in medicine in recent years, was rendered possible by the beneficent work of Lister. So long as clinical observa- tion was only capable of being checked by the experience of the post-mortem room, and by an occasional accidental experiment on the living subject, the important and com- plex part that the pancreas plays in the physiology of the body remained unsuspected, and descriptions of the diseases to which it is liable were confined to a few lines on malignant disease, cysts, and calculi. Animal ex- periments, now rendered safe by antiseptic surgery and improved technique, have thrown a flood of light on the physiology of the organ and elevated it from the position of a mere accessory digestive gland to the rank of a struc- ture indispensable for the metaboHc needs of the organ- ism. The numerous laparotomies undertaken in recent years have afforded the surgeon opportunities of observing and handling the living organ, both in health and disease, and a comparison of the conditions noticed, together with a closer investigation of the symptoms and after-histories of the cases, has very, considerably widened our concep- tion of the pathological changes that may occur in the gland and afforded a basis for their clinical differentiation. Histological and post-mortem inquiries, stimulated by the impetus thus given, have still further increased our knowledge, and confirmed the conclusions of the bedside and the operating theatre. There are as yet many points on which observers are not agreed, and there are questions 9 lo Preface which still call for elucidation, but the enormous literature of to-day as compared with that prior to 1886, when Pro- fessor Senn of Chicago published his valuable experimental work on the pancreas, shows the great advances that have been, and are being, made in the subject. The symptomatology and pathology of the pancreas are so intimately bound up with the physiology and anatomy of the gland, and these again are rendered so much more easily understood if the comparative anatomy and development of the organ are borne in mind, that we have prefaced the pathological and clinical sections of this work with a brief, but we hope sufficiently compre- hensive and accurate, account of those subjects. In the chapter on histology, and again later under the heading of diabetes, we have discussed the structure and supposed functions of those characteristic groups of cells known as the islands of Langerhans. In doing so we have endeavoured to impartially summarise the evidence for and against the contending views that are held with regard to them, but, as will be gathered from the text, we personally are of opinion that the balance of available evidence strongly points to their being independent struc- tures related to the control of carbohydrate metabolism within the body that the pancreas undoubtedly exerts. A thorough comprehension of the chemical changes induced in the body by diseases of the pancreas would include a knowledge of the pathology of diabetes, but at present we are still in the dark as to the true essentials of that condition. Our description of the chemical pathology of the pancreas is therefore largely confined to the condition of the urine and fseces found to accompany disease of the gland, and although we have now devoted special attention to this subject for six or seven years, we are conscious that as yet only the fringe has been touched upon. The so-called "pancreatic" reaction in the urine is still under investigation. The improved Preface 1 1 method described in these pages is undoubtedly a distinct advance on the original process described in the Arris and Gale lecture of 1904, but it is not yet as perfect as it might be. The difficulty of the investigation is considerable, for the quantity of material to be obtained from any one case, even when a well-marked reaction is given, has proved to be small, and it is only by collecting very large amounts of urine from suitable cases, whenever they have occurred, that we have been able to make slow advances. We do not consider that at present we are in a position to make more positive statements than those expressed in the chapter on chemical pathology, but we hope that w^e may shortly be able to do so. In our own practice we never rely upon the "pancreatic" reaction alone in making a diagnosis of pancreatitis or malignant disease of the pancreas, but always take into account the results of a complete analysis of the urine and a chemical examination of the faeces, as well as the clinical symptoms ; it is from neglect of these precautions, and under the false notion that the "pancreatic" reaction was claimed to be pathognomonic, that the mistakes made by some writers .have arisen. The examination of the faeces often gives im- portant confirmation of the presence or absence of disease of the pancreas, but this is not always the case, and the possible causes of unexpected results described in the text have always to be borne in mind. The question of the cause of the absence of colour in the stools in various patho- logical conditions has excited attention for many years, and, although we do not suggest that the explanations our investigations and observations have enabled us to make are true of all cases in which the faeces are white, they appear to be so for pancreatic disease. There seems to be an impression in the minds of nearly all members of the profession that diseases of the pan- creas, excepting some of the grosser lesions, are unrecog- nisable during life, but we venture to think that a careful 12 Preface perusal of the chapter dealing with general symptoma- tology and diagnosis will show that while no single sign or symptom is characteristic of disease of the pancreas, no more than of any other organ, the cumulative evidence to be obtained by a careful investigation of the history, clinical symptoms, and signs, and the indications to be obtained by the methods of the laboratory, should leave no doubt as to the presence or absence of pancreatic trouble in any particular case and, in the large majority, allow of a definite opinion as to the nature of the lesion being arrived at. The classification of inflammatory lesions of the pan- creas is the same as that outlined in the Hunterian Lectures of 1904. Increased experience has only served to demonstrate its clinical utility, and its adoption by subsequent writers shows that they recognise the numer- ous forms that inflammation of the pancreas may assume under various conditions. We have emphasised the in- timate etiological relation existing between gall-stones and pancreatitis and pointed out the conditions under which biliary calculi in the common bile-duct are likely to cause, and will fail to give rise to, pancreatic inflamma- tion. We have also laid stress upon the no less important, but less commonly recognised, association of inflammation of the pancreas with catarrhal conditions of the upper part of the gastro-intestinal tract. Pancreatitis resulting from a duodenal catarrh may, under certain conditions, give rise to more or less persistent jaundice, and, in our ex- perience, is the most common cause of the conditions usually known as acute and chronic "catarrhal" jaundice. In dealing with the subject of diabetes we have devoted much space to a consideration of its relations to the pan- creas, and have quoted the more important experimental evidence and clinical work bearing upon the subject. Between the ofttimes conflicting, and even contradictory, statements of different authors it is difficult to arrive at Preface 13 any very definite conclusions as to the frequency of pan- creatic lesions in diabetes and as to how these are related to the disease, but one fact that has been clearly estab- lished is that a small portion of normal gland is capable of averting the onset of the condition. It is therefore important that diseases of the pancreas should be recog- nised at the earliest possible moment, and that conditions likely to give rise to pancreatic lesions should be radically treated before they have had time to bring about perma- nent, and may be progressive, injury of the gland. For this reason we strongly advocate the early treatment of gall-stones, especially when they are present in the com- mon duct and an examination of the urine and faeces shows that a pancreatic lesion exists. The very striking increase in the death-rate from diabetes shown by the Registrar-General's returns is possibly not unconnected with the greater prevalence of digestive disturbances in recent years, and we therefore think that duodenal catarrh and the frequently associated catarrhal pancreatitis always call for prompt attention. One of the most important practical results that has followed from modern observations on the pancreas is the recognition of the very close similarity of the symp- toms of cancer and chronic pancreatitis in the head of the gland. Many cases of the latter have in the past been allowed to die unoperated on, under the mistaken impres- sion that they were suffering from cancer, and our ex- perience would suggest that, even at the present time, there are many who do not realize the importance of a differential diagnosis between the two conditions. We have dealt with this subject under the headings of chronic pancreatitis and cancer, and it is also referred to in the chapters on pathology and symptomatology. While a considerable number of the illustrations in this work are original and have been taken from prepara- tions in our own possession, we are deeply indebted to 14 Preface the various museums mentioned for permission to have drawings or photographs made from their specimens. Our thanks are also due to the authors and publishers, to whom acknowledgments are made, for the pictures appearing above their names. To the end of each chapter we have appended a list of the more important papers and publications bearing on the subjects therein discussed. These do not make any pretense at completely exhausting the bibliography, but as a rule merely represent an alphabetical list of the authors mentioned in the text. In writing this work we have laid under contribution all the monographs available to us, as well as those papers that have appeared on the subject in the current literature. We have endeavoured as far as possible to credit each author with the views and cases of Avhich he has written, and are particularly in- debted to Opie, Oser, Rolleston, and Flexner. A certain number of illustrative cases and illustrations have also been taken from a work on " Diseases of the Pancreas " by A. W. Mayo Robson and B. G. A. Moynihan (Saunders & Co.), which has been for some time out of print. Each case and opinion has, as far as possible, been attributed to its original author, but should we inadvertently have misrepresented any of the writers quoted, or attributed a view or opinion to some other than its original author, we crave forgiveness and ask for correction. A, W. Mayo Robson. P. J. Cammidge. London, Augtist, igoy. CONTENTS CHAPTER I Page Comparative Anatomy 17 CHAPTER II Anatomy 28 CHAPTER III Embryology 41 CHAPTER IV Anatomical Anomalies 46 CHAPTER V Surgical Anatomy 64 CHAPTER VI Histology 71 CHAPTER VII Physiology 96 CHAPTER VIII Pathology 126 CHAPTER IX Fat Necrosis 190 CHAPTER X Chemical Pathology 206 CHAPTER XI Diabetes 269 CHAPTER XII General Symptomatology and Diagnosis 311 CHAPTER XIII Injuries 345 CHAPTER XIV Inflammatory Affections of the Pancreas 361 15 1 6 Contents CHAPTER XV Page Acute Pancreatitis and Subacute Pancreatitis 384 CHAPTER XVI Chronic Pancreatitis 412 CHAPTER XVII Pancreolithic Catarrh and Pancreatic Calculi 473 CHAPTER XVIII Pancreatic Cysts 487 CHAPTER XIX Neoplasms 512 Index of Authors 529 l^OEX 537 THE PANCREAS ITS SURGERY AND PATHOLOGY CHAPTER I COMPARATIVE ANATOMY In unicellular organisms all the activities of life are embraced within the compass of a simple unit of living matter; it moves by contracting its substance, it draws back from hurtful influences, it absorbs oxygen, it en- gulfs and digests food, and gets rid of the waste products of its metabolism. But, early in the communal life of the multicellular metazoa, a tendency is seen to limit the physiological activities of the groups of cells, and to con- centrate special functions in particular areas. As we advance up the scale of life this tendency becomes more and more marked, and these aggregations of cells, set apart to subserve particular purposes in the economy of the body, are differentiated as distinct organs. The more complicated and active the life of the animal, the more numerous are its organs, for, by the concentration of the activities of the cells on the performance of some special work, energy is economised, in the same way as division of labour in the commercial world is found to contribute to economy of production. One of the most primitive areas to be set apart for the performance of a special function is that which deals with the absorption of food. First distinctly seen as a mere folding in of the surface to form a pouch, it assumes, 2 17 1 8 The Pancreas: Its Surgery and Pathology in most of the metazoa, the form of a canal passing through the tissues and connected with the external world by an apeture at either extremity. The cells lining the alimen- tary area, in its simplest form, show but little differentia- tion of structure, and, although they are predominantly digestive in character, they have not lost the primitive and many-sided qualities of the protozoon, as is shown by the ease with which a change of environment will bring about a change of function, so that those which were pre- viously subsidiary become predominant. In the higher forms the characters of the cells become more fixed, their structure is modified to suit their special work and sur- roundings, and their power of reverting to the primitively complex physiological, but anatomically simple, type is lost. Up to the echinoderms there is no indication of a par- ticular concentration of the digestive powers in any one part of the alimentary tract. The walls may be pouched or ridged, to increase the surface, and one portion may be more muscular or harder than another, in order that the food may be better ground or mixed, but the whole ex- tent of the tract is in contact with its contents, and no part is to be distinguished as especially set apart for the elaboration of digestive ferments. In a type such as the starfish, however, such glandular structures can be recog- nised. In this animal five branches are given off from the pyloric portion of the stomach, and each of these di- vides into two large digestive cseca. The glands them- selves do not come into contact with the food, but secrete a ferment, which is said to have tryptic, peptic, and diastatic powers, thus showing the first definite step in the further differentiation of the functions of digestion and absorption. Most of the Crustacea, Insecta, and Mollusca possess one or 'more pairs of similar digestive caeca, or glands, which join the alimentary tract in the region of the mid- Comparative Anatomy 19 gut, or stomach. Their function is still complex, and no advance to a higher and simpler physiological type than that seen in the starfish is to be recognised. In the lower members of the vertebrate series a similar state of things is found, and even in certain fishes the length of the intestine, a thickening of the mucous mem- brane of the duodenum, or the inactive nature of the species renders any special di- gestive glands unnecessary. In most osseous fishes, however, there is a well-defined gland having the characters and func- tions of a liver, and in addition an extension of the secreting surface of the intestine by the presence of a number of long, slender pouches, which are con- nected with the commencement of the duodenum. These differ in length and width, and, while the widest are sometimes found to be filled with the same con- tents as the intestine, the nar- rowest serve only as secreting organs, and are apparently specialised for the elaboration of digestive ferments. Their number and arrangement vary in different types. In the sand- lance there is only one, but in the whiting and salmon there are a hundred or more. In the herring, haddock, and salmon they are disposed in a line along the whole length of the duodenum, while in the whiting they are arranged in a circle around the distal end of the pylorus. A tendency to concentrate these intestinal outgrowths into a more typical glandular structure is seen in some Fig. I. — Pyloric ap- pendages of the salmon (after Giinther). 20 The Pancreas: Its Surgery and Pathology members of the class, especially in the more active types, in which rapid and complete digestion is a necessity. The 50 caeca of the pilchard open into the duodenum by 30 orifices, but the 120 of the whiting progressively unite into four or five groups, each communicating with the duodenum by a single duct. The swordfish has but two openings, and in the sturgeon a single wide duct terminates by a papilla on the internal sur- face of the duodenal wall, close to the ductus choledochus. The long, slender, ramified caeca are, in the two last, bound loosely together by connective tissue and possess a rich vascular sup- ply, the whole being enclosed in a capsule, thus corresponding in structure to a conglomerate gland of the type of the pan- creas. In many fishes that have typical pyloric caeca, and in some that do not possess these structures, there exists a conglomerate glandular organ opening by a duct into the duodenum, which is apparently the true homologue of the pan- creas in air-breathing animals. A large -lobed structure of this description, the duct of which is so intimately connected with the bile-duct that both appear as a single structure externally, is found in the salmon. In the catfish it is very large, and the bile-duct passes through its substance. In the plaice, flounder, gar-pike, etc., it is situated in the mesentery and is smaller, but its duct in each instance accompanies the terminal portion of the ductus chole- Fig. 2. — Alimentary canal of the whiting, showing the arrangement of the pyloric caeca (after Owen). Comparative Anatomy 21 dochus. The characters of this organ in the sturgeon have already been referred to. A similar glandular mass of considerable size, lying behind the stomach, and close to the spleen, is met with in sharks and other elasmo- branchs. The liver in reptiles is proportionally large, and they To. Fig. 3. — Pancreas and spleen of the turtle (after Owen). possess a distinct and well-defined pancreas. The lat- ter is a yellow or pink gland, consisting of many acini, each opening into a small duct. These ducts unite into larger ducts, and these again form a common channel, which opens into the intestine with, or close to, the bile- duct. The acini round the smaller ducts are aggregated 22 The Pancreas: Its Surgery and Pathology together to form lobules, and the lobules are again col- lected into lobes. In some members of the group the pancreas is spread out in the duodenal mesentery, but in most serpents and lizards it has a compact form; in the crocodile it is divided into two elongated lobes, and sometimes communicates with the duodenum by two distinct ducts ; in the turtle the pan- creatic duct terminates by a papilla opening into the expanded end or ampulla of the bile-duct. The pan- creas of the carnivorous types of the reptilia is more bulky and compact, forming a larger proportion of the total weight of the animal than in the vegetable-feeders. The pancreas of birds is usually firmer than that of reptiles. It is also relatively larger, probably to compensate for the absence of masti- cation and the salivary digestion of the food. It is long and narrow, lying in the space between the duo- denal loops, and generally consists of two, and sometimes of three, por- tions. It communicates with the in- testine by two, and occasionally by three, separate ducts, which open near the hepatic and cystic ducts. In the mammalia the pancreas is more plainly of the conglomerate type. It is paler, and of a firmer structure, than in birds, and also differs by the development of a part, stretching towards the spleen, which is more or less dis- tinct from that lodged within the duodenal loop. In the simpler members of the series, such as the mar- supials, the gland is bent upon itself, running from the Fig. 4. — Pancreas and duodenum of goose (after Owen). Comparative Anatomy 23 duodenum to the spleen, behind the stomach, and giving off into the duodenal mesentery and omental folds more or less numerous processes. The main duct opens into the bile-duct, the bile and pancreatic secretion reaching the intestine through a common opening. The main mass of the gland in rodents follows the curve of the duodenum, but sends numerous ramifying processes into the mesentery. The main duct, into which the minor fn Fig. 5. — Pancreas of the rat (after Owen). channels collect, enters the duodenum a considerable distance from the point of entry of the bile-duct. In the beaver 18 inches separates the papilla of the biliary pas- sage from that of the pancreatic duct, and the latter is some 21 inches from the pylorus. A small, and usually impermeable, duct, corresponding to the main channel of the pancreas of most of the mammalia, can, in some instances, be made out joining the intestine in the neigh- bourhood of the biliary papilla. The functioning duct 24 The Pancreas: Its Surgery and Pathology in this group probably represents one of the lower mem- bers of the series of digestive cseca found arranged along the length of the duodenum in some fishes. The pancreas of the aquatic mammals (Cetacea) is long, narrow, and compact. It crosses the spine at the root of the mesentery, the left end terminating near the spleen, and the right being expanded and adherent to the curve of the duodenum. The pancre- atic duct joins the bile-duct. The transverse or splenic portion of the gland is still better developed in the un- gulates, and, in this order, forms the larger part of the gland. Its duct, which is separate from that of the duodenal part, joins with the hepatic duct to form an am- pulla before entering the in- testine. The smaller, duode- nal portion of the gland lies at right angles to the trans- verse part. It expands down- ^"^ wards and backwards in the pouchlf the^lrham^'formed ^^odenal mesentery. Its duct by the union of the common enters the duodenum about bile-duct and pancreatic duct ,. j • j_ r j-i (after Owen). "the Same distance from the pylorus as the common bile and pancreatic opening, but by quite a distinct aperture. The divisions of the pancreas in the ruminants are somewhat less well defined, the gland being broader and flatter in character. The long, narrow pancreas of the carnivora shows a well-marked division into splenic and duodenal sections, which are of unequal length. The splenic part is straight Comparative Anatomy 25 and runs transversely across the spine ; the duodenal seg- ment follows the curve of the duodenum. Both are cov- ered by the peritoneum. In most members of the order the ducts of the transverse and descending portions anas- Fig. 7. — Pancreas of the dog, dissected to show the relations of the common bile-duct and pancreatic ducts and their openings into the duodenum. tomose at two points, and the main duct communicates with the bile-duct before entering the duodenum. The pancreas of the cat and dog calls for special men- tion, owing to the frequent use made of these animals in experimental work. The duodenal part of the gland in Fig. 8.- — Dissection of an abnormal pancreas of a dog, showing separate openings for the common bile-duct and two pancreatic ducts. the dog is larger than the splenic portion, which it joins at right angles. As a rule, the smaller duct joins with the common bile-duct within the walls of the duodenum, but is externally quite distinct. The larger then enters the 26 The Pancreas: Its Surgery and Pathology bowel half an inch or more below. Occasionally the bile- duct and two pancreatic ducts have separate openings, as in the case of a dog dissected by one of us. In the cat there is a large duct, communicating with both sections of the gland, which joins with the bile-duct, and enters the duodenum by a common orifice with it. A smaller duct is also present, which anastomoses with the main channel within the gland, but possesses a separate open- ing into the intestine a short distance below. Occasion- Fig. 9. — Pancreas of the cat, laid open to show the main ducts and their relation to a large vein near the junction of the two ducts (after DeWitt). ally a lateral reservoir, communicating with the main channel by a short duct just before its junction with the bile-passage, is found. The pancreas of the anthropoidea is less mobile than in any other group of animals, and is found to be more completely applied and fixed to the posterior abdominal wall, the more adapted the animal is to the upright posi- tion. The duodenal part is reduced to an enlargement termed the "head," while the splenic portion narrows at Comparative Anatomy 27 its termination near the spleen to form "the tail." The intervening portion forms "the body" of the gland, along the thick upper border of which run the splenic artery and vein. The main duct traverses the substance of the gland, nearer its lower than its upper border, and usually communicates, near its termination, with the lesser duct which drains the head. The latter may have a separate entrance into the duodenum, placed somewhat nearer the pylorus than the papilla by which the main pancreatic channel and bile-duct open, or it may be obliterated, the whole of the pancreatic secretion, including that from the head, then finding its way into the intestine by way of the common opening. Rachford has pointed out that the nearer an animal approaches to the purely carnivorous type, the more likely are the bile and pancreatic juice to be passed into the intestine through a common opening, and the closer is this opening to the pylorus. Literature Giinther : ' ' The Study of Fishes. ' ' Owen: "Comparative Anatomy and Physiology of Vertebrates." Rachford: Jour, of Physiol., xxv, 165. Schieffer: "Du Pancreas dans la serie animale," Th. Montpellier, 1894. Thompson: "Outlines of Zoology." CHAPTER II ANATOMY The greater part of the pancreas in man lies in the epigastrium, but a portion of the body and the tail extend into the left hypochondrium, and the head may project into the umbilical region. To expose the organ from the front the stomach must be detached from the great omentum and be turned up- wards. It is then seen as a long, pinkish, cream-coloured gland, stretching transversely across the posterior abdom- inal wall, from the concavity of the duodenum to the lower and inner border of the spleen. In the fresh condi- tion it has a firm consistency and a markedly lobulated appearance. In length it varies from 5 to 6 inches (12 to 15 cm.). Its average weight ranges from 2.25 to 3.5 ounces (66 to 102 grams) . The general shape of the gland is aptly compared by Birmingham, in Cunningham's " Text-Book of Anatomy," to the letter J placed upon its side, rH , the loop being thickened to represent the head, the thickened stem corresponding to the body, and the narrow bend joining the two indicating the neck. The enlarged right extremity, or "head," extends down- wards and to the left, lying in the concavity of the duode- num in contact with its second and third parts, and oppo- site to the second and, upper part of, the third lumbar vertebra. The short and comparatively narrow portion of the gland termed the "neck" arises from the upper and right part of the head. It runs upwards and to the left, and, after a course of about one inch, merges into the "body." This, which is the longest section of the gland, runs backwards and to the left at the level of 28 Fig. 10. — Relations of the pancreas (Sobotta and McMurrich). Anatomy 29 the first lumbar vertebra. The pointed left extremity, or "tail," is the least firmly attached portion of the organ. It merges so gradually into the body that no sharp line of distinction can be drawn between the two. The disc-shaped head is flattened from before back- wards. Its right and lower borders are closely united to the duodenum, one-third of the circumference of which may te enveloped by the gland substance in a well- pig_ ji_ — Transverse section of the abdomen at the first lumbar verte- bra, to show the relations of the pancreas (after Braune). developed organ. The right half, above, is continued into the neck. To the left, it is separated from the neck by a deep groove, the "incisura pancreatis." In this groove lie the superior mesenteric vessels, which are continued over the anterior surface of the head, near its left border. That portion of the gland which lies to the left of the ves- sels, along the third part of the duodenum, is termed the "uncinate process," and when, as happens occasionally. 30 The Pancreas: Its Surgery and Pathology it is separated from the rest, it is known as the "lesser pancreas." The superior and inferior pancreatico-duode- nal vessels also course over the head, near its right and left borders respectively, to break up on its anterior sur- face. Above and to the right, the anterior aspect of the head is in contact with the commencement of the trans- verse colon, the posterior surface of which is directly attached to the pancreas by areolar tissue. The lower part of the anterior surface of the head of the gland is covered by peritoneum, reflected from the lower surface of the colon and entering into the formation of the greater sac. This part is in contact with portions of the small intestine. The posterior surface of the head is devoid of perito- neum, and is directly applied to the front of the inferior vena cava, the left renal vein, and the aorta. The com- mon bile-duct also lies in a groove, or canal, in this surface. The neck springs from the upper border of the anterior surface of the head. It passes slightly upwards, forwards, and to the left, to join the body. It is rarely more than an inch (25 mm.) long, is usually about 0.75 inch (18 mm.) wide, and less than 0.5 inch (12.8 mm.) thick. Its junc- tion with the anterior surface of the head is generally grooved by the gastro-duodenal and superior pancreatico- duodenal arteries on the right side. Anteriorly, and to the right, it is in contact with the first part of the duode- num, and also with the pylorus when the stomach is distended. Behind, and to the left, is a groove in which lie the terminations of the superior mesenteric and splenic veins to form the portal vein. The body and tail together measure about 4 to 5 inches (10 to 14 cm.). They are of a pyramidal shape and pre- sent three surfaces of about equal width, averaging 1.25 inches (31 mm.). The body runs from right to left, and slightly upwards. It is moulded to the adjacent organs, and is thickest in Anatomy 31 front of the left kidney. The anterior surface is concave, and looks upwards and forwards. It is separated from Fig. 12. — Vertical section of the body at full term, showing the relation of the uterus to the pancreas (after Braune). the stomach by the lesser sac of the peritoneum, the posterior wall of which is intimately attached to it. At 32 The Pancreas: Its Surgery and Pathology the right extremity of the anterior surface, where the body joins the neck, there is often a well-marked promi- nence, the "omental tuberosity," so called from its com- ing into contact with the small omentum when the stom- ach is distended. The posterior surface looks directly back, and lies upon the aorta, the origin of the superior tnesenteric artery, the pillars of the diaphragm, the splenic artery and vein (which run a tortuous course along its upper border in a single channel or may be two separate grooves), the left kidney and renal vessels, and the left suprarenal capsule. This surface, like the posterior aspect of the head, is devoid of a peritoneal covering, and is connected to the abdominal wall and adjacent organs by areolar tissue. The inferior surface looks downward and slightly forward. It is narrowest at the right end, which rests upon the duodeno- jejunal flexure, but widens towards the left extremity, where it comes into contact with the splenic flexure of the colon. At the full term of pregnancy the uterus rises and comes into contact with the lower border. The middle portion is covered by the jejunum. The whole surface is completely invested by peritoneum, derived from the descending layer of the transverse mesocolon. The tail turns sharply upwards, and backwards. As a rule, it comes into contact with the lower part of the inner surface of the spleen, but occasionally it is separated by a portion of mesentery containing a lymph nodule. The blood-supply of the body and tail of the pancreas is mainly derived from the splenic artery. The hepatic division of the coeliac axis and the inferior pancreatico- duodenal branch of the superior mesenteric supply chiefly the head. The superior (anterior) pancreatico-duodenal artery is a branch of the gastro-duodenal ; passing on to the front of the head, it sends branches into the sub- stance, and also on to the duodenum. The inferior (posterior) pancreatico-duodenal artery arises from the Anatomy 33 upper part of the superior mesenteric, or occasionally from the middle colic, artery; it passes upward, and to the right, across the back of the head, and sends branches to it and to the neighbouring duodenum. The two pan- creatico-duodenal vessels frequently anastomose around the lower border of the head of the pancreas and form a vascular loop. The inferior pancreatic branch of the Fig. 13. — Arteries and veins of the pancreas. superior mesenteric artery runs to the left, along the lower border of the pancreas, often as far as the tail. A large number of small branches are given off by the splenic artery to the body and tail as it courses along the upper border of the gland. Small pancreatic branches are also given off by the hepatic artery as it rests upon the upper border. 3 34 The Pancreas: Its Surgery and Pathology The veins are all tributaries of the splenic and superior mesenteric, the blood from the pancreas being thus car- ried to the portal system. The anterior (superior) pan- creatico-duodenal vein lies on the front of the head, and joins the superior mesenteric. The posterior pancreatico- duodenal runs on the back of the head to open into the portal vein. A number of small tributaries of the splenic vein, corresponding to the arterial branches from the Gastrohepatic omentum. Aorta. Tuber omentale. Layers of transverse mesocolon. Colon Uncinate process. Mesentery. Mesenteric artery and vein. Fig. 14. — Peritoneal reflexions on the pancreas (after Testut). splenic artery, collect the blood from the body of the gland. There are also many small veins arising in the head and neck which run into the portal vein. There is a complex network of lymphatic vessels in and around the gland, which opens into glands situated on the head of the pancreas, in the hilum of the spleen, and along the superior mesenteric vessels. The nerves of the pancreas are provided by cerebro- spinal fibres coming from the vagi, and sympathetic Anatomy 05 fibres derived from the solar plexus. They accompany the arteries through the coeliac, splenic, and superior mesenteric plexuses, and, travelling in the substance of the gland with the ducts, terminate round the acini in rich plexuses of fibres which send fibres to the secreting cells (Miiller) . The nerve fibres are almost entirely non- medullated and have minute ganglia on them (visceral sympathetic ganglia cells— R. y Cajal) as they traverse the gland, and near their dis- tribution to the alveoli small cells, ap- parently of a nervous nature, are also found. Common bile- duct. Duct of San- torini. Orifice of the duct of Santorini. Orificeof the duct of Wirsungand the common bile-duct. Duct of Wirsung. Fig. 15. — The excretory ducts of the pancreas (after Testut). Peritoneum. — The transverse mesocolon is attached to a line running along the anterior border of the pancreas from the neck to the tail. The anterior layer passes upwards and backwards, over the superior surface, to form the posterior wall of the lesser sac, the posterior going downwards and backwards, along the inferior sur- face to form the greater sac. At the neck, and on the head, the two sheets of peritoneum have separate lines of attachment, so that a somewhat variable area is devoid of a peritoneal covering and is only separated from the colon by areolar tissue. In many cases, however, the 36 The Pancreas: Its Surgery and Pathology transverse mesocolon is continued as far as the hepatic flexure, so that the head and neck receive a complete peritoneal investment. The posterior surface is quite uncovered by perito- neum. Ducts. — The pancreas has normally two ducts which open separately into the duodenum. The main duct, or duct of Wirsung, commences in the tail by the union of the small tribu- taries draining that region, and grad- ually increases in size as it courses through the body of the gland from left to right. In the neck it alters its course, bending downwards and backwards, to reach the head of the organ. In the latter it lies nearer the posterior than the anterior surface, and comes into relation with the common bile-duct, beside which it runs to the duodenum. The two ducts pierce the wall of the second part of the duodenum obliquely, about 3 to 4 inches (8 to 12 cm.) below the pylorus, to open into the lumen of the gut by a common ori- fice, situated on a papilla-like fold of the mucous membrane called the ' ' papilla or caruncula major. ' ' Above this there is constantly found a small fold of mucous membrane, which must be raised in order that the caruncle and its orifice may be seen, and running downwards from the caruncle is a small vertical fold known as the "frenum carunculas" or "plica longitudinalis. ' ' Shortly before their Fig. 16. — Photo- graph of a specimen in the Hunterian Museum of the Royal College of Surgeons, showing the separate lobules of the pan- creas with their ducts opening into the duct of Wirsung (anatomical series 277). Anatomv 37 termination the common bile-duct and pancreatic duct usually unite to form a common channel, known as the " ampulla or diverticulum of Vater. ' ' This is a small oval or triangular cavity lying in the wall of the duodenum, having its apex at the duodenal orifice, and its base at the openings of the two ducts. Its average length, Neck of gall-bladder and cystic duct. Hepatic duct. Portal vein. Common bile duct. Duoden Hepatic artery Head of pan creas. Uncinate process. Superior mesenteric artery and vein. Fig. 17. — Head and neck of the pancreas, viewed from the front (after Testut). according to Opie, is 3.9 mm. Occasionally it may be as long as II mm., while in other cases it is non-existent, the two ducts opening side by side upon the common papilla. The orifice of the common bile-duct into the ampulla is above that of the pancreatic duct, and the two are separated by a small transverse fold of mucous 38 The Pancreas: Its Surgery and Pathology membrane. The average diameter of the duodenal open- ing of the ampulla, which is always the narrowest part of the bile channel, is 2.5 mm. (Opie), but in some instances it is equal to, or greater than, the length of the divertic- ulum. The ampulla, and the terminations of the two ducts, are surrounded by a thin layer of unstriped muscle fibre, forming a sphincter (Oddi) . Fig. 18. — Preparation showing the common bile-duct and pan- creatic ducts and their common point of entry into the duodenum (Royal College of Surgeons Museum, anat. series 275 B). The accessory duct, or duct of Santorini, is a very variable structure. For a long time it was regarded as inconstant, but more extended and thorough investiga- tion has shown that it is always present, although, at times, it is small, or partly obliterated, especially in the neighbourhood of the intestine. Opie in the examination of 100 bodies found that the duodenal orifice of the lesser Anatomy 39 duct was obliterated, or so constricted as to be of little or no functional service, in over half the cases investi- Common bile-duct. Valvula; coiiiii ventes. of Vater. carunculae. Fig. 19. — Diagram showing the formation of the ampulla of Vater by the union of the common bile-duct and pancreatic duct and their opening into the duodenum (after Testut). Caruncula major. Frenum carunculae. Fig. 20. — Opening of the ampulla of Vater on the caruncula major in the duodenum (after Testut). gated. The opening, when present, is situated on a small papilla, "the papilla or caruncula minor," lying 0.75 to i 40 The Pancreas: Its Surgery and Pathology inch above, and somewhat ventral to, the papilla major on which the ampulla of Vater or the main duct opens. The duct of Santorini is morphologically, and, in some instances, anatomically, the duct of the head of the pan- creas, representing what, in the lower vertebrates, is the excretory duct of the duodenal portion of the organ. In man, as in many mammals, the main and accessory ducts communicate with each other within the gland by branches of varying size. Literature Birmingham: Cunningham.'s" Text -book of Anatomy," 1906. Cajal, R. y. : "Terminacion de los nervos y tubos glandulaires del pancreas de los vertebrados," Barcelona, 1891. Miiller: Archiv. f. mile. Anat., xi, 405, 1892. Schafer and Symington: Quain's "Anatomy," iii, Part 4. Testut: "Traite de Anatomic humaine," viii, 1894. CHAPTER III EMBRYOLOGY It was formerly taught that the pancreas arises in verte- brates by two outgrowths from the walls of the duodenum, the one dorsal and the other ventral, but it has now been shown, for most members of the group, including man, that the ventral bud in its early stages is double, so that Liver Gall-bladder Common bile- duct Duodenum Stomach Dorsal pancreatic bud Ventral pancreatic bud Fig. 21. — The pancreatic and hepatic processes of a fourth-week em- bryo (after Kollmann). a triple origin of the primitive rudiment, or anlage, of the pancreas is now generally accepted. The first indication of the pancreas in man is seen in the fourth week of intrauterine life, as a process from the dorsal wall of what will later become the second part of the duodenum. It grows out between the layers of the dorsal mesogastrium and eventually reaches the spleen, 41 42 The Pancreas: Its Surgery and Pathology as it lies above the cardiac end of the stomach. From the opposite wall of the duodenum the two ventral buds take their origin, on either side of the hepatic diverticu- lum, which has made its appearance at an earlier date. As they increase in size they fuse together to form a sin- gle mass, which later unites with the larger dorsal out- Gastrohepatic omentum Spleen Dorsal meso- gastrium — Stomach Mesentery // Fig. 2 2. — The relation of the pancreas, spleen, and liver to the xneso- gastrium in the embryo (after Keith). growth. The greater part of the adult pancreas is de- rived from the dorsal process, the ventral buds only giving rise to the lower part of the head (Fig. 23). Originally, the gland lies parallel to the dorsal border of the stomach, the head occupying the bend of the duo- denal loop and the tail being directed forwards against the spleen. The whole gland is then completely invested Embryology 43 with peritoneum. As the stomach rotates to the left, and the great omentum is developed, the pancreas comes to lie transversely across the abdominal cavity. The former right surface now becomes posterior, and is closely applied to the wall of the abdomen. Its peritoneal cover- ing gradually disappears and is replaced by a connecting layer of areolar tissue. The anterior aspect of the gland, which was formerly its left surface, comes to lie behind Dorsal Gastrohepatic mesentery omentum Common bile-ducl "~ Pancreas Mesentery Duct of Wirsung Duodenum Ventral pancreas Fig. 23. — Diagram of the pancreas showing its relation to the dorsal and ventral mesenteries, the parts formed from the ventral and dorsal outgrowths ; and the formation of the duct of Wirsung by a union between the ducts of the dorsal and ventral buds (after Keith). the stomach and retains its peritoneal coat, so that the adult arrangement of the pancreas, outside the peritoneal cavity, is reached. In many animals a process from the dorsal outgrowth extends into the gastro-hepatic omen- tum as an omental lobe, and, in man, this is sometimes found to be represented by a well-marked omental tuber- osity. Each primitive pancreatic outgrowth is provided with a 44 The Pancreas: Its Surgery and Pathology duct opening into the duodenum. That from the ven- tral process opens by an orifice common to it and the hepatic diverticulum, close to which it originated. The duct of the dorsal bud communicates with the duodenum by an opening situated nearer to the pylorus. The two ducts almost always anastomose within the substance of •the gland at an early stage, and it is found that, as a con- sequence of this, the chief excretory channels of the adult pancreas are usually of complex origin. The main duct, or duct of Wirsung, of the adult is partly derived from the duct of the dorsal process and partly from that of the ventral outgrowth. That part which lies in the body of the gland represents the main portion of the dorsal em- bryonic duct, and that which courses through the head, opening into the duodenum along with the common bile- duct, is derived from part of the dilated channel of the ventral pancreatic process. The remaining section of the dorsal duct, lying between the point of anastomosis of the two primitive channels and the duodenal opening, usually undergoes partial atrophy and becomes the acces- sory pancreatic duct, or duct of Santorini, of the adult organ. The primary pancreatic processes are hollow, but the secondary, tertiary, and succeeding buds which arise from their walls consist of solid masses of cells. Later these acquire a lumen and the typical structure of the acinotubular pancreas is gradually developed. Literature Brachet: Journ. de I'Anat. et de la Phys., 1896, xxxii, 620. Brunn, Von: Merkel-Bonnet Ergebnisse, 1894, Abt. ii, iv, 87. Choronschizky : Ref., Anat. Hefte, Merkel-Bonnet, Ergebnisse, 1899, Abt. ii, ix, 669. Felix: Arch. f. Anat. u. Phys., 1892, Anat. Abt., 281. GSppert: Morphol. Jahrb., 1891, xvii, 100. Gotte: Leipzig, 1875. Hamburger: Anat. Anzeiger, 1892, vii, 707. Hammar: Arch. f. Anat. u. Phys., Anat. Abt., 1893, 123. Helly: Arch. f. mik. Anat., 1900, Ivi, 291; 1901, Ivii, 271. Jankelowitz: Inaug. Diss., Berlin, 1895. Embryology 45 Keith: "Human Embryology and Morphology, " 1904. Laguesse: Bibliogr. Anat., 1894., ii, loi. Opie: "Diseases of the Pancreas," 1903. Schafer: Quain's "Anatomy," 1896. St5hr: Anat. Anzeiger, 1893, viii, 205. Stoss: Anat. Anzeiger, 1891, vi, 666. V5lker: Arch. f. mik. Anat., 1902, lix, 62. Wlassow: Morpholog. Arbeiten herausgeg. von Schwalbe, 189'; iv, 67. Zimmermann: Anat. Anzeiger, 1889, iv, 139. CHAPTER IV ANATOMICAL ANOMALIES When considering the anatomical anomaHes of the pan- creas it is important that the embryology and compara- tive anatomy of the organ should be borne in mind, for by this means arrangements and distributions of the glandular substance, which would otherwise appear cap- ricious, are explained, and abnormalities of the ducts are simplified. The abnormality around which most of the literature of the subject centres, and which has aroused the greatest amount of controversy, is the occurrence, in from 0.5 to I per cent, of persons, of one or more accessory masses of glandular tissue. These accessory pancreases have been classified by Glinski into three divisions : 1. ''Pancreas minus,'' in which a supernumerary lobule, or lobe, is present in the head of the gland, separated by a more or less marked constriction. 2. ''Pancreas accessorium," where isolated nodules of pancreatic tissue are found embedded in the walls of the gastro-intestinal tract, or in other situations. 3. "Pancreas divisum," in which parts of the gland may be found separated from the main mass, but still connected by means of their ducts to the chief excretory channel of the organ. Pancreas Minus. — Examples of the first variety can hardly be classed as instances of an accessory pancreas, for they only represent, as a rule, an exaggeration of a normal condition, in which a portion of the pancreas is separated from the remainder by a more marked depres- sion than usual. 46 Anatomical Anomalies 47 The commonest is that to which reference has already been made when considering the anatomy of the gland. In this variety a portion of the head, lying behind the mesenteric vessels, is divided from the rest by a deep cleft to form the lesser pancreas ("pancreas parvum" of Winslow). Occasionally, however, the cleft, in which the superior mesenteric vessels lie, is bridged over so that they are contained in a canal in the head of the pan- creas, and the descending lobe is thus firmly fixed to the body. A rarer anomaly, of much surgical interest, is an exag- geration of another normal condition. It has already been pointed out that the head of a w^ell-developed gland may embrace one-third of the circumference of the second part of the duodenum ; in rare cases the overlapping is so great that the whole circumference of the bowel is en- closed in a ring of pancreatic tissue. Either at birth, or later if the gland should be invaded by growth or become enlarged from inflammatory changes, it may lead to symptoms of obstruction resembling those due to pyloric stenosis. Shirmer collected four examples of this con- dition from the older literature, quoting cases by Tiede- mann, Becourt, Moyse, and Ecker. More recently in- stances have been recorded by Symington, Generisch, Tieken, Santos, and Vidal. That of the last named was in a child, and the symptoms, which appeared immediately after birth, suggested congenital stenosis of the pylorus. At operation the true state of things was discovered, and gastro-enterostomy was performed to relieve the ob- struction. The patient recovered and steadily gained in weight after the operation. The case operated on by Santos was a woman of twenty-six, who sufTered from constant vomiting and was much emaciated. Gastro-enterostomy was performed, but the patient died and the anomaly of the pancreas was confirmed post-mortem (Fig. 24), Symington's case was discovered post-mortem in an adult 48 The Pancreas: Its Surgery and Pathology male, and is described as follows in the "Journal of Anat- omy and Physiology" for 1885: "On distending the intestine with air, in order to facilitate the dissection of the head of the pancreas, it was noticed that the upper part of the descending portion did not become dilated like the rest of the intestine, and on examination this was found to be due to its being completely surrounded in that situation by pancreatic tissue. Two processes of the pancreas passed from the upper part of the head of Fig. 24. — Congenital malformation of pancreas compressing the duodenum and leading to obstruction which required gastro-enter- ostomy. Stomach and duodenum laid open, showing the gastro- enterostomy opening and the stricture caused by the pancreas (Santos). the gland towards the right, one in front and the other behind the duodenum. They blended on its outer side so as to form, with the head of the gland, a ring of pan- creas encircling the duodenum. The processes became somewhat narrower as they passed outwards, and the portion of the gland on the right side of the duodenum was about half an inch in vertical extent. On dissecting out the ducts of the pancreas nothing unusual was ob- served in their arrangement. The common bile-duct Anatomical Anomalies 49 opened into the duodenum below the seat of the constric- tion. The circumference of the distended duodenum, where it was surrounded by the pancreas, was two and a half inches, while above and below that it was more than three times as large. In a case operated on by one of us, a prolongation from the head of the pancreas was found extending upwards, in front of the common bile-duct and the hepatic duct, and exerting pressure on both, owing to its being inflamed and swollen. The body and tail of the pancreas are rarely the seat of anatomical abnormalities. Occasionally the latter is bifid, and a case has been recorded by Klobin in which an enlargement of the tail was found on investigation to contain an accessory spleen. Glinski's third division, "pancreas divisum," also hardly merits the description of accessory pancreas, for it is really represented by portions of the gland which have become separated by the mechanical pressure of blood-vessels, etc., during development. Hyrtl has de- scribed cases belonging to this class in which the head of the gland was separated from the body, a portion of the head lay behind the mesenteric vessels, and the tail was separated from the body of the organ. Engel records an instance in which a portion of the pancreatic tissue was situated under the head, and at the inner side of the descending portion of the duodenum, but was connected with the main pancreatic duct. Pancreas Accessorium. — The condition to which the term "accessory pancreas" strictly applies, and which most writers describe under that name, is the occurrence in connection with some part of the gastro-intestinal tract of one or more masses of pancreatic tissue in an abnormal situation and independent of the main mass of the gland. It is this which forms the second division of Glinski's classification, the "pancreas accessorium." Accessory masses of pancreatic tissue have been described in the 50 The Pancreas: Its Surgery and Pathology walls of the stomach, duodenum, jejunum, and ileum. Thorel has maintained that they are most commonly met with in the stomach, but Glinski states that the intestinal wall is the more frequent site. A survey of the literature of the subject tends to support Glinski 's contention. We have been able to meet with records of thirty-seven cases, in which forty-one masses of accessory pancreatic tissue were present, and to these we have to add the hitherto unpublished specimen shown in Fig. 25, from the Leeds Pathological Museum. It was discovered at a post- mortem examination made by W. H. Maxwell Telling and was situated in the wall of the in- testine at the duodeno- jejunal junction. In eight instances (Klob, Gegenbaur, Weichsel- baum, Glinski, Schirmer, and three by Opie) there was a sin- gle nodule in the wall of the stomach. In one, recorded by Opie, there was a mass of pan- creatic tissue in the stomach wall, 8 cm. from the pylorus, and a second mass at the py- lorus, which on microscopical examination was only found to contain a dilated duct. Opie also describes a case in which an accessory pancreas was found in the stomach, 2 mm. from the pylorus, and another nodule in the wall of the duodenum, 9.5 cm. below the pylorus. Wagner records an instance where an accessory pancreas was present on the anterior wall of the stomach, midway between the pylorus and the cardiac end, and a second nodule, the exact situation of which is not described, in the intestine. Twenty-nine accessory masses of pancreatic tissue E.n-WRlCjHTr Fi^:. 25. — Accessory pancreatic nodule in the intestinal wall at the duo- den o-jejunal junction. Anatomical Anomalies 51 have been described in the intestine ; all, however, were situated above the ileo-ca;cal valve. Six were met with in the walls of the duodenum (Weichselbaum, Zenker, and four by Opie). One of Opie's cases has already been referred to in connection with a similar nodule in the stomach. Four lay in the wall of the duodenum above the pancreas, one was situated on the convex border opposite the head of the gland (Zenker), and one was below the pancreas (Opie). The nodule in Telling's case lay at the duodeno- jejunal junction. We have been able to find records of nine cases in which a single mass of pancreatic tissue was present in the wall of the jejunum (Klob, Turner, Xicholls, Lewis, Zenker three, and Opie two). In all but one case it lay within two or three feet of the origin of the gut. The exception is described by Opie, and here it was situated 4 metres from the stomach. Zenker quotes a case in which two accessory pancreases were present in the jeju- nal walls, one 16 cm. below the duodenum, and the second 32 cm. lower down. The five cases in which an accessory pancreas was found associated with the ileum are peculiar in that the glandu- lar tissue was in each instance situated at the end of a slender or funnel-shaped diverticulum of the intestinal wall (Zenker, Neumann, Nauwerck, Hansemann, Schirmer) . This peculiarity is not, however, confined to the ileum, for in Weichselbaum 's case of an accessory pancreas in the stomach wall the nodule was situated at the bottom of a diverticulum near the pylorus, and in cases described by Roth, Opie, and RoUeston diverticula were found on the left side of the duodenum, running into the substance of the pancreas. The fact that the ileal outgrowths have been most commonly found about two feet from the ileo- cascal valve has naturally suggested that they were the remains of the vitelline duct and examples of Meckel's diverticulum. The occurrence of similar csecal appen- 52 The Pancreas: Its Surgery and Pathology dages apart from, and in connection with, pancreatic tissue in other situations along the walls of the gastro- intestinal tract, and the fact that the vitelline duct is already formed when the pancreas begins to develop, are, however, opposed to such a theory. The discovery of a true Meckel's diverticulum, in addition to the intes- tinal outgrowth containing pancreatic tissue, in three cases has cast further doubt on the suggestion. The shorter forms may be possibly explained by the weaken- ing of the muscular wall of the gut, produced by the inclusion of the pancreatic tissue, which would allow the mucous membrane to bulge outwards under the pres- sure of the intestinal contents, and to carry before it the pancreatic nodule and remains of the muscular tissue. It is more probable, however, that they are the result of traction exerted during development. The similarity, at least of the pyloric and duodenal diverticula, to the caeca found around the pylorus and along the duodenum of some fishes raises the interesting question as to whether some of them may not be a partial reversion to an an- cestral type. A unique case has been recorded by Wright, in which a mass of pancreatic tissue, 3.5 mm. in diameter, was found embedded in the wall of a congenital umbilical fistula, and apparently connected with the persistent re- mains of the vitelline duct. Letulle met with five cases in which an accessory pancreas was present in two hundred post-mortems, but unfortunately he gives no details. Opie collected ten examples from eighteen hundred autopsies, which have been included in the foregoing survey of the subject. The investigations of Helly have shown that in many individuals a small mass of pancreatic tissue, forming a true accessory pancreas, lies in the papilla of Santorini's duct, entirely isolated from the remainder of the gland, and either communicating directly with the duodenum by a separate channel, or draining by a small tributary Anatomical Anomalies 53 into the lesser duct near its termination. Opie has con- firmed these observations and described a similar condi- tion in connection with the duct of Wirsung. The accessory pancreas in all the recorded cases, except those of Klob and Wright, was connected to the adjacent lumen of the alimentary tract, or diverticulum from it, by one or more ducts, which in some instances opened on to a well-defined papilla. In Klob's case it is possible that the duct was overlooked. The size of the pancreatic nodule varies considerably in different cases. No exact measurements are given in the older records, but a rough idea can be formed from the comparison with a bean and a hempseed, by Weichsel- baum, and a pea b}^ Neumann. The largest recorded accessory pancreas is that described by Glinski, which measured 4.5 by 3.5 by i.o cm. They are usually much smaller than this, however, and range about i.o cm. in diameter. The smallest is described by Opie, and measured only 3 mm. As a rule, they lie embedded in the muscular tissue of the gut wall, projecting more or less into the submucosa, and beneath the peritoneum, but occasionally they lie in the submucosa only. Microscopically these accessory nodules have the char- acters of ordinary pancreatic tissue. Islands of Langer- . hans were present in cases described by Wright, Opie, and Lewis, but Le tulle and Turner were unable to dis- cover them in their cases. In many instances there has been an increase of the interstitial fibrous tissue, indi- cating that the gland substance has undergone chronic inflammatory changes, and in some the fibrosis has advanced to such a stage that few or none of the glandular elements remained. In such cases the nodule may present characters suggesting an adenoma, or may only consist of fibrous tissue with a few dilated ducts (Opie). It has been suggested that the presence of one or more 54 The Pancreas: Its Surgery and Pathology masses of accessory pancreatic tissue in the walls of the gastro-intestinal tract may stay, or prevent, the onset of diabetes in cases where the main gland is diseased, but, when the small size of even the largest recorded examples is taken into account, and the frequency with which their contained gland substance shows evidence of disease is considered, it would appear to be unlikely that they can exert any material influence in that direction. There has been much speculation as to the origin of these accessory pancreatic nodules. Zenker has explained them by supposing that an additional pancreatic rudi- ment occasionally arises from the duodenum, close to the origin of the normal buds, and that, becoming attached to the stomach, or duodenum, as the case may be, it is carried upwards, or downwards, by the growth of the gastro-intestinal tract, eventually becoming separated from its origin to form a distinct mass of pancreatic tissue in one or other situation. More recently Glinski has suggested that the pancreas of persons in whom accessory pancreatic masses are met with, as well as those of normal individuals, develops from two only of the three buds which are now known to be present in the embryo. Under ordinary circumstances the third out- growth remains as a rudiment, but occasionally it persists, and, undergoing a limited amount of development, be- comes attached to the gastric or intestinal wall as an accessory pancreas. Both these hypotheses only succeed in explaining those cases in which two masses are present by supposing that, in rare instances, a third or fourth pancreatic bud is present in the embryo. Glinski sup- poses that in such cases the dorsal as well as the ventral outgrowth is double. In disproof of these explanations Opie quotes two cases observed by him. In one there was pancreatic tissue in the walls of the stomach, duode- num, and the lesser papilla, and in the other in the duode- nal wall above the pancreas and in the papilla of the duct Anatomical Anomalies 55 of Santorini. He points out that, should an accessory pancreas arise by persistence of one of the two ventral outgrowths, it can only be carried by lengthening of , the intestine downwards towards the duodenum, while, should it arise from part of a double dorsal outgrowth, it can only be carried upwards towards the stomach, but that, in these two cases, there were two accessory glands above the pancreas, which can only be explained by the occurrence of a triple primitive dorsal rudiment — a condition which is unknown in the development of any vertebrate animal. The explanation which he himself offers is similar to that given by Helly for the presence of pancreatic tissue in the papilla of the duct of Santorini. Helly believes that lateral branches from the dorsal embryonic outgrowth may, at an early stage of development, penetrate the wall of the intestine, and later, becoming separated from the rest, acquire new ducts, thus giving rise to the con- dition found in some cases after birth. Opie, extending this theory, supposes that accessory masses of pancreatic tissue in all parts of the gastro-intestinal tract arise by entanglement of lateral branches of the primitive buds in the developing walls of the alimentary canal, those originating in the dorsal growth being carried upwards to form the gastric and upper duodenal nodules, and those from the ventral bud being carried downwards to consti- tute the accessory masses of gland substance met with in the lower duodenum, jejunum, and ileum. Other anatomical anomalies of the pancreas have been described, although they are not numerous. Hertz re- cords a case in which there was falling forwards of the gland ; Cacchini describes an instance in which there was congenital displacement of the head of the gland, asso- ciated with gastroptosis ; and cases have been reported in which the tail of the organ, normally its most movable part, has been dragged into abnormal situations by a 56 The Pancreas: Its Surgery and Pathology wandering spleen. Klebs states that the pancreas may be pushed downwards by tight lacing, and that retro- peritoneal tumours and aneurysms of the adjacent vessels may carry it upward. Although the pancreas is one of the most firmly fixed organs in the abdominal cavity it has been occasionally met with in hernial sacs, and Dobrzycki describes a case in which a movable pancreas, giving rise to symptoms resembling those of movable kidney, was present in a man as the result of a fall from a height. The pancreas formed part of the contents of 27 out of 276 cases of diaphragmatic hernia collected by Lacher, and in one case described by Claessen, it had passed through a rent in the diaphragm into the thoracic cavity. Two cases of congenital umbilical hernia in which the pancreas was found in the sac are recorded, and Rose met with a similar relation of the gland in an umbilical hernia in a woman of sixty-four. One case in which the pancreas was appa- rently entirely wanting has been described. The anatomical variations of the pancreatic ducts, and also of the common bile-duct, have important bearings upon the pathology of the pancreas, and particularly upon the pathology of pancreatitis. It is therefore important that they should be discussed in detail. The common bile-duct may be divided into four por- tions : 1. The supraduodenal part. 2. The retroduodenal part. 3. The pancreatic part. 4. The intraparietal part. Starting by the junction of the cystic and hepatic ducts, it courses along the free border of the lesser omentum, associated with the portal vein and hepatic artery. Then passes behind the first part of the duodenum, and soon comes into relation with the pancreas. Finally it pierces the wall of the second part of the duodenum along with Anatomical Anomalies 57 the duet of Wirsung. The first two portions are unim- portant as regards the pancreas, but the relations of the remaining sections have considerable bearing upon the etiology of diseases of that organ. The third, or pancreatic, portion of the common bile- duct measures from 20 to 25 mm. in length. It extends from the inferior border of the first part of the duodenum to the point where the duct penetrates the wall of the second part. This portion of the common duct crosses a small quadrilateral area, bounded above by the in- ferior border of the first part of the duodenum, below by the superior border of the third part, externally by the inner border of the second part, and internally by the su- perior mesenteric vein. Anteriorly it is closely applied to the posterior surface of the head of the pancreas. iVccording to Helly, this portion of the common duct is com- pletely embraced by the head of the gland in 62 per cent, of bodies, and lies in a deep groove in the remaining 38 per cent. Bunger, in a careful examination of fifty-eight subjects, found that in 2 5 per cent, the duct ran in a groove and in 75 per cent, was entirely enclosed in pancreatic tissue. Wyss investigated the relation of the common duct to the pancreas in twenty-two bodies, and found Fig . 2 6 . — Diagram showing rela- tions of the common bile-duct to the duodenum (viewed from behind) : I, Supraduodenal portion of the common bile-duct; 2, retroduo- denal portion of the common bile- duct; 3, pancreatic portion of the common bile-duct; 4, intraparietal portion of the common bile-duct (after Testut). 58 The Pancreas: Its Surgery and Pathology that it was surrounded by the tissue of the gland in seven (31.7 per cent.), and grooved the posterior surface of the head in fifteen (68.1 per cent.). These variations in the relations of the duct to the pancreas are important, for it is obvious that swelling of the gland, when the duct passes through the substance of the head, may compress Portal vein Hepatic artery Hepatic duct Neck of gall- bladder and cystic duct Hepatic artery Splenic vein Body of pancreas Superior mesenteric arterv and vein Infrrior pancre- atic and duo- denal artery Common bile- duct Uncinate process Fig. 27. — Diagram showing the common bile-duct passing through the head of the pancreas, a portion of which has been reflected (viewed from behind) (after Testut). it and lead to occlusion, while when it is contained in a groove it may be pushed aside and escape compression. The intraparietal, or interstitial, portion of the common bile-duct comprises that portion of the canal which is contained in the thickness of the wall of the duodenum. Its relation tO the duct of Wirsung, and its union with the termination of the chief excretory channel of the Anatomical Anomalies 59 pancreas to form the diverticulum of Vater, have already been described. The mode of formation of the ampulla of Vater, and the terminations of the common bile-duct and pancreatic duct, are liable to great variation, and these variations will be seen to be of considerable importance when we come to consider the diseases of the pancreas. Letulle and Nathan Lorrier distinguish four types : 1. The first, or normal, arrangement, in which the ducts unite to form the ampulla of Vater, has already been dealt with (Figs. 19 and 28, a). 2. In the second type the pancreatic duct joins the I^RD. Fig. 28. — Diagram of the four methods by which the common bi'e- duct and duct of Wirsung enter the duodenum: C. D, Common bile- duct; P. D, pancreatic duct; V, ampulla of Vater; O, common orifice; C, cup-shaped depression in the wall of the duodenum; P, papilla. common duct some little distance from the duodenum; the ampulla of Vater is absent, and the united ducts open into the duodenum by a small, flat, oval orifice (Fig. 28, b). 3. In the third type the two ducts open into a small fossa in the wall of the duodenum, while the caruncle and ampulla of Vater are both absent (Fig. 28, c). 4. In the fourth type the caruncle is well developed, but the ampulla of Vater is absent, the two ducts opening side by side at the apex of the caruncle. In eleven out of one hundred specimens examined by Opie the arrange- ment described in this last type was present (Fig. 28, d). 5. Rarely the common bile-duct unites with the duct 6o The Pancreas: Its Surgery and Pathology of Santorini instead of with the duct of Wirsung, as in a specimen preserved in the Museum of the Royal College of Surgeons (Fig. 29). The A^ariations met with in the two pancreatic ducts are well shown in a series of one hundred cases investi- gated with regard to this point by Opie. In every case he found that both ducts were present, although occa- sionally one or the other was so small that it was demon- strated with difficulty; the duct of Wirsung and the Fig. 29. — Photograph of a specimen in the Hunterian Museum of the Royal College of Surgeons, showing the common bile-duct joining the duct of Santorini (anatomical series 277 A). common bile-duct always entered the duodenum together, while the duct of Santorini invariably opened into the intestine at a higher level. In ninety specimens the two pancreatic ducts anastomosed within the substance of the gland; in ten there were two wholly independent ducts. On investigating the relative size and the patency of the ducts, he found that out of the ninety cases in which ducts anastomosed the duct of Wirsung was the larger in eighty-four ; in these the duct of Santorini was patent in sixty-three, and impervious in twenty-one. The duct of Anatomical Anomalies 6i Santorini was larger than the duct of Wirsung in six, but the latter was jjatent in all. Of the ten in which no anastomosis between the ducts could be discovered, the Fig. 30. — Diagram to show the variations in the ducts of Wirsung and Santorini (after Opie). duct of Wirsung was the larger of the two in five, and the duct of Santorini in the other five. So that in 89 per cent, of the cases the duct of Wirsung was the main excre- 62 The Pancreas: Its Surgery and Pathology tory channel of the pancreas, and in 21 per cent, the duct of Santorini was apparently obliterated near its termina- tion, and, even in those instances where it was patent, it was found to diminish in size as it approached the duode- num. Thus the duct of Santorini could not be relied upon to supplement the duct of Wirsung in at least 31 per cent, of the cases if the latter was obstructed. More- over, it must be borne in mind that the duct of Santorini. Large accessory duct Fig. 31. — Drawing of a preparation showing a large accessory pancreatic duct opening into the ampulla of Vater (Royal College of Surgeons Museum, 277 B). even if patent and communicating with the duodenum, may itself be compressed by a moderate sized gall-stone passing down the pancreatic portion of the common bile- duct. In an earlier observation Schirmer obtained somewhat similar results. He examined the pancreas in one hun- dred and four bodies and found that in sixty-six (6^ per cent.) there were two ducts opening into the intestine and communicating with the substance of the gland, Anatomical Anomalies. 63 while in thirty-seven cases (35 per cent.) the two ducts did not anastomose, or one or other did not open into the duodenum. In one case three ducts were present. A specimen showing a similar anomaly is preserved in the Museum of the Royal College of Surgeons (Fig. 31). It was discovered during dissection by the prosector to the college, W. U. Pearson. Literature Bunger: Med. Press, 1902, p. 523. Ecker: Zeitschr. f . rationelle Medicin, 1862. Engel: Medicin. Jahrb., Wien, 1840. Gegenbaur: Reichert's Archiv, 1863. Generisch: Verhandl. internat. mad. Congress, 1890. Glinski: Virchow's Archiv, 1901, clxiv, 132. Helly: Archiv. f. mik. Anat., Hi, 773. Hyrtl: "Topograp. Anatomic." Klob: Zeitschr. d. Gesellsch. d. Aertze, Wien, 1859. Lens: Boston City Hosp. Rep., 1905, p. 172. Nauwerck: Zeigler's Beitrage, xii, 1893. Neumann: Archiv. f. Heilkunde, xi, 1870. Nicholls: Montreal Med. Journ., Dec, 1900. Opie: "Diseases of the Pancreas," 1903. RoUeston: Journ. Anat. and Physiol., xxviii, 12. Santos: Medical Congress, Lisbon, 1906. Schirmer: "Beitrag zur Geschichte und iVnatomie des Pancreas," 1893. Inaug. Dissert., Basel, 1893. Symington: Journ. Anat. and Physiol., xix, 292. Turner: Lancet, Dec. 3, 1904, p. 1566. Vidal: Dix-huitieme Congres de Chirurgie, Paris, 1905. Wagner: Archiv. f. Heilkunde, 1862. Weichselbaum : Bericht. d. Rudolf stiftung., 1884. Wright: Journ. Boston Soc. Med. Sci., 1901, v, 497. Zenker: Virchow's Archiv, xxi. CHAPTER V SURGICAL ANATOMY The intimate relations of the head of the pancreas to the duodenum may lead to invasion of that part of the 4i»^ Fig. 32.- — Invasion of the duodenum by carcinoma of the head of the pancreas (Leeds Path. Museum, EE 204 A). intestine by disease of the gland, and, conversely, a pri- mary growth of the duodenum may secondarily involve 64 Surgical Anatomy 65 the pancreas. Gallaudet has described a case in which a cancer of the head of the pancreas so far obHterated the lumen of the duodenum as to call for gastro-enterostomy, and we have recently had under our observation a case in which a malignant growth of the duodenum gradually invaded the pancreas and eventually gave rise to a severe grade of diabetes. Specimens showing invasion of the duodenum by pancreatic growth are preserved in the Museum at St. George's Hospital (201 A) and in the Leeds Pathological Museum (EE 204 A, EE 204 B). An example of the converse condition is to be seen at St. Fig- 33- — Sarcoma of the pancreas invading the duodenum (Leeds Path. Museum, EE 204 B). Mary's. The duodenum may also be compressed or dis- torted by cysts or tumours of the pancreas, or may be involved in a pancreatic abscess which may discharge itself into the lumen of the gut. The proximity of the pancreas to the stomach (Fig. 34) renders it liable to invasion by ulcer or cancer of that organ, and fixation of the stomach by adhesions, whether they arise from disease of the pancreas or of the stomach itself, may lead to a train of symptoms when the latter organ is distended with food, owing to the limitation of its move- ments in a downward direction, as well as giving rise to 5 66 The Pancreas: Its Surgery and Pathology pain, from interference with its normal peristaltic move- ments. Adhesion to, or invasion of, the pancreas by a cancerous growth of the stomach or pylorus not only adds to the danger of operations undertaken for the relief or cure of the condition, but renders a return of the disease much more probable if removal is attempted. Von Mikulicz's experience on this point is most instructive; in ninety-one partial gastrectomies, without injury to the pancreas, twenty-five died as the result of operation, a mortality of 27.5 per cent. ; but in thirty cases in which the pancreas was injured or partly removed the mor- tality was 70 per cent. {i. e., twenty- one deaths), most- ly from peritonitis. Nevertheless, as part of the oper- ation of gastrec- tomy, a partial pancreatectomy has been success- fully performed by one of us, by Mik- ulicz, by Kocher and others. Fen wick investigated one hundred cases of cancer of the pylorus and found that the pancreas was adherent in six ; in another series of one hundred cases of malignant disease of the cardiac end of the stomach the pancreas was adherent in sixteen, and in the same num- ber of cases of cancer of the lesser curvature or posterior wall, it was adherent in nineteen. Chronic ulcers of the stomach, when they become ad- herent to the pancreas, may set up pancreatitis, and even give rise to an abscess, as is shown in the case of a man Fig. 34. — Diagram showing the relations of the stomach to the pancreas (after Testut). Surgical Anatomy 67 who was operated upon by one of us six years ago. In this instance the pancreatic abscess had burst into the stomach, giving rise to acute gastritis, with extremely foul stomach contents and incessant vomiting. His symptoms were relieved, and he was eventually cured, by drainage of the stomach into the jejunum through a gastro-enterostomy opening. In another case, also a man, an ulcer of the posterior wall had become adherent Fig. 35. — Chronic ulcer of the posterior wall of the stomach eroding the pancreas (Fenwick, London Hospital Museum). to the pancreas and produced a cavity in the substance of the gland into which the tip of the finger could be passed. A third case may be cited in which a middle- aged man had suffered from symptoms of chronic gas- tric ulcer for several years, with vomiting of coffee-ground material. On exposing the stomach no evidence to ac- count for the trouble could be found, but when it was opened a large ulcer, one and a half by three inches in diameter, was discovered on the posterior wall, eroding 68 The Pancreas: Its Surgery and Pathology the pancreas. Posterior gastro-enterostomy was followed by complete and permanent recovery. The relations of the pancreas to the peritoneum are of the utmost importance, both from a surgical and patho- logical point of view. The retroperitoneal position of the organ is of great importance, for it explains not only the course taken by pus in some cases of suppurative pancrea- titis, upwards to the diaphragm and downwards towards the left iliac fossa, but also how such collections may be reached from the right or left loin, especially the latter, by an incision in the costo-spinal angle, or from the left iliac fossa, or between the ribs, when it has travelled up- wards and presents as a subdiaphragmatic abscess. The fact that the anterior surface of the pancreas pro- jects into the lesser sac renders it easy to explain how this cavity is invaded in inflammatory affections or injury of the gland, and, from its shape, it is not difficult to see how, when it is filled with fluid, it is in many instances mistaken for a true pancreatic cyst. The real nature of this variety of pseudo-cyst was demonstrated many years ago by Jordon Lloyd. The sharply limited surfaces of the pancreas, as well as the indefinite site of origin of true cysts of the gland, cause considerable variation in the relations of any tumour which may develop. These relations and the variations induced by the origin of a cyst above or below the transverse mesocolon, and to the right or left of the mesentery, as well as the mistakes in diagnosis which are likely to be caused thereby, will be fully considered in a subsequent chapter (Chapter XVIII). The situation of the pancreas at the back of the abdom- inal cavity makes the technique of operations upon it somewhat difficult, unless it is approximated to the abdom- inal wall by disease, as in the case of pancreatic cysts, or by some special method, such as has been described for exposing the biliary passages.^ Various routes have ^ Robson: " Dis. of the Gall-bladder and Bile-ducts." Surgical Anatomy 69 to be adopted according to the situation of the diseased part and the direction of enlargement of the organ. The operative methods by which the gland can be exposed may be divided into transperitoneal and retro- peritoneal. In the transperitoneal methods it is reached by a median or lateral incision in the anterior abdominal wall, and then either through the gastro-hepatic or through the great omentum, or after pushing up the omentum and transverse colon, through the mesocolon. In each case the omental bursa is opened. Another transperitoneal route, employed by Korte, and by one of us in pancreatic lithotomy and when removing a por- tion of the gland for microscopical examination in sus- pected cancer, — by which, however, only the head of the organ can be reached, — is to force a way along the side of the duodenum, the peritoneal covering of which must first be incised. A third method, which is also useful in exposing the pancreatic portion of the common bile- duct, is to incise the parietal peritoneum, lateral to the descending portion of the duodenum, to detach the duode- num from the abdominal wall and then lift it inwards, separating it from the front of the kidney, thus exposing the posterior surface of the head of the gland. The retro- peritoneal methods, by incisions in the lumbar regions, only allow of the head or tail of the organ being dealt with, and should therefore be employed only when, through the effects of disease, the affected part is enlarged and pushed to one side or the other, as by abscess, cyst, or tumours. The anatomical relations of the pancreas to many structures, including the aorta and vena cava, the coeliac plexus, the spleen, the left suprarenal capsule, the left kidney, the portal vein, the duodenum, the stomach and colon, and even the uterus during pregnancy, as well as the common bile-duct and the middle colic artery, injury of which is followed by gangrene of the transverse colon 7o The Pancreas: Its S.urgery and Pathology (Kronlein), have all to be remembered in undertaking operations upon the pancreas. Its relations to these important structures, its fixation, and its great vascularity would render an operation for the complete extirpation of the pancreas extremely diffi- cult, even if it were justifiable on physiological grounds, but where disease is invading the distal part of the body, or tail, the removal of that portion is both justifiable and safe in the case of cystic or solid, benign or malignant, growths. The variations in size of the ampulla of Vater have been already referred to, and the bearing of these upon acute pancreatitis will be dealt with when that subject is con- sidered subsequently. CHAPTER VI HISTOLOGY The structure of the pancreas at once recalls that of the salivary glands, hence the names "abdominal salivary gland," "gland salivaire abdominale," "bauchspeichel- druse, ' ' that have been applied to it. The resemblance is, however, only a superficial one, for although in its broad lines the pancreas is constructed on the same plan as a serous salivary gland, such as the parotid, its minute anatomy is much more complex, as was first clearly demonstrated by the researches of Langerhans in 1869. Like the parotid, the pancreas is a compound tubular gland, composed of branching ducts terminating in acini of a tubular form. The acini about the terminal ducts are grouped together to form primary lobules, which in man are usually more or less fused together to form larger secondary lobules about the medium sized ducts. These are again grouped together to form tertiary lobules, which represent the smallest subdivisions of the organ seen on the surface with the naked eye. The larger lobes of the gland are formed in a similar manner by the union of the lobules around the larger ducts. The lobules are less definitely polygonal than in the salivary glands, and they are also less compactly arranged, so that the gland is of a looser and softer texture. The alveoli are much larger and more tubular than in the parotid, and, since they are also relatively more numerous, fewer ducts are seen in a given sectional area. The arrangement of the connective-tissue framework of the normal pancreas is of importance in view of the changes that occur in it as the result of chronic inflam- 71 72 The Pancreas: Its Surgery and Pathology matory affections. Our present knowledge concerning its arrangement and distribution is chiefly due to the researches of J. Marshall Flint, who has made a number of valuable observations on this subject, chiefly by means of the Spalteholz digestion process and with Mallory's stain. The surface of the gland, as we have seen, has no true capsule, but is covered by a loose thin coat of connec- tive tissue. Within the substance of the organ the con- nective tissue is arranged in an interlobular framework Fig. 36. — Piece digestion of a human pancreas, showing the Hmiting membrane of a lobule and the reticulated basement membranes of the alveoli. In the center is an island of Langerhans with its capsule of trabecula; (X 26) (Flint). of relatively large strands, which separate the lobes and lobules, and an intralobular network of finer fibrils, which lie between the individual acini and so form a plexus within the areas bounded by the coarser interlobular bundles. According to Flint, the interlobular connective tissue is much more delicate and less abundant than in the salivary glands, and is not fasciculated, except in the neighbourhood of the duct of Wirsung, nor are the connec- tive-tissue bundles so regularly arranged. The amount Histology 73 varies in different parts ; in some places only a few strands are found spanning the fissures, while in others relatively thick processes bind adjacent lobules together. The secondary lobules or lobule groups are usually separated by relatively wide bands of loose connective tissue, but the primary lobules, as pointed out above, are, as a rule, not clearly defined. The intralobular framework is of approximately the same form and size as in the salivary glands, although it is somewhat more delicate, but its arrangement is quite different. It con- sists of a fine network of delicate interlacing fibres stretch- ing across the lobules between the limiting membranes, and forming a reticulated basement membrane which supports the alveolar cells. The constituent fibres pursue an irregular course and are unequally distributed, being collected in some parts into small bundles, while in others they are seen as narrow strands. Near the islands of Langerhans the processes between the alveoli become thicker and stouter, forming septa which run into the capsule of the island. There is a slight amount of elastic tissue mixed with the fibrous framework, but it is almost exclusively confined to the interlobular regions, excepting around the ducts, which, even in the intralobular septa, are surrounded by a delicate network of elastic fibres. As the ducts unite and become larger the elastic tissue becomes heavier and thicker, but is never laminated as in the submaxillary gland. The connective tissue lying between the lobes and lobules contains a fair amount of fat. Connective-tissue cells, and occasionally mastzellen, are seen in the inter- lobular framework, and numerous cells with elongated or polygonal nuclei lie in the interalveolar connective tissue, as a rule on the side away from the lumen of the alveolus. The lobules do not posses a definite hilus, like those of the submaxillary gland, but receive their blood-vessels 74 The Pancreas: Its Surgery and Pathology and ducts by separate portals. These structures run, together with the nerves, in the intralobular framework as far as the spaces separating the secondary lobules, but within the secondary lobules themselves the ves- sels course independently of the ducts and enter the primary lobules at a different point. Both the blood-vessels and ducts are much finer structures than in the sali- vary glands. In radio- graphs of the pancreas taken after the ducts have been injected with mercury their extremely fine and delicate character is well demonstrated. By the cruder methods the ducts can probably be injected only as far as their lobular sections, but by forcing in coloured injections under pressure fine intercellular passages between the se- creting cells (Saviotti's ca- nals) can be made out. It has been contended that these fine ramifications are artifacts produced by the pressure, but the fact that Golgi's silver chromate method shows similar fine processes between the cells, and even extending into the cell substance (Schaffer), lends support to the results obtained by injection methods. The walls of the larger ducts consist of an inner thick, and an outer loose, coat of connective and elastic tissue. The epithelial lining is formed by a single layer of colum- Fig. 37. — Skiagram of a pan- creas after injecting the ducts with mercury (Royal Coll. of Surg. Museum). Histology 75 nar cells, which show only faint longitudinal striation. As the ducts diminish in size the connective-tissue coats become less marked and the epithelium assumes a more cubical character, until in the intermediate or intercalary portions it is seen as a single layer of flattened epithelium, the constituent cells of which appear spindle-shaped in 'i>^\ .^) "^-^' Fig. 38. — Origin of the ducts of the pancreas, as shown by the chromate of silver method (E. Miiller): A, Duct cut longitudinally, lined by columnar epithelium giving off laterally the intercalary or lobular ductules, m, to the alveoli, e. The manner in which these commence within the alveoli is shown under a higher power in B. section and do not stain w^ell with either acid or basic dyes. In the largest trunks small mucus-glands can be seen in the walls.- The minute structure of the organ is best studied in the lower animals, for preparations made from the human gland are rarely satisfactory, owing to the rapid changes that take place after death and the interval which usually 76 The Pancreas: Its Surgery and Pathology elapses before the material can be fixed in a hardening solution. It is also possible to investigate the condition of the gland in animals under various experimental con- ditions, which, while reproducing more or less closely those obtaining in the human subject in a variety of physiological and pathological states, cannot be secured at will in man himself. Although it is not strictly jus- tifiable to argue from the condition of an organ under any tory cells '^^^ ^J @ V %', Connective i^^^.^!^'^gs;i=:^^/\(^Mv ,s^~ tissue ' T7/^^'^^S^^ f,^S^f^&m\ 7 ^ Inner gran" ular zone of secre- tory cells Fig. 39. — From section through human pancreas; X450 (subHmate) (Bohm and Davidoff). particular set of circumstances in one animal to what may be expected in its homologue in another under similar conditions, and there are undoubtedly some differences to be observed in the histology of the pancreas in different animals, there is a sufficiently close resem- blance in all the higher members of the series to make the advantages of the method outweigh its possible defects. The alveoli of the pancreas are tubular or flask-shaped, Histology 77 and are lined by a single layer of columnar cells which taper somewhat towards their central extremities, where they abut upon the small irregular lumen of the acinus. The nuclei of the cells are centrally placed, as in the serous salivary glands, and there is also generally a spheri- cal para-nucleus. The latter consists of a portion of the protoplasm which stains more deeply than the rest, and is said to be formed by extrusion of material from the nucleus (Gaule, Nocolaider). The protoplasm of the cells contains numerous granules, which stain deeply with acid dyes, such as eosin. The quantity and distri- bution of these has been found to depend upon the state of the gland as regards its condition of "rest" and "activity." In the "resting," "charged," or "loaded" gland they occupy the inner or central two-thirds of the cells, while in the ' ' active " or " discharged ' ' gland they are compara- tively scanty, and are limited to the inner half. According to Heidenhain, during the first stage of diges- tion (six to ten hours) the granules gradually disappear and the granular inner zone diminishes in size ; in the sec- ond stage (ten to twenty hours) the inner zone is granular and greatly increased in size, while the outer is small, and during hunger the outer zone again enlarges. All the cells are not, however, in the same stage at the same time, and while in some the granular zone is narrow, in others it may be comparatively broad. The changes observed in the granules during digestion point to their being the zymogen, or precursor of the digestive ferment secreted by the pancreas. Kuhne and Sheridan Lea, Fig. 40.— Alveoli of rabbit's pancreas during rest (a) and dur- ing activity (b) (Kuhne and Lea). 78 The Pancreas: Its Surgery and Pathology watching the effect produced in the Hning gland of the rabbit by the injection of pilocarpin, found that secretion of pancreatic juice is accompanied by a diminution in the size of the cells and a discharge of the granules of the inner zone. According to Macallum and Steinhaus, the nuclei possess safranophilous nucleoli, and as the nucleus loses its safranophilous substance the cell substance ac- quires safranophilous granules. These authors conclude that the chromatin of the nucleus gives rise to a substance, pro-zymogen; sometimes it is dissolved in the nuclear substance, sometimes collected in masses (plasmosomes) ; finally it diffuses out into the cell protoplasm, and there meets with a constituent of the latter to form zymogen proper. The protoplasm between the granules only stains faintly with nuclear or basic dyes, but the outer clear zone stains well. The latter is of a homogeneous character, although in some instances it is seen to be faintly striated. Lying in the lumen of the acini, and sending processes between the secreting epithelium, are small spindle- shaped or branched cells, which, from their position and relation, are known as "centro-acinar cells." They act as supporting elements for the walls of the acini, and, according to Langerhans, are a continuation of the cells of the smaller duct radicles, to which they bear a strik- ing resemblance. Langerhans, in his description of the pancreas in 1869, first drew attention to those characteristic structures now known as "intertubular cell-clumps," "interacinar islands," or "the islands, or areas, of Langerhans." These are ovoid groups of small spherical or polygonal cells, which, in man, are apparently irregularly scattered through the gland substance, but in some animals, such as the cat, occupy a definite position in the centre of the lobules (Opie). In adult life no connection between the islands and the duct system of the gland can be made out, but Histology 79 they are found to be intimately related to the blood- vessels. The structure and relations of the interacinar islands have been the subject of numerous researches on the part of a large number of investigators, who, while agreeing on some points, differ in their descriptions in many impor- tant particulars. All those who have devoted attention to the subject agree that very similar structures are found in all vertebrates, but, while some regard them as permanent bodies prob- ably endowed with spe- cial functions, others look upon them as be- ing of a temporary na- ture and consider that they are in reality rest- ing acini. Harris and Gow in 1894 described three main types in different animals : 1 . Those in which the islands were not unlike lymphoid tissue, con- sisting of many deeply stained nuclei with lit- tle or no distinct cell protoplasm (e. g., the guinea-pig). 2. Masses of non-granular cells with distinct out- lines, which were joined in an irregular network {e. g., armadillo) . 3. Compound cell-groups in which the islands were divided by strands of connective tissue into smaller groups (e. <^., human). in 1899, however. Von Ebner stated that all these types could be found in one and the same animal, and suggested that the different appearances depended upon Fig. 41. — Microphotograph of nor- mal human pancreas showing an island of Langerhans and its relation to the blood-vessels ( X 50). 8o The Pancreas: Its Surgery and Pathology the amount of blood in the capillaries. Subsequent investigation has shown that the cells are of a similar type in all mammals and that the classification adopted by Harris and Gow does not hold good. The cells are always smaller than the gland cells, and each possesses a centrally placed round or oval nucleus. The nuclei differ from those in the secreting cells by being usually larger, relative to the amount of cell protoplasm, and having a very fine chromatin network with small nucleoli. The protoplasm of the well-defined cell bodies is very finely granular, containing numerous very small fat droplets. It does not stain at all with basic nuclear dyes, such as haematoxylin, but has some affinity for eosin and other acid stains. The appearance of the cells differs somewhat in some members of the vertebrate series ; thus, in birds the cells are generally small, oblong in shape, and stain very poorly ; in the frog the preponderant cells are tall and columnar, and are arranged in single rows between the blood-vessels, so that each cell is in contact with blood capillaries on two sides. This arrangement of the cells in rows between the blood-vessels, so charac- teristically seen in the amphibia, is found in the mammalia to some extent, although in them several rows of cells usually intervene between two adjacent capillaries. All observers are agreed that the islands of Langerhans are richly vascularised, but there is some divergence of opinion as to the nature of the vessels. The most recent observations are those of Pensa (1905) and Lydia M. De Witt (1906) . Pensa states that, as the result of injections of the blood-vessels in a large number of different animals, he was able to show that the islands are mostly supplied by a rich capillary network which is continuous with the intertubular capillary plexus. In some animals, such as birds, guinea-pigs, and dogs, the larger islets may have, in addition, a small afferent artery breaking up into a capillary plexus and then collecting again into a Histology 8 1 single efferent vein. As Lydia De Witt points out, however, he does not explain how he distinguishes the arteries from the veins, and while he states that the con- nection is, as a rule, purely capillary, one of his figures seems to indicate that the connection with larger vessels is common. By reconstructing the islands by the Born wax-plate method, and by a study of serial sections from injected and uninjected preparations, Lydia De Witt has made most valuable contributions to our knowledge of their morphology and histology. She comes to the conclusion that in all the animals she has investigated, including man at different ages, cats, rabbits, rats, birds, guinea- pigs, and frogs, the cords of cells forming the areas are separated by large irregular anastomosing vessels, having a complete endothelial wall, but little or no adventitia, thus corresponding to Minot's definition of "sinusoids." The endothelium of the sinusoids is directly applied to the epithelium of the islands, intervening connective tissue when present, as in the human adult pancreas, being secondary. The vascular network, according to her observations, is derived from the branching, winding, and anastomosing of several large venous channels and many capillaries which communicate intimately with the interacinar capillaries. The largest sinusoids are situated at the centre of the islands, where the cells are smallest. The periphery of the areas is much less vascular. Von Ebner also considers that the large blood-vessels of the islands are venous, and he points out that they are sur- rounded on all sides by cells like the blood capillaries of the liver lobules. By means of preparations made by Golgi's method Pensa showed that the islands of Langerhans are supplied with a very rich network of nerve fibres, which pass along the blood-vessels and between the cells. The number 6 82 The Pancreas: Its Surgery and Pathology u Fig. 42.— A, Wax reconstruction of areas of Langerhans from hu- man pancreas (X about 245); B, wax reconstruction of blood-vessels with surrounding connective tissue in same area (X about 245) (De Witt). Histology 83 and arrangement of the fibres were found to be quite different from those met with in the acini. Fig. r. A « Fig. 2. Fig. 43. — I, Interior of the model shown in A in the preceding figure; 2, interior of the model shown in B in the preceding figure (DeWitt). The relation of the cell islets to the excretory ducts of the pancreas has been investigated by Von Ebner, Kuhne 84 The Pancreas: Its Surgery and Pathology f( f. \ ^^^^5^4^,/^ and Lea, Lewaschew, and Dogiel. The last named made use of Golgi's method, while the others forced injection masses into the ducts. They all, excepting Lewaschew, came to the conclusion that, in the adult, the cell islets are not connected by permeable ducts with the excretory system of the gland, but Lewaschew found that some of the injection mate- ^"7,^ , rial passed within the islands. His results, however, are generally re- garded as having been due to acci- dental escape of the injection mass. Lydia De Witt, in her paper, states that the reconstruc- tion method and her study of serial sec- tions shows "that the cords of cells have the external form of branching and anastomosing tubules, with occa- sional alveolus-like enlargements ; they are, however, solid structures with no lumen and no arrangement of the cells and nuclei which would suggest a lumen." In speaking of the intralobular framework of the gland it was mentioned that Flint's investigations, by means of the Spalteholz digestion process, showed that near the islands of Langerhans the processes between the alveoli become thicker and stouter, forming septa which run c ^^, Fig. 44.- — Section of an island of Lan- gerhans from the pancreas of a rat in which the veins were filled with blood, showing the connection with the large vein and the arrangement of the sinusoids within the island (X 200)' (De Witt). Histology 85 into the capsule of the island. According to Flint, this capsule is a well-defined structure, and the connective tissue forming the framework of the islands has a char- acteristic arrangement in sharp contrast to that of the remainder of the lobule. He found that the capsule is composed of thousands of ultimate fibrils, which, on the one side, are connected with the alveolar network, and, on the other, with the septa or trabeculae which stretch across the space within the island, subdividing it into smaller lacunse and acting as a support for the cells of which it is composed. Every island in the gland has the same characteristic appearance and general conformity in the arrangement of its framework, and no transition stages can be found between the two. Laguesse, however, was unable to find a fibrous tissue capsule, but describes a thin, homogeneous layer, thick- ened in places, which forms a "pseudo-capsule." He also states that a thin amorphous sheath accompanies the principal vessels, and may be continued over the cap- illaries. Von Ebner was also unable, as a rule, to find any connective tissue, or membrana propria, between the capillaries and cells of the islets. These apparently contradictory statements are to some extent explained by the obser\^ations of Lydia De Witt. This observer found that in the frog no connective tissue can be demonstrated, either surrounding the areas or around the intra-insular sinusoids, but that in the guinea-pig, rat, and rabbit, sections stained with Mallory's stain show a very thin connective-tissue capsule separat- ing the island from the surrounding pancreatic acini, and delicate sheaths of connective tissue covering the blood- vessels, while delicate fibres also follow the contour of the cells. In the human subject the age and condition of the body appear to be most important factors in deter- mining the amount and distribution of the intra-insular connective tissue. The pancreas of the new-born infant. 86 The Pancreas: Its Surgery and Pathology according to her, shows most of the islands to be situated in the inter-lobular connective tissue, by which they are surrounded, but no connective tissue could be made out within the cell islets themselves. In a four-year-old child, although no connective tissue could be demonstrated with ordinary stains, Mallory's stain revealed a delicate capsule and delicate sheaths surrounding the blood- vessels. Around most of the cell-islands of the adult a rather definite capsule of nucleated connective tissue was found, and in the interior definite trabeculas, divid- Fig. 45_. — Piece digestion of a human pancreas, showing the connec- tive tissue of an island of Langerhans from Fig. 36 (X 135) (Flint). ing the islands into smaller compartments containing the cells, could be seen. In some instances the connective tissue formed, with the larger blood-vessels, one or many large trabeculge passing through the centre of the area, and from them smaller branches were given off to the sides, much in the same way as that described and fig- ured by Flint (Fig. 45). The very small amount of connective tissue found in the islands of Langerhans in animals and young persons, and the increase which apparently accompanies advanc- ing age in man, as a rule, suggest that it is probably a Histology 87 secondary effect analogous to the fibrosis which is usually associated with advancing years in other organs. The size and distribution of the islands is not uniform. Laguesse has distinguished five different types in man, varying from a very small form, less than 100 ji in diam- eter, to very rare, giant forms of over 400 // in diameter. He found, from an average of six bodies, that there is rather less than one island to each square millimetre of pancreatic tissue, and that about 0.0 1 per cent, of the gland is formed by the islands of Langerhans. Opie found that they were more numerous in the tail, or splenic end, than elsewhere in the human pancreas. He also agrees with Kasahara that the pancreatic tissue of the foetus and very young children shows a larger num- ber of islands than that of the adult. He states that this can be explained if it is assumed that they are formed during embryological development, and persist unchanged while the secreting tissue increases in bulk. Lydia De Witt, however, found that while about 0.02 per cent, of the pancreas of the adult consisted of insular tissue, it formed 0.04 per cent, of that of the four-year-old child and only 0.008 per cent, of the pancreas in the new-born infant, and that the average size of the islets in the adult was ac- tually greater than in either of the other two. The position of the islands with regard to the rest of the pancreatic tissue is not constant in the human sub- ject, although they are often situated in the centre of a more or less clearly defined lobule, but in the cat they occupy a position near the centre of the lobule, each of which, in the splenic portion, contains an island (Opie). Their distribution in certain bony fishes, particularly in Lophius piscatorius and Scorpoena scropha, is of consid- erable interest and importance, as bearing upon the questions of their origin and significance. Rennie states that very large islets were found in the areas of pancreatic tissue scattered along the abdominal vessels in all of the 88 The Pancreas: Its Surgery and Pathology twenty-five species he investigated, and that in Lophius piscatorius and Scorpoena scropha there was constantly present a very large, so-called "principal islet," indepen- dent of the pancreatic tissue, and surrounded by a fibrous tissue capsule, in the mesenteric fold between the portal vein and mesenteric artery, a short distance in front of the spleen. These principal islets are sufficiently large to be distinguished by the naked eye and can be dissected out free from pancreatic tissue. In the pancreas of the guinea-pig Lydia De Witt met with large, relatively isolated islets lying in the connective tissue around the large ducts, especially about the junction of the splenic and middle thirds, and, occasionally, in the mesenteric fat near the periphery of the gland, cell islets, which appeared to be free from the pancreatic tissue, were met with. The majority of the islets were, however, closely related to the pancreatic acini. The same observer noticed that in all the sections from the new-born infant examined by her the islets were situated in, and were surrounded by, the interlobular connective tissue. Light on the vexed question of the significance of these remarkable structures has been sought by a study of their development, but here again there is considerable difference of opinion. Hansemann believed that they arose from the interstitial tissue and had no connection with the pancreatic acini. Laguesse, studying sheep embryos, described a double origin for the islands. The so-called "primary islands" are said by him to arise from deeply staining units in the single layer of cells forming the wall of the primitive pancreatic tubules. By their proliferation these particular cells form solid outgrowths, which, later, becoming surrounded by the hollow outgrowths which bud out from the primitive tubules, constitute the primary cell islets. The secondary islands were believed by Laguesse, following Lewaschew, to be transitory structures developed from the acini and Histology 89 changing back into them again. Kiister states that they are derived from the ducts ; Pearce beHeves that they are developed from the pancreatic tubules, and are at first solid and later become vascularised, a reticulum develop- ing still later. Renaut states that the primitive dorsal and ventral duodenal outgrowths form solid branches which ramify in the mesentery. These later acquire a lumen, and from their walls groups of bhnd pouches arise, which constitute the secreting acini, each group of pouches representing a primary lobule of developed pancreas. In each group of pouches there appears a cell, similar to those described by Laguesse, and by the continued growth and multiplication of this the cell islet of the lobule is formed. The constant presence of these structures at all ages and in so many different animals, their early appearance in embryonic life, the manner in which they retain their vitality under varying conditions, their different staining reactions, and peculiar arrangement, have suggested that they are independent vascular glands, derived from the same embryonic rudiments as the secretory acini, but endowed with some special function. Although this is the view held by most recent writers who have devoted attention to the subject, there are others, as we have already mentioned, who regard them as temporarily changed acini which may again assume their former appearance and characters. The principal arguments advanced in favour of the latter hypothesis by its sup- porters are: (i) that the islets are closely related to the acini, from which they are not separated by any definite capsule, both structures have a common blood-supply, and the islets open into the pancreatic ducts ; (2) in the same sections various transition stages between typical acini and typical islets can be found; (3) the number of islets increases during activity of the gland and dimin- ishes during rest; (4) by prolonged stimulation of the 90 The Pancreas: Its Surgery and Pathology '^U^ y' gland, either by overfeeding or by the administration of pilocarpin, it is possible to transform secreting acini into islands of Langerhans; (5) if the pancreas of a guinea-pig is ligatured near the splenic end in two places, and portions are examined at intervals from between, behind, and in front of the ligatures, while all traces of gland substance disappear from the tissue between and behind the points of constriction, the cirrhosed portion in front shows as many, if not more, islands of Langer- hans than secreting acini, so that, al- ''"'* ^'* ^I^4p> though the results seen in the last named may tend to support the theory that the islands are independent and more resistant struc- tures than the acini, the same cannot be said for the remain- ing parts (Mankow- ski). The points raised under the first heading have .been discussed in consid- ering the structure of the islands, and it has been shown that the most recent observations do not lend support to these contentions. There is no doubt that in some of the lower animals, such as the rabbit, in which the capsule of the islands is very thin, places can be seen in which the island cells and the gland cells appear to be continuous, and that in the foetal pancreas affected with congenital syphilis the islands of Langerhans may be continuous with the surrounding secreting structures (Opie). But this can be explained, in the one case by :^'' Fig. 46. — Section through the center of an island of Langerhans from the pan- creas of a rabbit, showing a connection with the pancreatic tubules at "p" (X 200) (DeWitt). Histology 91 the more or less rudimentary condition of the organ, and in the other by the retarding effects of the pathological condition on development, for, as we have shown, it is probable that the secreting cells and the islets originate from a common epithelial anlage. With regard to the second point, it cannot be disputed that at times struc- tures suggesting transitional forms are met with, but by the study of serial sections it can generally be made out that these are either gland acini in which the staining reactions are abnormal and the characteristic differentia- tion into zones is absent, or cell islets in which the eosino- phile cells are more numerous than usual. Lydia De Witt states that in her experience she has seen none that could not be explained in some other and more rational way than by supposing that they were transition forms between secreting acini and cell islets, and that the so-called transition forms have proved to be merely resting pan- creatic tubules. The same observer has examined, measured, and counted large numbers of islands from a considerable number of guinea-pigs to determine the changes brought about by digestion and diet. The animals were killed about fourteen hours after eating and were kept upon (i) normal full diet, (2) without food or drink, (3) on pure carbohydrate diet, (4) on pure meat diet. She concluded that "while some qualitative changes were noted in the islets, — such as an increase or diminution of the eosinophile cells, a granular change in the cells, atrophy of the cells with increase of the inter- cellular substance, — there were none which could be regarded as constant for any one experiment and con- stantly increasing with the duration of the experiment." Opie, Schulze, Diamare, and Jarotzky have arrived at similar conclusions, while Hansemann believes that the apparent increase during digestion is due to a more marked differentiation arising from the changes in the acini. The statement that prolonged stimulation of the 92 The Pancreas: Its Surgery and Pathology gland by overfeeding, or by the administration of pilo- carpin, causes transformation of secreting acini into islands of Langerhans is based upon experiments carried out by Lewaschew in Heidenhain's laboratory. They have not, however, been confirmed by other observers. Statkewitsch, who has described alterations in the secret- ing acini in several specimens of animals under various conditions, thinks that they are merely the results of intense changes in the gland cells, and are not stages in a transition to cell islets. Jarotzky, as the result of his experiments, comes to the conclusion that the islands of Langerhans are independent structures, and are not connected with the altered gland acini met with as the result of altered dietetic conditions. He attributes the results obtained by Lewaschew to imperfect fixation. After the administration of pilocarpin Opie found that no increase in the number of cell islets could be detected, and that no transition stages between glandular acini and cell islet could be seen. He points out that in Lew- aschew's experiments the normal variations in the num- ber of islands in various parts of the gland, and in different glands, are not sufficiently taken into account. The more powerful physiological stimulus afforded by injection of secretin has recently been employed by Dale in investigating this subject. He states that the pro- longed administration of secretin produces changes in the gland cells of such a kind as to assimilate them in arrange- ment and properties to those forming the epithelium of the ductules and centro-acinar cells, thus bringing about reversion to an embryonic type. The lumina of the acini disappear and the cells are brought into more inti- mate relation with the blood-vessels. These altered masses of cells are regarded by Dale as being islands of Langerhans, and he states that numerous intermediate forms, retaining obvious traces of their former alveolar structure, can be found. He therefore agrees with Histology 93 Lewaschew that the cell islets are not independent structures, but, as Laguesse has suggested, represent an internally secreting stage in the life of the pancreatic tissue. Further investigation of the subject by this method is, however, desirable before the true interpreta- tion of the results described by Dale can be arrived at, and it is more particularly desirable that serial sections stained by appropriate methods should be examined, and that models prepared by the Born wax-plate method should be compared with those made from normal cell islets, for the evidence at present available cannot be regarded as conclusively demonstrating a structural con- nection of the secreting acini ~ with the cell islets in the adult forms of the higher types of animals. The experiments of Mankowski, in which the pancreas was ligatured in two places, were undertaken to disprove observations made by Schulze, in which it was found that the cell islets remained embedded in connective tissue, after the glandular acini had been destroyed as the result of the changes produced by tying the ducts. Man- kowski's conclusions have not, however, been supported by the experiments subsequently undertaken by Ssobolew, Sauerbeck, Zunz, and De Witt, w^ho confirmed the obser- vations originally made by Schulze as to the atrophy of the gland tissue and preservation of the cell islands after ligature of the excretory duct of the gland. Ssobolew also found that if portions of the gland are transplanted, the glandular parenchyma disappears, but the islands of Langerhans are extremely resistant and re- main for long unchanged. An interesting case has recently been carefully ex- amined and described by S, G. Scott, in which a condition, similar to that induced experimentally in animals by ligature of the pancreatic duct, was brought about by an obstruction due to a malignant growth of the head of the gland. The body of the organ was markedly atrophied 94 The Pancreas: Its Surgery and Pathology and the duct was dilated. Under the microscope an extreme degree of fibrosis was found, and a number of cell-groups, which, in serial section, had the appearance and character of cell islets, were seen embedded in the fibrous tissue, but no secreting glandular tissue could be distinguished. There was no evidence of diabetes, and * ^l^ Fig. 47. — Obstruction of the pancreatic duct by carcinoma of the head of the gland, giving rise to atrophy and fibrosis of the body with persistence of the islands of Langerhans (Scott) (X ca 30). the urine gave no reaction for sugar during life, in spite of the almost complete disappearance of the secreting parenchyma of the gland. Some observers, basing their theory on the microscop- ical characters of the cells, have regarded them as lymphoid structures, but their origin from an embryonic Histology 95 anlage, the arrangement of the cells, and their appearance in well-fixed preparations, at once differentiate them and suggest that such an opinion can only have originated from the study of imperfectly prepared specimens. That they are not embryonal remains, as some have supposed, is shown by the fact that they exhibit no evi- dence of degeneration in adult life, and further that, although they appear to be more numerous in the embryo and in early life than in the adult, the disproportion is only relative and not absolute. Literature. Dale: Proc. Roy. Soc, Ixxiii, 1904; and Phil. Tr. Roy. Soc. (B), cxcvii, 1904. De Witt: Journ. of Exp. Med., 1906, viii, 193. •Diamare: Internat. Monatschr. f. Anat. u. Phys., 1899, xvi, 155. Dogiel: Arch. f. Anat. u. Phys., 1893., Anat. Abt., 117. Ebner: Kolliker's "Handbuch der Gewebelehre des Menschen," 1899. Flint: Johns Hopkins Hosp. Rep., xii, 1904. Hansemann: Zeit. f. klin. Med., 1894, xxvi, 191. Verhand. der deutsch. path. Gesellsch., 1902, iv, 187. Harris and Gow: Journal of Physiol., 1894, xv, 349. Jarotzky: Virchow's Archiv, 1899, ^Ivi, 409. Kasahara: Virchow's Archiv, 1896, xxliii, iii. Kiihne and Lea: Untersuch a. d. phys. Instit. d. Univ. Heidelberg, 1882, ii, 488. Kiister: Arch. f. mik. Anat., 1904, Ixiv, i. Laguesse: Compt. Rend, de la Biol., 1893, xlv, 819; Ibid., 1894, xlvi, 667; Ibid., 1895, xlvii, 669; Ibid., 1905, Iviii, 564. Journ. de I'Anat., 1894, xxx, 591; Ibid., 1S96, xxxii. Langerhans: Inaug. Dissert., Berlin, 1869. Lewaschew: Arch. f. mik. Anat., 1886, xxvi, 452. Macallum: Quoted by Adami, Brit. Med. Journ., Dec. 22, 1906, 1763. Mallory: Journ. of Exp. Med., 1900, v, 15. Mankowski: Arch. f. mik. Anat., 1901, lix, 286. Opie: "Diseases of the Pancreas, " 1903. Pearce: American Journ. of Anat., 1903, ii, 445. American Medicine, 1903, vi, 1020. Pensa: Internat. Monatschr. f. Anat. u. Phys., 1905, xxii, i. Rennie: Quarterly Journ. of Micros. Science, 1904, xlviii, 379. Zeit. f. Phys., 1905, xviii, 23. Sauerbeck: Virchow's Archiv, 1904, clxxvii, Suppl. Heft. i. Verhandl. der deutschen path. Gesellsch. Erganzungsheft. Centl. f. path. Anat., 1904, XV, 217. Schaffer: Quain's "Elements of Anatomy," iii, iv. Schulze: Arch. f. mik. Anat., 1900, Ivi, 491. Scott: Journal of Pathology and Bacteriology, January, 1907. Ssobolew: Virchow's Archiv, 1902, clxviii, 91. Statkewitsch : Arch. f. exp. Path. u. Pharm., 1894, xxxiii, 415. Steinhaus: Quoted by Adami, Brit. Med. Journ., Dec. 22, 1906, 1763. Zunz: Zent. f. allg. Path. u. path. Anat., 1905, xvi, 5. CHAPTER VII PHYSIOLOGY The anatomical similarity of the pancreas to the salivary glands led the early observers to consider that their functions were also of the same nature, and it was not until Bernard pointed out, in 1849, that the pancreatic juice was concerned in the digestion of fats, and, in 1856, that it was also capable of acting upon proteid material, that the vastly greater importance of the pancreas as a digestive organ came to be recognized. The subsequent investigations of other workers upon the processes of digestion have shown that the pancreas is the digestive organ of the body par excellence, it is capable of dealing with all the chief forms of food material, its action is more energetic and complete than any other, and, more- over, it prepares for absorption substances, such as fat, which are little, if at all, changed by the secretions of the other digestive organs. The researches of Pawlow and his colleagues on the work of the digestive glands demon- strated in a masterly manner that the processes of diges- tion are not made up of a series of isolated phenomena, but that each step follows in an orderly manner as the result of the one which precedes it, and that to this rule the pancreas furnishes no exception. A study of the mechan- ism of the pancreatic secretion by Bayliss and Starling has resulted in the enunciation of a new principle concerning the co-ordination of its digestive functions with those of other parts of the alimentary tract, which has opened up a fresh field for research, and promises to throw light upon a number of hitherto obscure problems in other re- gions of the body. 96 Physiology 97 Although the digestive functions of the pancreas are undoubtedly of great importance in the due maintenance of the health of the organism, there is reason to believe that it exerts a still more important influence upon the internal metabolism, particularly through the control it exerts upon the assimilation of carbohydrate material by the tissues. The salivary glands, stomach, and intesti- nal bacteria may to a certain extent replace or supple- ment its digestive work, but, so far as we know at present, no other organ can take on its functions in carbohydrate metabolism. Analyses of the pancreas by Oidtmann show that it consists of 74.53 per cent, of water, 24.57 P^r cent, of organic matter, and 0.95 per cent, of inorganic substances. The same observer found in the salivary glands of the dog 79 per cent, of water, 20 per cent, of organic and i per cent, of inorganic matter. The organic matter of the pancreas consists of proteids (albumin, glob-ulin, and nucleo-proteid) , zymogens, nuclein, leucin, xanthin (1.8 p. m.), hypoxanthin (3 to 4 p. m.), guanin (2 to 7.5 p. m.), adenin, inosit, lactic acid, volatile fatty acids, and fat. The principal constituent of the cells, however, appears to be a complex nucleo-proteid, which Hammarsten regards as identical with trypsin. This when boiled gives a coagulated proteid and a phospho-gluco-proteid, and the latter on treatment with dilute acid yields a reducing substance having the characters of a pentose. Although small quantities of a pentose (1-xylose) can be obtained from most organs in the body, the pancreas yields over four times the proportion that can be obtained from any other structure. Neuberg found 2.48 per cent, of the dry weight of the pancreas was xylose, while from the liver and from the thymus only 0.56 per cent, could be obtained, the submaxillary gland yielded 0.53 per cent., the thyroid 0.5 per cent., and the kidneys, spleen, brain, and muscles under 0.5 per cent. 7 98 The Pancreas: Its Surgery and Pathology During life the organ is alkaline in reaction, but it very rapidly becomes acid after death ; at the same time small quantities of tyrosin make their appearance. Activity of the pancreas has been shown by Barcroft and Starling to be accompanied by an increased oxygen absorption, which is not due to the augmented blood- flow through the organ. Normally the oxidation in the pancreas is greater than in the body generally, being about the same as in the submaxillary gland. . Increased meta- bolism in the pancreas has been found by Bainbridge to be accompanied by increased lymph formation, and he has shown that there is a close relation between the secretion of pancreatic juice and the increased flow of lymph. The mechanism of the flow of the pancreatic secretion was first satisfactorily studied by Pawlow, by a method of obtaining the juice under practically normal conditions, which he described in 1879. In this method, which differs only slightly from that reported by Heidenhain in the following year, an oval piece of the duodenal wall containing the orifice of the pancreatic duct is cut out, and, after the lumen of the bow^el has been restored, is brought to the surface and stitched into the slit in the abdominal wall. The wound heals quickly, and, after two weeks, when the animals are ready for observation, shows a roundish elevation of mucous membrane in which the cleft-like orifice of the duct appears about the centre. By paying strict attention to cleanliness, regulating the diet, and adding a certain quantity of sodium bicarbonate to their food, to make up for the loss of alkali through the pancreatic fistula, such animals can be kept in good health for a lengthy period. Pawlow 's method overcame the difficulties which had beset attempts to investigate the mechanism and rate of secretion of the pancreatic juice under varying conditions by previous experimenters, for it allowed sufficient time for the animal to recover from the effects of the operation and the transitory interference Physiology 99 with its functions which had been found to result from the formation of a temporary fistula, while the inflam- matory changes which followed the older methods of forming a permanent fistula were likewise avoided. Employing dogs, provided with a pancreatic fistula in this manner, Pawlow investigated the effects of stimu- lating the nerves going to the pancreas, and results of variations of diet upon the secretion. He found that if the vagus in the neck be cut and left under the skin for four days, so that the cardiac fibres may degenerate, stimulation by a slow induced current, or by mechanical blows, causes a gradually increasing flow, after a latent period of three minutes. When the stimulation of the nerve is discontinued the flow does not cease at once, but continues in diminishing amount for four to five minutes. This part of the experiment can be done without an ansesthetic, thus avoiding any disturbing influences which might thereby be introduced. A slow induced current was employed, as it does not stimulate the vaso-constrictor nerves, excitation of which would diminish the blood supply and stop the secretion. If instead of resecting the vagus it is exposed and at once stimulated below the origin of the cardiac branches, after the cervical spinal cord had been cut to prevent reflexes from the sensory nerves, a similar flow of pancreatic secretion follows. In this so-called "acute method" it was found that simultaneous stimulation of the oppo- site vagus often had an inhibitory action, suppressing the secretion after a latent period, and that stimulation of the sympathetic at first slightly increased the amount of secretion, but soon brought it to a standstill. As the result of these experiments Pawlow came to the conclu- sion that the mechanism of the pancreatic secretion is arranged upon the same plan as that of the stomach and salivary glands, being determined reflexly, or psychically, through the cortex by impulses leaving the central loo The Pancreas: Its Surgery and Pathology nervous system and travelling by way of the vagi and splanchnic nerves to the gland. The failure of Heiden- hain and other observers to obtain a flow of pancreatic juice on stimulating the vagi and splanchnic nerves was attributed by him to the unphysiological conditions under which their experiments were carried out. A pupil of Pawlow, Popielski, found that the intro- duction of acid into the duodenum brought about a flow of pancreatic juice after section of both vagi and splanch- nics, or destruction of the spinal cord, or complete extirpation of the solar plexus, and came to the conclu- sion that there were local centres presiding over the secretion in the scattered ganglia of the pancreas, and that, since there was no secretion if the duodenum was cut across a short distance from the stomach, the most important part was situated near the pylorus. Wert- heimer and Lepage confirmed Popielski 's observation with regard to the effect of acid in the duodenum, and further reported that a similar result followed the intro- duction of acid into the jejunum, but that the intensity of the reaction diminished as the distance from the duode- num increased. This they endeavoured to explain by suggesting that the local centre for the duodenum lay in the pancreas, but that for the jejunum was probably situated in the solar plexus. These observers, in their attempts to unravel the problem of the means by which the acid produced its effect, also found that if it were injected directly into the circulation no secretion of pan- creatic juice ensued. The most serious defect of Pawlow's method of investi- gation was one against which he had guarded in his researches into the secretory mechanism of the stomach. In his experiments upon the gastric secretion he provided against the entry of food into the stomach by an oesopha- geal fistula, but no provision was made to guard against the entry of acid chyme from the stomach into the duode- Physiology loi num in the pancreatic experiments. This, however, has been shown by the researches of Bayliss and StarHng to be the most important, if not the only stimulus that induces the flow of pancreatic juice, Pawlow and his fellow- workers were well aware that the introduction of dilute hydrochloric acid into the duodenum brought about active secretion, so much so in fact that they made use of it as a crucial test for deciding the normal relation of the alimen- tary canal to the pancreas, but their minds were so imbued with the idea of a nervous control that they failed to recognize the true importance of their own observations in this direction. They attributed the result produced by the acid to excitation of the peripheral nerve-endings in the mucous membrane of the intestine. A second hypothesis considered by Pawlow, only to-be rejected, was that the acid was absorbed into the blood and carried to the secretory centres or gland cells, where it acted as a stimulant for the production of the secretion. His reasons for setting aside this explanation were that if it were correct, the alkalinity of the blood would be dimin- ished during digestion and not increased, as it is known to be; further, that experiment shows that when acid solutions are injected into the rectum the pancreas remains at perfect rest, and in the same way acids do not act upon the pancreas so long as they remain in the stomach. A third explanation has been offered by Bayliss and Starling and supported by convincing experimental evidence. After setting aside the nervous theory, by proving that the presence of hydrochloric acid in a liga- tured loop of the upper part of the jejunum, the nervous connections of which had been completely destroyed, brought about a copious flow of pancreatic juice, and, accepting the observations of Wertheimer, that the intro- duction of acid into the blood failed to excite secretion, they concluded that the acid must act upon the cells of I02 The Pancreas: Its Surgery and Pathology the intestinal mucous membrane and produce some sub- stance which, being absorbed into the blood, travels to the pancreatic cells and arouses them to activity. To prove this they scraped off the cells lining the mucous membrane of the upper part of the jejunum, rubbed them up in a mortar with sand and 0.4 per cent, hydrochloric acid, filtered the extract, and injected the filtrate into a vein. The result was a brilliant confirmation of their surmise, for a flow of pancreatic juice was produced which was even greater than that excited by the introduction of acid into the lumen of the intestine. This effect was Fig. 48. — The efifects produced by the injection of acid into a loop of small intestine after destruction of the nerves: a, Blood pressure; b, drops of pancreatic juice; c, signal marking injection of 50 c.c. of 0.4 per cent. HCl; d, time in ten minutes (Starling). found not to be specific, for an extract prepared from the upper part of the intestine of any vertebrate animal induced pancreatic secretion in the same, or any other, species into which it was injected. The activity of the extract prepared from various parts of the intestine was shown to diminish as the distance from the pylorus was increased, that from the duodenum being most active and that from the lower part of the ileum being entirely ineffective, thus agreeing with the observations of Wert- heimer and Lepage on the secretory activity induced by the introduction of acid into various parts of the gut. The chemical messenger, or harmone, which brings Physiology 103 about this reaction, and to which its discoverers have given the name of ''secretin,'' has not been isolated. Once formed by the action of acid, or boiling water, on the intestinal mucous membrane, it can be boiled, showing that it is not a ferment, neutralised or made alkaline, without being destroyed. It is readily oxidized, is not precipitated by the ordinary reagents for proteids, but is soluble in 90 per cent, alcohol in the presence of ether, although it is insoluble in absolute alcohol and ether. Pig 45. — Effects produced by the injection of secretin prepared from the intestinal mucous membrane: a, Blood pressure; 6, drops of pancreatic juice; c, signal marking injection of secretm; d, time m ten minutes (Starling). It is diffusible through animal membranes, and can be filtered through a gelatinised Chamberland filter. It is not precipitated by tannic acid, thus excluding bodies of an alkaloidal nature and di-amido compounds. This evidence points to secretin being a body of relatively small molecular weight, and not a colloid. It may be compared to the active principle of the suprarenal gland, adrenalin, which has been obtained in a crystalline form and the chemical constitution of which has been deter- mined. This is indeed what might be expected of a I04 The Pancreas : Its Surgery and Pathology substance which has to be turned out into the blood at repeated intervals in order to produce in some distant organ a physiological response proportional to the dose. Even after coagulation of the mucous membrane by heat or alcohol, secretin can be extracted by the action of warm dilute acid, but mere extraction with water or al- cohol, in which secretin is freely soluble, does not give an active solution. It is therefore concluded that the epithelial cells contain a precursor of secretin, which is insoluble in water, alcohol, and salt solution, termed ''prosecretin,'' and that this, on hydrolysis with acids, gives rise to the active substance. It has been found impossible to prepare secretin, or a substance having a similar action on the pancreas, from any organ or tissue of the body other than the mucous membrane of the duo- denum and the jejunum. The effect of secretin appears to be limited to the pancreas and liver, for while a solution, free from bile salts, on injection into a vein induces a marked flow of pancreatic juice and some increase in the excretion of bile, it is found to have no action upon any other gland. Bayliss and Starling believe that it acts as a direct chemical stimulant to the secretory cells of the pancreas, since the flow of secretion is still obtained when the gland has been cut off, as far as possible, from all nervous connections. The question as to whether secretin can be produced from the mucous membrane of the upper part of the small intestine by any other substance than hydrochloric acid, which is undoubtedly the most effective, has received attention at the hands of several observers. Fats were believed by Pawlow to be independent exci- tors of the pancreatic flow, for he pointed out that, since they restrain the secretion of gastric juice, the output of pancreatic juice which follows their administration is not likely to be indirectly due to acid in the gastric contents. Oil when rubbed up with duodenal mucous Physiology 105 membrane does not give rise to secretin, but Fleig has shown that if a solution of soap is employed instead, the mixture, on injection into the blood stream, gives rise to active pancreatic secretion. It is therefore possible that fats owe their activity as excitors of the pancreatic secretion to the formation of a certain amount of soap in the intestine, which in its turn sets free secretin. Fleig regards the secretin produced by the action of soap as different from that formed by acid, and has named it "sapocrinin," but there is no evidence to justify such a conclusion. Irritating substances such as oil of mustard, or ether, do not produce secretin from the scraped-off mucous membrane (Starling), but they produce a flow of pancre- atic juice on being introduced into a loop of small intestine, which Wertheimer's experiments show is due to the presence of secretin in the blood. Starling explains this by supposing that the secretin is formed by a process of hydrolysis in the over-stimulated cells of the intestine, possibly as a stage in their death. Most investigators have admitted that secretin is the most important exciter of the pancreatic secretion, but there are some who still believe that nervous activity plays some part in the process. Starling considers it doubtful whether the vagus has any direct secretory action on the cells of the pancreas, and points out that the normal effect of stimulating the vagus is to bring about movements of the stomach, which may cause its contents to flow into the duodenum, and there set up the chemical mechanism of secretion which he and Bayliss discovered. When due precautions are taken to prevent the stomach contents passing into the intestine, stimula- tion of the vagus produces such a slight flow from the pancreatic duct that it can hardly be regarded as evidence of the presence of secretory fibres in the nerve. Some io6 The Pancreas: Its Surgery and Pathology observations on the pancreatic secretion in a man have been described by Clayton-Greene, which he thinks support Pawlow's original theory that secretion is in- fluenced by a nervous mechanism. The case was one in which, during pylorectomy for malignant disease of the stomach, a portion of the pancreas was torn across, and three days later a pancreatic fistula • formed. Food given by the mouth was followed, some few seconds after it had been swallowed, by a definite secretion of pancre- atic juice, and the sight of food was also found to set up secretion. He considers that the conditions under which the flow started were such that it could not be explained as the result of the formation and absorption of secretin, and that in the observations made in this case there was support for the theory that stimulation of a sensory nerve could evoke a secretion from the pancreas as it does from the salivary glands. The composition and characters of the external secre- tion of the pancreas have been chiefly studied in dogs. That obtained by inserting a cannula into the duct two or three hours after a meal is found to be a clear, odour- less, colourless, syrupy fluid of a strongly alkaline reaction and a specific gravity of about 1.030. It contains from 2 to 15 per cent, of solid matter, of which a variable, but often considerable, proportion is coagulable proteid. The alkalinity, which is equal to 0.2 to 0.4 per cent, of sodium hydrate, is generally said to be due to carbon- ates and phosphates of sodium. Alkaline chlorides, and small quantities of calcium and magnesium phosphates, leucin, fat, and soaps are also present. The fluid readily decomposes on exposure to the air. The following analy- ses of the temporary secretion, obtained directly after operation by Schmidt, show that in the same form of fistula the proportion of total solids varies very considera- bly: Physiology 107 (a) («) Water 900.8 99.2 90.4 8.8 884.4 115. 6 Total solids Organic matter Ash The secretion from a permanent fistula, collected a few hours to several days after the operation, resembles that from a temporary fistula in its general characters, but is poorer in solids and coagulable proteid. It con- tains from 1.5 to 3.5 per cent, of the former and 0.5 to 2.5 per cent, of the latter. The specific gravity is also less, being generally about i.oio to i.oii. Schmidt's analyses of three specimens gave the following results : (.a) W (c) Water 976.8 23.2 16.4 6.8 979-9 20.1 12.4 7-5 984.6 15-4 Total solids Organic matter Ash 9.2 6.1 Starling states that the juice from a permanent fistula, after feeding, is similar in all respects to that flowing from a temporary fistula as the result of injecting secretin, or introducing acid into the duodenum. According to him, it is a somewhat viscid, clear, colourless fluid, of a specific gravity of about 1.030, and contains from 2 to 3.5 per cent, of total solids. About i per cent, of the solid consists of salts and the remainder of coagulable proteid. It is always strongly alkaline, 10 c.c. of the juice requiring from 10 to 15 c. c. of decinomal acid to neutralise it. That is to say, its alkalinity is equivalent to from 0.365 to 0.547 per cent, of hydrochloric acid, figures which correspond closely to the acidity of the gastric juice (0.48 per cent.). A certain proportion of the pro- teid is precipitated on neutralisation. In neutral solu- io8 The Pancreas: Its Surgery and Pathology tion about half the total proteid is coagulable at between 55° and 60° C, the remainder coagulating at about 75° C. Opportunities for studying human pancreatic juice are rare, and even in such cases as have occurred the material can hardly be regarded as normal, for it has usuall}^ been either the contents of a cyst or the drainings from a wound. In a case where the entry of the secretion into the intestine was prevented by the pressure of a malignant growth upon the duct of Wirsung, Herter found that it was a clear, alkaline fluid, without odour, and contained 2,41 per cent, of solids, of which 0.64 per cent, was soluble in alcohol. He separated 1.15 per cent, of peptone (and enzymes), but no other proteid, and 0.62 per cent, of mineral substances. The ash was found to be very rich in alkaline phosphates. Zawadsky analysed the pancreatic secretion of a young woman with a fistula that remained after a cyst of the pancreas, and found 864.05 p.m. of water, 132.51 p.m. of organic matter, and 3.44 p. m. of inorganic substances. The quantity of proteid was 92.05 p. m. The investigations reported by Glaessuer, on a case operated on by Korte, in which the pancreatic duct and common bile-duct were drained for eight days after an operation undertaken to relieve a simple stricture of the bile-passage following duodenal ulcer, will be referred to later, when the functions of the pancreatic juice are considered, but it may be mentioned here that the fluid was clear, alkaline in reaction, and of a specific gravity of 1.007. It contained 6 per cent, of ash and 15 per cent, of proteid, of which more than half was albumin. Different observers vary considerably in their esti- mate of the total daily output of pancreatic juice. Bidder and Schmidt state that the dog, under normal condi- tions, secretes 2.5 grams per kilo of body-weight a day, while Pawlow gives 21,8 c. c. per kilo as the normal out- put for twenty-four hours. On the basis of Bidder and Physiology 109 Schmidt's findings an average man of 154 pounds weight might be expected to secrete 175 grams of pancreatic juice a day, but in Glaessuer's case the daily amount collected was from 500 to 800 c.c, and in Wohlgemuth's 400 c.c. It is generally assumed that the amount lies between 200 and 500 c.c. Excepting in herbivora, such as the rabbit, in which digestion is uninterrupted, the secretion of pancreatic juice is intermittent. In a dog with a pancreatic fistula there is no secretion while the animal is fasting, but the administration of food, or even the sight of food, brings about a flow, after a latent period of two to three minutes. The flow of secretion induced by a meal gradually in- creases in amount until it reaches a maximum in two to three hours; it then diminishes, the lowest reading being reached in five to seven hours ; a second rise, reaching its maximum in the ninth to the eleventh hour, may then take place, after which it again gradually sinks until it finally stops about the eighteenth to the twentieth hour, unless a fresh supply of food is ingested. Both the amount and rate of secretion have been shown by Pawdow to vary with the nature of the food, Walther, working in Pawlow's laboratory, found that, taking the hourly quantity of pancreatic juice poured out for corresponding nitrogen equivalents of flesh, bread, and milk, the follow- ing results were obtained : With 100 grams of flesh . . . .38.7 44.6 30.4 16.9 0.8 131.45 c.c. With 250 grains of bread . . . 36.5 50.2 20.9 14. i 16.4 12.7 10.7 6.9 168.4 c.c. With 600 c.c. of milk 8.5 7.6 14.6 11. 2 3.3 i.o . . . . 46.1 c.c. When these results are plotted out in curves the varia- tions in amount and rate of secretion induced by the food materials experimented with are well seen (Fig, 50), Wohlgemuth, working with a patient with a pancreatic fistula the result of an operation after injury of the pancreas, also found that, employing fixed quantities of food and collecting the juice from hour to hour, the quanti- no The Pancreas: Its Surgery and Patholog}' tity secreted was greater with carbohydrates, smaller with albumin, and least with fats. The secretion was usually most active in the second hour. According to the secretin theory, the variations pro- duced by different classes of food material are not due, as Pawlow supposed, to some specific influence they V.Hi'ir '' I II HI IV V 1 iiiiiiv vvivnvirn niniv v v; IsBshasssssssBSi ■■■r~ Resh, 100 grins. Ifru.-ul, 'irrf) j-nns. \ Fig. 50. — Curves of secretion of pancreatic juice after 100 grams of flesh; 250 grams of bread; 600 c.c. of milk (Pawlow). exert upon the pancreas by way of its nerve supply, but upon the acidity of the chyme and its rate of dis- charge into the intestine. The first portion of the acid stomach contents passed into the duodenum will continue to excite the formation of secretin from the epithelial cells until the alkaline pancreatic juice, secreted in re- Physiology 1 1 1 sponse to its stimulating effect, has completely neutra- lized the acid; a second supply of chyme will then be ejected by the stomach, giving rise in its turn to the formation of secretin and of an alkaline pancreatic juice by which it will be neutralised, and this process will he continued so long as the products of the activities of the stomach continue to pass into the intestine. As a cer- tain amount of bile is secreted at the same time as the pancreatic juice, and the secretions of the intestine are also alkaline, a somewhat smaller quantity of pancreatic juice will be formed than would, by itself, be sufficient to neutralise the acid contents of the stomach. Fig. 51. — Effects of the injection of secretin on the flow of pan- creatic juice and bile: a, Blood pressure; b, drops of pancreatic juice; c, drops of bile; d, signal marking injection of secretin; e, time in ten minutes (Starling). Ferments. — The pancreatic juice contains four, or pos- sibly five, ferments or digestive enzymes: (i) "Amylop- sin," or pancreatic diastase, a diastatic ferment which converts starch and glycogen into dextrin and maltose, (2) "Trypsin," a proteolytic ferment which converts proteids into albumoses, peptone, and amino-acids. (3) "Steapsin," or "pialyn," a steatolytic, lipolytic, or fat-splitting ferment, Avhich converts neutral fats into fatty acids and glycerine. (4) A milk-curdling ferment, which in the presence of calcium salts changes caseinogen into casein; and, possibly, (5) a ferment which has been named "lactase," because of its supposed power of split- ting milk-sugar into galactose and dextrose. The collective pancreatic ferments are thus capable 112 The Pancreas: Its Surgery and Pathology of acting upoh all forms of food-stuffs and carrying to a final issue the changes commenced by other digestive glands. The same ferments are found in the pancreatic secretions of all vertebrates, but in the human subject, although trypsin is present during the last third of foetal life, the diastatic ferment does not make its appearance until a month or more after birth. The process of digestion of starches commences in the mouth through the agency of the ptyalin of the saliva, but it is quickly, although not immediately, stopped in the stomach through the precipitation of the diastatic ferment by the gastric juice. In the intestine the amy- lopsin of the pancreatic secretion continues the process, but much more vigorously and rapidly, exerting some action even upon unboiled starch. The main results of the activity of the two ferments are the same, however — - namely, dextrin and maltose. Only small quantities of dextrose are produced, even by the action of the more vigorous pancreatic ferments. The final conversion to this substance is brought about either through the activ- ity of epithelium of the intestinal wall or, possibly, by the action of a special ferment in the pancreatic juice, to which the name "maltase" has been given. The diastatic ferment of the pancreatic juice acts most satis- factorily at a temperature of 30° to 45° C. in a neutral or very faintly acid medium. The optimum reaction, according to Melzer, is about 0.0 1 per cent, of hydro- chloric acid. It is quickly destroyed by strong mineral acids and its action is suspended by 0.05 per cent, of lactic acid or 0.08 per cent, of acetic acid (Hofmeister) . Although Claude Bernard discovered the digestive powers of pancreatic juice upon proteids, he believed that the presence of bile was necessary, and it was not until Coivisart, in 1857, demonstrated that the juice alone exerted a powerful solvent action at the temperature of the body, and that infusions of the fresh gland possessed Physiology 113 the same property, that our knowledge of this function of the pancreas was placed on a secure footing. The latter observer also showed that the products of the pancreatic digestion of proteids had the same general characters as those resulting from the action of gastric juice. In 1877, Kiihne carried the investigations a step further by excluding the effects of bacterial action, of which account had not been taken by previous experi- menters. He found that the addition of a small quantity of salicylic acid to a pancreatic digestion-mixture pre- vented the growth of micro-organisms but did not stop the digestive process. He therefore concluded that the action was due to an enzyme and named it "trypsin^ It is now known that the pancreatic juice does not con- tain trypsin but trypsinogen, and that it is by the action of a substance contained in the succus entericus, known as " enter okinase," on the latter that the active ferment is produced. Different views are held as to the natiore of enterokinase. By its discoverers, Pawlow and Schepo- walnikow, it is regarded as a "ferment of ferments," and in this they are supported by Starling, but Delezenne, and others of the French school, consider that it is rather of the nature of an amboceptor binding the ferment to the proteid, Delezenne has recently stated that inactive pancreatic juice acquires an extremely powerful proteo- lytic action on being incubated for several hours with a suitable quantity of a soluble calcium salt ; barium, stron- tium, and magnesium have little or no effect and the ac- tion of lime salts in this respect is specific. Julius Wohlge- muth has also shown that the trypsinogen contained in the juice from a fistula is activated by glycine, alanine, and leucine, and feebly by tyrosine. Schiff, Herzen, Gachet, and Pochon have maintained that the spleen is of importance in the production of trypsin, but their conclusions have been disputed by Ewald and Heidenhain, and are not generally accepted. 114 The Pancreas: Its Surgery and Pathology The proteolytic ferment of the pancreatic juice can act in an alkaline, neutral, or faintly acid medium. The optimum is about i per cent, of sodium carbonate, other alkaline carbonates being found to be much less effective. Its action is prevented by the presence of free mineral acids, even in small quantities. Free hydrochloric acid destroys the ferment more rapidly if pepsin is also present, but hydrochloric acid combined with albumin, in not too large amounts, appears to rather increase the rapidity of its action. Organic acids exert a much less harmful effect than mineral acids, and it is said that in the presence of 0,2 per cent, of lactic acid, bile, and salt, the proteolytic action is very energetic. According to Wohlgemuth, the tryptic activity of pancreatic juice is doubled by the presence of bile. Small quantities of salicylic acid have no effect, and only saturated solutions interfere with its activity. The nature of the proteid also exerts some influence on the ease with which the digestive process proceeds ; fresh unboiled fibrin is attacked and dissolved exceedingly rapidly, but boiled fibrin or coagulated white of egg are digested much more slowly, hence it is advisa- ble to employ one or other of the latter when testing the digestive power of a fluid containing pancreatic ferments. The optimum temperature for the action of trypsin lies between 30° and 45° C. Beyond that its digestive power rapidly increases up to 60° C, but the ferment is at the same time quickly destroyed and loses its power, so that at 75° to 80° C. it ceases altogether. On exposing a proteid to the action of activated pan- creatic juice it is found to be attacked and eroded from the outside, without undergoing any swelling, or becom- ing clearer, as in gastric digestion. The alkali albumin first formed, when the digestion takes place in an alkaline medium, is quickly converted into deuteroalbumose ; this in turn gives rise to peptone, and from this again various amino-acids, and relatively simple nitrogenous Physiology 115 bodies, are derived. No proto-albumose or hetero- albumose can be detected, as in the earher stages of gas- tric digestion, probably because the action of trypsin is so much more rapid and energetic than that of pepsin that these bodies are broken down as quickly as they are formed. Kiihne as the result of his work on digestion considered that the peptone formed in the stomach differed from that produced in pancreatic digestion. The former he named "amphopeptone," because he believed that it consisted of two united groups ("hemi- peptone," which could be broken down by trypsin into simpler bodies, and "antipeptone," which was resistant to the action of trypsin) ; the latter, he stated, consisted only of "antipeptone." Kiihne himself, however, had doubts as to the unity of antipeptone, and it has now been shown that it is really a mixture of various amino-acids and hexone bases. Emil Fischer, working at the cleav- age products of proteid digestion, has isolated com- binations of amino-acids, which he terms "polypeptides," occupying an intermediate position between the pro- teoses and peptones, on the one hand, and the final products of digestion on the other. It is now generally accepted that "ampho-peptone," "hemi-peptone," and "anti-peptone," in the sense used by Kiihne, do not exist, and that there is no essential difference between the prod- ucts of the activity of trypsin and of pepsin. Both split up the proteid molecule by a process of hydrolysis into simpler combinations of a similar nature, but while the whole series of processes is rapidly and easily performed by the powerful tryptic ferment, the gastric juice only acts completely upon a variable fraction, which can be broken off with comparative ease. The small quantities of leucine, tyrosine, and other bodies of low molecular weight, now known to be formed during gastric digestion, are what were grouped together by Kiihne as anti-peptone. The only essential difference, therefore, between the ii6 The Pancreas: Its Surgery and Patdology proteolytic activity of trypsin and pepsin is in their velocity of action. It was formerly believed that the leucine, tyrosine, and similar bodies of relatively simple composition pro- duced during the digestion of proteids, were merely waste substances formed by the excessive activity of the pan- creatic juice, which were normally conveyed to the liver and there rapidly destroyed, but evidence is now rapidly acciunulating which tends to show that the main part of the proteid taken in as food is broken up into these simple cleavage products before it is absorbed, and that the body proteids are built up synthetically from them. That it is possible to maintain the weight, health, and nitrogenous equilibritun of animals by feeding them on the crystalline cleavage products of the pancreatic digestion of proteids has been experimentally demonstrated by Loewi and others, and the probability that such is the natural pro- cess by which proteids are absorbed does away with the difficulty of explaining how the constant chemical compo- sition of the various tissues of the body is maintained in spite of the widely differing nature of the food materials from which they are derived. Fresh pancreatic juice, which has not been activated by enterokinase, possesses slight digestive powers for proteids. This action, which is quite distinct from that due to trypsin, is akin to the feeble proteolytic property possessed by many animal tissues, particularly the kidney (Vernon) . Cohnheim has demonstrated the presence of a ferment in the succus entericus of the dog, which has the property of splitting proteoses and peptone into simpler products, but has no action on native proteids, and to it he attaches considerable importance. According to Kutscher, how- ever, it is a comparatively feeble and unimportant ferment. To this intestinal enzyme Cohnheim gave the name of " erepsin," but the same term has also been used Physiology 117 by Starling to describe the proteolytic ferment met with in fresh pancreatic secretion and in the tissue juices gen- erally. A similar ferment, found in the intestinal juice of suckling infants, may possibly be of value, as it is said to speedily break up caseinogen into casein. Collagen, the chief constituent of connective tissue, is not acted upon by pancreatic juice, unless it has pre- viously been boiled with water, or has been acted upon by dilute acid, hence connective tissue is not digested if the stomach has been removed, or if the secretion of acid is interfered with by disease. Gelatin ingested as such, or derived by previous digestive changes from collagen, is converted by trypsin into gelatin-peptones. Elastin is attacked and dissolved. Mucin and nucleo-proteids, after a preliminary cleavage into their constituent pro- teids and organic radicles, undergo digestive changes in the pancreatic secretion. In the fasting animal a mix- ture of pancreatic juice with the intestinal secretions is said to exert an exceedingly powerful action upon the wall of the intestine, giving rise to extensive inflammatory changes and erosions (Starling). The digestion of fat is peculiarly a function of the pancreatic juice. Neutral fats are entirely unaffected by the secretions of the salivary glands, and in the stomach are only slowly changed, yielding but i.o to 2.7 per cent, of fatty acid after some hours. Fine emulsions of fat, such as occur in the yolk of eggs and in milk, may be more completely digested, however, for Volhard found that the former, after one to four hours' stay in the stom- ach, might contain as much as 78 per cent, of free fatty acid, a fact probably of some practical importance in young infants before the lipolytic function of the pan- creatic secretion is fully developed, and in adult patients whose pancreas is disorganised by disease. The fat- splitting power of the stomach is believed to be due to a gastric lipase, although it has also been attributed to the ii8 The Pancreas: Its Surgery and Pathology action of bacteria. No fat-splitting ferment has been obtained from the intestinal mucous membrane or from the chyle. Eberle, in 1834, was the first to observe that pancreatic juice had the power of emulsifying fat, but it was Bernard who, in 1849, discovered that it had the property of split- ting fats with the liberation of fatty acids. The steapsin of the pancreatic juice is much less stable than the tryptic and diastatic ferments. It is very sus- ceptible to the action of acids, being quickly destroyed by all except the higher fatty acids. Strong alkalies also affect it unfavourably. It is most active in a neutral or weakly alkaline medium, even 0.25 per cent, of sodium carbonate retarding its activity. Unlike the other pan- creatic ferments, it is insoluble in water and in glycerine, so that its effects can only be studied by employing the fresh gland or the secretion, but that it is an enzyme is proved by its being destroyed on boiling and by its activity being maintained in the presence of antiseptics. The digestive power of the pancreatic juice for fats is considerably increased by the presence of bile, and still more by the presence of bile and hydrochloric acid. Bile of itself has little or no digestive power, but a mix- ture of bile and pancreatic juice can digest more than three times as much fat as the pancreatic solution alone. According to the experiments of Wohlgemuth, the lipase of the pancreatic juice exists partly in an inactive form, and this is activated by the accession of bile. The enzyme acts best at the temperature of the body, and, although more vigorous at higher temperature, up to a certain point, is quickly destroyed, like the other ferments of the secretion. Different views have been held from time to time as to the exact part the pancreatic secretion plays in the diges- tion and absorption of fat. Although Bernard discov- ered the saponifying action of pancreatic juice upon Physiology 119 neutral fats, he did not attach much value to it, but described a "ferment emulsif" to which he attributed the chief importance in the preparation of fats for ab- sorption. Briicke, however, found that the presence of a certain amount of free fatty acid was sufficient to emul- sify the remaining neutral fat, and arrived at the conclu- sion that the chief function of the fat-splitting ferment of the pancreatic juice was probably to provide the fatty acid for that purpose. Rachford has shown that under favourable conditions a sufficient amount of fatty acid is formed in the presence of bile and hydrochloric acid at room temperature to form a spontaneous emulsion in two minutes, thus explaining Bernard's results without invoking the aid of any special emulsifying ferment. In consequence of these and other observations it was held that only a small proportion of the ingested fat is split up into fatty acids and glycerine, and that this, aided by that naturally present in most fatty foods, converts the remainder into a fine emulsion which is absorbed by the intestinal epithelium and passed to the lacteals. The structure of the epithelial cells is not, however, suited for such a function, and fat globules have not been ob- served in their broad striated border, so that it is highly probable that the fat passes through the walls of the cells in a soluble form and is afterwards thrown down in visible particles. Zawarykin's suggestion that the fat is absorbed from the intestine by lymph cells and carried by them to the lacteals has been discredited, as also has the theory of Munk that a considerable portion of the fat is absorbed as emulsified fatty acids. It is now gen- erally acknowledged that fat is absorbed in the form of soluble soaps, although in some animals a certain por- tion may also leave the intestine as dissolved fatty acids. The neutral fats in the intestine are believed to be split up by the action of the pancreatic juice into fatty acids and glycerine ; the fatty acids then unite with the sodium, I20 The Pancreas: Its Surgery and Pathology potassium, calcium, and magnesium of the intestinal juices to form soluble alkaline soaps, and these are ab- sorbed, together with the glycerine, by the epithelial cells of the intestinal wall, within which they are again synthesised to form neutral fat. Miiller has shown that the macerated pancreas of the pig can split 86.4 per cent, of the fat of milk in twenty-four hours, and Rachford states that the steapsin of the pancreatic juice is prob- ably quite capable of splitting up all the fats of a full meal in the time digestion usually takes within the body. Radziejewski has also proved that alkaline soaps are absorbed by the intestine, and Perewoznikoff has dem- onstrated that alkaline soaps and glycerine are synthe- sised to neutral fats. The presence of bile makes the free fatty acids, which are entirely insoluble in water, soluble, and increases the solubility of the alkaline soaps. Its solvent power is greatly augmented by the presence of lecithin, but is mainly due to the bile salts it contains. Pawlow pointed out that there is a close relationship between the amount and rate of secretion of the bile and pancreatic juice, as he showed by the diagrams reproduced in Fig. 52. He also states that the amount of fat-split- ting ferment in the pancreatic secretion is dependent upon the quantity of fat in the food, basing his conclu- sion on experiments carried out by Walther. This observer found that in the first two hours after a meal of milk, a juice is furnished which is uncommonly rich in fat-splitting ferment, but that if the milk is deprived of its fat by filtration the juice presents a very low fat- splitting power, without any other alteration in the progress, or rate, of secretion; on again mixing the fat with the milk-filtrate, the fat ferment in the pancreatic juice is again increased to the previous amount. Recent advances in our knowledge of the influences governing pancreatic secretion make it doubtful whether such an adaptation of the enzymes to the quality of the Physiology 121 food as Pawlow and others have described does occur. In this connection, the investigations that have been carried out on the presence in the pancreatic secretion of a fer- ment which is said to convert lactose into galactose and dextrose are of interest. Weinland, working at this subject, found that a chloroform -water extract of the pancreas of dogs fed on a diet free from milk did not affect lactose, but that a similar extract made from the pancreas of animals which had been fed on milk for several days possessed the power of converting as much Fig. 52. — Curves representing the hourly secretion of I, Pancreatic secretion; II, entry of bile into the intestine: a, a, After ingestion of milk; b, b, after ingestion of flesh; c, c, after ingestion of bread (Pawlow). as 50 per cent, of the added milk-sugar. He also found that the pancreas of dogs fed on meat with the addition of milk-sugar contained lactose, but that the subcuta- neous injection of lactose for several days in succession had no effect in producing the ferment. He consequently came to the conclusion that the formation of "lactase" was not dependent upon any direct chemical action exerted by the lactose or its products upon the pancreas, but that the adaptation of the organ to a milk-sugar diet was brought about by a nervous mechanism. 122 The Pancreas: Its Surgery and Pathology After the discovery of secretin Bainbridge re-investi- gated this question, and agreed with Weinland that dogs fed on biscuits did not secrete a lactose-sphtting ferment in their pancreatic juice, whereas those that were fed on milk secreted it in every instance. Extracts of the mucous membrane of milk-fed dogs injected into biscuit-fed dogs were found by him to induce the appearance of lactase in the pancreatic secretion of the latter, so that the formation of the special ferment by the pancreas appeared to depend upon some substance produced in the intestinal wall in response to the milk diet. That this was not secretin was shown by the negative result obtained on injecting secretin prepared from the duodenal mucous membrane of milk-fed dogs into biscuit-fed dogs, and the positive reaction induced by the injection of secretin from the latter into animals previously fed on a milk diet. He finally arrived at the conclusion that the adaptation of the pancreas to a diet containing lactose was brought about by a chemical, and not a nervous, mechanism resulting from the action of the milk-sugar upon the intestinal mucous membrane, and, further, that the reac- tion was only slowly produced. He considered that secretin evokes the secretion of all the ferment present in the pancreas at the time it is injected, and that the composition of the juice as regards its ferments for any given meal depended mainly on the previous diet of the animal, and little, if at all, upon the composition of the particular meal, excepting in so far as the nature of the food determined the amount of hydrochloric acid secreted by the stomach. These conclusions with regard to the adaptation of the pancreas to the presence of lactose in the food have been adversely criticised by Bierry and Plimmer, who attribute the results obtained to faulty methods. Their criticism appears to be well founded and is supported by a consid- erable amount of careful experimental work. They agree Physiology 123 with PopeHski that the composition and amount of the pancreatic secretion are determined solely by the inten- sity and duration of the stimulus. Plimmer also points out that the work of Vasilieff and Walther, on which Pawlow based his inferences concerning the lipase and trypsin, was carried out before the discovery of entero- kinase, so that its value is seriously diminished. It may therefore be concluded that, at present, there is no evi- dence of any adaptation of the pancreatic secretion of the diet, the only factor that influences it probably being the amount of secretin formed and the duration of its flow into the blood. The milk-curdling ferment of the pancreatic juice is probably not of much physiological importance. In its general action it resembles renin, but the results produced by the two ferments are not exactly alike. Kiihne was the first to demonstrate that an extract made from the pancreas of the dog produced clotting in milk, but Gamgee pointed out that this does not prove that the secretion of the organ possesses the same property. Halliburton and Brodie found, however, that pancreatic juice does pro- duce a change in the caseinogen of milk, although the action differs from that of rennet in some particulars. They showed that the addition of pancreatic juice to milk at a temperature of 35° to 40° C. caused a finely granular precipitate of casein to form, but that the milk still remained fluid. On being cooled, however, it formed a coherent curd, which again broke up into fine granules when the temperature was raised. By the action of rennet this "pancreatic casein" can be transformed into true casein. Wohlgemuth, experimenting with the secre- tion from a pancreatic fistula, states that the rennetic ferment is mainly present in the form of a pro-ferment which is actuated by the intestinal juice and hydrochloric acid. From being considered as merely an organ accessory 124 The Pancreas: Its Surgery and Pathology to the digestive tract, the pancreas has come, in recent years, to be regarded as having a most important influence upon the metabohsm of the body. Not only may disease of the pancreas lead to wasting from imperfect digestion and absorption of food, but, in certain cases, it gives rise to glycosuria and other symptoms of diabetes, which suggest that it is of fundamental importance in the assimilation of carbohydrate materials by the tissues of the body. The mechanism of this influence is as yet a matter for surmise, but it is now generally assumed that, in addition to its external secretion of digestive ferments into the alimentary canal, the pancreas forms an internal secretion through which it controls carbohydrate meta- bolism. The elaboration of this internal secretion has been attributed by some to the islands of Langerhans, but as the discussion of this subject is intimately bound up with the relationship of the pancreas to diabetes, it will be considered later, under that heading. Literature Bainbridge : Proc. Royal Soc, Ixxii, 1903. Journ. of Physiol., xxxi, 1904. Brit. Med. Journ., Dec. 31, 1904, p. 1742. Barcroft and Starling: Journ. of Physiol., xxxi, 1904, 491. Bayliss and Starling: Proc. Royal Soc, Ixix, 352 : Ibid., Ixxiii. Journ. Physiol., xxviii: Ibid., xxix; Ibid., xxx; Ibid., xxxii. Bernard: Compt. Rend. d. sc, Paris, 1849, xxviii. Arch. gen. de Med. Paris, 1849, xix, 4. Ann. de chem. et de Physique, xxv, s^r. iii. "Lefons de Physiol, experimentale, " ii. "Mdmoire sur le pancreas," Paris, 1856. Bidder and Schmidt: "Die Verdauungssafte in der Stoffwechsel, " S. 245-. Brucke: Sitzungsb. d. k. Akad. d. Wissensch., 1870, Ixi. Clayton-Greene: Lancet, Feb. 11, 1905, p. 359. Cohnheim: Zeitsch. f. Biol. (1897), xxxvi, 129; Ibid. (1899), xxxvii, 443; Ibid. (1899), xxxviii, 418. Coivisart: "Collect de memoire sur une function peu connue du pan- creas," Paris, 1857. Delezenne: Brit. Med. Journ., Dec. 22, 1906, p. 1785. Eberle: v. Frey, in Arch. f. Anat. u. Physiol., Leipzig, 1881, S. 382. Glaessuer: Zeits. f. phys. Chem., Jan., 1904. Halliburton: Trans. Path. Soc. of London, Ivi, 159. Chem. Soc. Ann. Rep., i, 171. Halliburton and Brodie: Journ. of Physiol., xx. Hammarsten: "Text-book of Physiol. Chem.," tr. Mandel, 1902. Zeitsch. f. physiol. Chem., xviii, 213; Ibid., xix, 19. Physiology 125 Herter: Zeitschr. f. phys. Chem., iv. Hill: "Recent Advances in Physiol, and Bio-chemistry," 1906. Kossel: Zeitsch. f. physiol. Chem., viii. Kiihne: Lehrbuch, Virchow's Arch., xxxix. Verhandl. d. naturb. med. Ver. zu Heidelberg, 1876, N. F., i, 190; Ibid., 1877, N. F., i. 233. Melzer: Inaug. Diss., Erlangen, 1894. Neuberg: Ergebnisse d. Physiol., iii, 1904, 373. Oidtmann: Hammarsten's "Text-book of Physiol. Chem.," 1902, 283. Pawlow: "The Work of the Digestive Glands," tr. W. H. Thompson, 1902. Perowoznikoff: Centralbl. f. d. med. Wissensch., 1876, S. 851. Plirnmer: Journ. of Physiol., xxxiv, 1906. Popielski: Gazette clinique de Botkin, 1900. Rachford: Journ. of Physiol., 1891, xii, 72. Schafer: "Text-book of Physiology," i. Schrnidt: Maly, Hermann's "Handbuch, " v (2), 189. Starling: "Recent Advances in the Physiology of Digestion," 1906. Vernon: Journ. of Physiol., xxvii, 288; Ibid., xxviii, 378. Weinland: Zeitschr. f. Biologic, 1899, xxxviii, 607; Ibid., 1900, xl, 386. Wertheimer and Lepage: Journ. de Physiologic, lii, 365. Wohlgemuth: Biochem. Zeit., 1906, ii, 264. Zawadsky: Centralbl f. Physiol., v, 179. Zawarykin: Arch. f. d. gen. Physiol., Bonn, 1883, xxxi, 231. CHAPTER VIII PATHOLOGY Our present knowledge of the nature and etiology of the diseases to which the pancreas is liable has resulted •mainly from clinical observation and experimental re- search upon the lower animals. Relying upon the naked- eye appearance of the organ as seen on the post-mortem table, it has been assumed, until recently, that pancreatic lesions were among the curiosities of medicine, and were therefore of little or no practical importance. Since 1889, when Pitz published his work upon acute pancreati- tis and von Mering and Minkowski first brought forward the results of their investigations upon the relation of the pancreas to diabetes, it has been slowly recognised, however, that diseases of the pancreas are far from being uncommon, and that disturbances of its functions, besides interfering with the normal processes of digestion, may be responsible for signs and symptoms which had hitherto been unexplained or were usually referred to other causes. Hale White, searching the post-mortem records of Guy's Hospital for fourteen years, 1884 to 1897 inclusive, found that in the 6708 post-mortems performed during that period the pancreas had been regarded by the mor- bid anatomists in charge as diseased or injured in 142 instances; that is, in 2 per cent, of all cases dying in a large general hospital. In 55 there was primary or sec- ondary malignant disease of the pancreas; a cirrhotic hard or small atrophied gland was found in 45, and there were noticeable fatty changes in 9. Three of the remain- ing T,T, were instances of hemorrhage into the pancreas, 126 Pathology 127 3 were cases of pancreatic calculi, 2 were examples of pancreatic cysts, i was a hydatid cyst of the pancreas, in 3 there was suppuration of the gland, and 4 showed evidence of tuberculosis, while the balance was made up of a miscellaneous collection of more or less rare diseases and injuries presenting obvious lesions. Recent statis- tics based upon microscopical investigations have shown that in many instances the gross appearance of thp organ is unchanged in the presence of considerable alterations in its minute anatomy, and that acute and chronic in- flammations may bring about only slight and easily overlooked microscopical changes. Bosanquet found on examining sections of the pancreas from 170 cases dying in hospital from a variety of diseases, and ranging in age from four days to over ninety years, that a certain amount of fibrosis was present in 13 per cent. ; allowing, as he suggests, that some increase of fibrous tissue is the usual accompaniment of old age and that 10 per cent, of his cases, who were over forty years of age, may be excluded on that score, there still remained 5 per cent, under that age in which the microscope revealed a fibrosis, probably of pathological origin, as compared with the 0.7 per cent, recognised by the morbid anatomists at Guy's Hospital as showing evidence of chronic inflamma- tory changes. It is evident, therefore, that post-mortem records cannot be relied upon either for precise informa- tion as to the relative frequency of diseases of the pancreas or for data on which to base an accurate estimate of the true importance of the various morbid conditions to which it is liable, unless the results of naked-eye observations have been checked by microscopical examinations, and, since this is so rarely done even at the present time, there is not a sufficient mass of systematic observations to allow of any satisfactory conclusions being reached upon these points, apart from clinical evidence. The difficulties met with in the recognition of morbid 128 The Pancreas: Its Surgery and Pathology conditions of the pancreas in the post-mortem room are in part due to the rapid changes that take place in the organ at and after death, and to the appearance and colour of the gland as seen at autopsy. Apart from the putre- factive changes which rapidly supervene under suitable conditions, the pancreas is said by Chiari to be more or less altered by a process of auto-digestion that sets in immediately before, or shortly after, death in some 50 per cent, of cases. In this condition the organ is white and flaccid and, under the microscope, is found to stain uniformly with acid dyes. These difficulties are to a large extent obviated in experimental researches upon animals and in observations made on man in the course of abdominal operations. Experiments upon animals can be conducted under the most favourable circumstances, and give at least some indication of what may be expected in the human subject under similar conditions, while the surgeon has the advantage over the pathologist that he can examine and handle a living organ unaffected by self -digestion and post-mortem changes, although his investigations must naturally suffer from not being as searching and conclusive as the pathologist is able to carry out. Only those who have seen and handled an inflamed pancreas can realise how an engorgement and swelling of the gland which was perfectly evident at the time of operation may, if the organ is examined a few hours later, on the post-mortem table, have entirely disappeared, leaving a structure that is pale and flaccid, presenting, in fact, no marked deviation from the normal. The pathological conditions met with in the pancreas resemble in many respects those encountered in the liver, and in some instances a similar morbid state is found in both organs. Their parenchyma is liable to be affected by the same degenerative changes, and chronic inflamma- tion in the one gives rise to results very similar to those induced in the other. The ducts of both organs are not Pathology 129 uncommonly involved in the same pathological processes, and the morbid influences reaching the one are also liable to affect the other. There are, however, certain affections of the pancreas to which no strictly analogous condition can be found in the liver, and it is to these that attention has been largely devoted in most text-Vjooks of medicine and pathology. In view, however, of the important part that the pancreas takes in the processes of digestion and internal metabolism, the neglected lesions, that are often masked by the more obvious and striking affections of other organs, are no less worthy of attention, for, although at first sight the pancreas may appear to play but a subsidiary part in the production of the symptoms met with in some of these conditions, it becomes increasingly plain, as our methods of investigation are improved, that the pancreatic lesions are responsible for much that has hitherto been attributed to disease of other organs, or been altogether unexplained. Atrophy. — In common with other organs of the body, the pancreas is liable to diminish in size and weight with advancing years, and as the result of chronic diseases and marasmus its bulk may be reduced considerably below the normal standard. Senile atrophy of the gland is found, in many instances, to be accompanied by sclerotic changes in the vessels supplying it with blood, while in others it is no doubt the consequence of general malnutri- tion, such as also gives rise to the condition in chronic wasting diseases. Atrophy of the pancreas is met with in a considerable number of cases of diabetes. According to Hansemann, diabetic and cachectic atrophy can be distinguished both macroscopically and microscopically, and while the former always gives rise to diabetes, the latter only does so in advanced cases. The pancreas in diabetic atrophy is said to be usually flabby and somewhat dark in colour. Its bulk is espe- cially diminished in a transverse diameter, so that it 9 130 The Pancreas: Its Surgery and Pathology assumes a fiat shape. The lobules are small, and the sur- rounding connective tissue and fat extend into the organ so that it is often only removed with difficulty. At times large adhesions and new-formed bands connect the pan- creas with surrounding structures. Microscopically the secreting cells show no particular change, apart from the atrophy ; there is no marked opacity, fatty degeneration, or pigmentation. The stroma is scanty, but the gaps caused by the diminution in size of the gland lobules are more or less obliterated. Although the organ is largely fibrous, there are here and there patches of recent cellular infiltration, giv- ing rise to a condition resem- bling that met with in certain forms of granular atrophy of the kidney. In cachectic atrophy, on the other hand, the adjacent fat tissue has disappeared to a de- gree corresponding to the gen- eral emaciation, and the gland is sharply defined from the sur- rounding structures. In shape it is cylindrical, its thickness and its height being about equal. It is not flabby, but of a firm or moderate consistency. Under the microscope both the lobules and the individual cells are small, the stroma is atrophied and scanty, and the cells are not especially pigmented. Although these observations of Hansemann's summarise the points by which cachectic atrophy may be distin- guished from other conditions in which diabetes is accom- panied by changes in the size of the gland, it has not been shown that his "diabetic atrophy" is a pathological entity, Fig- 53-— Atrophy of the pancreas (Univ. Coll. Hosp. Museum, 3194 A). Pathology 131 or that the condition he describes under that name is in- variably accompanied by diabetes; in fact, it is probable, as we shall show later, that there is no special form of dia- betic atrophy, but that atrophy of the pancreas arising from a variety of causes may be accompanied by glycosuria. One of the commonest causes both of pancreatic atrophy and of diabetes is chronic interstitial pancreas, and it is to this probably that the so-called atrophy of diabetes, described by Hansemann, is in most instances due. Williamson, Opie, and others have described cases of diabetes in which the pancreas was diminished n size to an extent bearing no relation to the wasting of other organs, although no changes could be observed in the structure of the gland. Opie suggests that in these instances the condition is possibly congenital, and that, since the pancreas is unusually small, it fails to meet the demands upon it at some period of life, so that diabetes results. He considers that when the weight of the pan- creas falls below 65 grams (2 ounces) it is abnormal. A condition apparently due to congenital deficiency of the pancreatic functions has been described by Byrom Bramwell under the name of "pancreatic infantilism." The patient in whom this diagnosis was made was a youth of nineteen whose bodily development had apparently been arrested about the age of eleven years. He was bright and intelligent, perfectly formed, and presented none of the physical alterations suggestive of sporadic cretinism. The abdomen was swollen and tympanitic, and for nine years before he came under observation he had suffered from chronic diarrhoea. The urine was free from sugar. From careful investigations of the urine and fceces it was concluded that the pancreatic secretion was defective or completely absent. That this was the case was proved by the remarkable improvement brought about by the administration of a glycerine extract of pancreas, for as the result of this treatment the stools 132 The Pancreas: Its Surgery and Pathology were reduced from five or six loose motions a day to two, one of which was formed; in two years he grew five inches, and increased i stone 8 pounds in weight, although for the previous eight years he was said not to have grown at all; the sexual development, which before treatment was begun was infantile, progressed in a normal manner; the patient looked much older, and his voice, which had previously been high-pitched and childish, be- came of low tone and rough. Thomson has since described two apparently similar cases in males and Ren- toul has recorded the case of a fe- male with similar symptoms. The first of Thom- son's cases was a man of twenty- four who was about the size of a boy of ten ; the second was a boy of eighteen who resembled a child of eight or nine years in size and development; both suffered from chronic in- tractable diarrhoea. In Rentoul's case the patient was a girl of eighteen whose parents complained that for seven years she had not grown, and that she had been troubled all her life with diarrhoea. This patient was Fig. 54. — Pancreatic infantilism; a, Be- fore treatment; b, after treatment (Byrom Bramwell). Pathology 133 also much benefited by pancreatic extract, putting on 9^ pounds in weight and adding almost 2 inches to her height in a little over four months, besides developing sexually and improving in her general condition. Whether pancreatic infantilism is due to congenital atrophy of the pancreas or, as has been suggested, to general fibrosis from congenital syphilis, it would appear that while in patients who present such symptoms the pancreas is sufficient for the metabolic needs of the body for the first eight or nine years of life, it is subsequently un- able to keep pace with the calls upon it, so that the gen- eral nutrition and development suffer in consequence. It is worthy of note, however, that the deficiency appears to be confined to the • digestive func- tions of the gland and does not interfere with carbohydrate metabolism, for none of the re- corded cases have had glyco- suria. Whether diabetes will subsequently develop or not it is as yet too early to say, but the history of a case in this connection would be well worth following up as growth advanced. As the result of pressure exerted from without by aneurysms, new-growths, etc., the pancreas may undergo secondary atrophic changes, and similar consequences may also follow the chronic interstitial inflammation accompanying pancreatic calculi, pancreatic cysts, hsemor- rhage, or abscess formation. In some instances the changes may be so great that the gland tissue almost Fig- 55- — Fibrosis of the pancreas (St. George's Hospital Museum). 134 The Pancreas: Its Surgery and Pathology entirely disappears and the organ is only represented by a small mass of fibrous tissue. Fatty Infiltration and Degeneration. — The interstitial connective tissue of the pancreas normally contains a certain amount of fat, and this is liable to increase under pathological conditions. The increase that takes place in simple cachectic atrophy has already been re- Fig. 56. — Chronic atrophic pancreatitis. Island of Langerhans, sur- rounded by fat, instead of acini (Deaver and Miiller). ferred to. In general obesity, especially when con- nected with alcoholism, a similar overgrowth of fat is frequently encountered. This condition is often com- bined with fatty degeneration, and, in extreme cases, may result in the whole organ being transformed into a mass of fatty tissue. It is then of a yellow or yellowish- white appearance, soft, and somewhat larger than normal. Pathology 135 On section it is found to be lobulated and to consist of masses of fat separated by more or less well-marked strands of fibrous tissue in which the remains of the larger ducts, and perhaps some remnants of the gland structures, are embedded. Fatty degeneration is caused most frequently by inflammation of the gland, but it also occurs in infectious diseases and toxaemias, and may result from poisoning by phosphorus or mineral salts. Extreme degrees of fatty change are frequently found associated with pancreatic lithiasis. Fatty degeneration of the parenchyma is preceded in the first instance by cloudy swelling, the cells microscopically being found to be somewhat enlarged, opaque, and very granular. To the naked-eye the gland is hypersemic and enlarged. At first it is hard to the touch, but, as the degenerative process advances, becomes softer and of a white or yellowish-white colour. Under the microscope the epithe- lium is then found to contain numerous fat globules and the interstitial tissue to be oedematous. Amyloid degeneration of the pancreas occurs under the same conditions as in other organs, and is associated with a similar lesion in other tissues of the body. It primarily affects the small blood-vessels, but may eventually involve the larger vessels and the membrana propria of the acini. The gland cells undergo fatty degeneration and, in part, disintegrate. Hennigs found, in one hun- dred and fifty-five cases of general amyloid disease, six in which the pancreas was affected. Rokitansky states that sometimes the degenerative changes are limited to the pancreas and may affect the secreting epithelium, but this is denied by Friedrich and Kyber. Opie suggests that the condition observed by Rokitansky in these cases was in reality hyaline degeneration. Hyaline degeneration of the parenchyma of the pancreas has been described by Saunby in a case of diabetes, and Opie has also published an account of a diabetic whose pan- 136 The Pancreas: Its Surgery and Pathology areas showed patches of hyaHne degeneration, apparently replacing the islands of Langerhans, but also affecting the secreting parenchyma. Other cases of hyaline degenera- tion have since been recorded by a number of investigators. All have been associated with diabetes and the degenera- tive changes have been said to have been limited to the islands of Langerhans. According to Opie, hyaline de- § I / % ' r J* :r ««.© "«©! rv V Ac#%.%*^#l. ,/ \wi^ -^-\/-^ T..'/'?' Fig. 57. — Hyaline degeneration of the pancreas (Opie). generation first manifests itself by an increase in the size of the cells of the islets and an alteration of their pro- toplasm. With the death of the cells their nuclei disap- pear, and the cell protoplasm, which stains with acid dyes, remains for a time granular, but subsequently becomes homogeneous. The small masses of hyaline material then fuse with one another and form large collec- tions which lie in contact with the fibrous septa of the Pathology 137 island. After complete transformation of its cells the island is found to be represented by a hyaline mass, penetrated by the remains of altered capillaries. In some instances the degenerative process may spread to the secreting parenchyma, but, as a rule, it is limited to the cell islets. The hyaline material stains with eosin, picric acid, and other acid dyes, but shows no affinity for nuclear stains; on treatment with iodine, gentian- violet, methyl-violet, or iodine-green it does not give the amyloid reactions, and it resists the action of strong acids and alkalies. Unlike amyloid changes, hyaline degenera- tion of the pancreas is limited to that organ, for the blood- vessels of the liver, 'spleen, and kidneys appear to be unaffected. Its etiology is uncertain, but the fact that, in most cases, there has also been chronic interstitial pancreatitis of the interacinar type has suggested that it may possibly be due to interference with the circulation in the cell islets. Opie, however, is of opinion that both lesions are due to some irritant carried to the pancreas by the blood. Focal Necrosis. — In the body of a male negro, who had suffered from diabetes, Opie met w4th a condition of the pancreas resembling the focal coagulation necrosis fre- quently observed in the liver in typhoid fever and other infections. To the naked eye the organ presented no notable abnormality, but microscopically there was some increase of the connective tissue, and foci of necrosis, in- volving a considerable number of acini, were found in the parenchyma. In some places the islands of Langerhans were found to be implicated in the process, and, rarely, the cells of an island had undergone necrosis while the surrounding acini were normal. The affected cells pre- served their identity and were not fused into homoge- neous masses, as in hyaline degeneration, but they had lost their nuclei and stained deeply with eosin. Local hcEfnorrhages into the tissues of the pancreas are 138 The Pancreas: Its Surgery and Pathology relatively frequent. In some instances they are asso- ciated with extravasations of blood in other organs and result from circulatory disturbances, due to diseases of the heart, lungs, and liver, or from altered conditions of the blood, such as occur in infectious diseases, purpura, scurvy, phosphorus-poisoning, etc. Diseases of the blood- vessels, such as atheroma, and fatty degeneration, or alcoholic or syphilitic arteritis, may also be associated with haemorrhage into the gland. Among other causes of pancreatic haemorrhage may be mentioned fatty degeneration of the gland cells, with a deposit of fat in the pancreas, the result of alcoholism or of general obesity, fat necrosis in the gland or its vicinity, disintegration of neoplasms, embolism of a pancreatic artery, and inflam- mations of the gland, which last will be considered in detail subsequently. Although the pancreas lies in a sheltered position within the abdominal cavity, its tissues are comparatively soft and easily bruised, so that even slight injury takes more effect upon it than upon firmer organs, and may give rise to an effusion of blood. This susceptibility of the pancreas to injury is shown by the effects of manipulations of the gland in animals, and must be borne in mind in conducting operations for gall-stones in the common duct, for it is then often necessary to manipulate the head of the pancreas rather freely. Large pancreatic haemorrhages are of great clinical interest and are probably more common than is usually thought. They may occur in the substance of the gland and disintegrate it, or on the surface and lead to extensive effusion, either beneath the peritoneum or into the lesser sac. Besides the local hemorrhages not associated with inflammatory change, which may be termed "pancreatic apoplexy," and the form accompanying acute inflamma- tion, known as "hemorrhagic pancreatitis," there is, in many pancreatic affections, a tendency to" general haemor- Pathology 139 rhage from wounds or mucous surfaces, and to petechial haemorrhage into the skin, or to more extensive bleeding into the subcutaneous tissue. It is well recognized that a hasmorrhagic tendency coexists with cancer of the head of the pancreas, and it is generally thought to be alto- Fig. 58. — Haemorrhage into the pancreas produced by the injection of zinc chloride, showing how the blood infiltrates and breaks up the gland tissue: a, Blood-vessel distended with blood; b, extra vasated blood in the parenchyma between the lobules; c, normal pancreatic tissue (Oser). gether dependent upon the attendant cholasmia. There is, however, much less danger from haemorrhage in patients jaundiced from gall-stones than in those in whom the jaundice depends upon disease of the pancreas. The haemorrhagic tendency is, moreover, also present, although perhaps not to quite the same extent, in some pancreatic affections not associated with cholasmia. Investigations I40 The Pancreas: Its Surgery and Pathology of the blood that we have carried out in a number of cases of cancer of the head of the pancreas and in inflammatory affections of the gland, with and without jaundice, have shown that the coagulation time of the blood, as estimated by Wright's method, is considerably delayed and the number of blood platelets markedly diminished. It is well known that a diminution in the lime salts of the blood interferes with its power of coagu- lating, and that the prolonged coagulation time and tendency to haemorrhage in pancreatitis are probably due to a great extent to this cause is suggested by the benefi- cial effects resulting from the administration of calcium chloride. Further evidence, pointing in the same direc- tion, is also afforded by the composition of pancreatic calculi, which are peculiarly rich in calcium, and by the very frequent presence of a large deposit of calcium oxa- late crystals in the urine in cases of pancreatitis. Inflammatory Affections. — Judging from post-mortem records, inflammatory affections of the pancreas must be considered as amongst the rarest of diseases, but recent clinical observations and operative experience have shown that such a conclusion would be far from being the truth. As far back as 1672, Tulpius described a diffuse pancreatic abscess of pyaemic origin, and Matthew Baillie, in a work on "Morbid Anatomy," figured what he called a hard pancreas, with the lobules distinct, but which would now be considered as an example of chronic pancreatitis. In the same work Baillie also gives a drawing of a pancreas in which concretions were discovered in the ducts post- mortem, and which shows the changes that accompany them, as well as the relations of the bile and pancreatic ducts, in a striking manner. Portal in 1804 described a case of acute suppurative pancreatitis, following an attack of gout in the feet, and Percival in 1818 recorded a well- marked case of pancreatic abscess associated with jaun- dice. Pathology 141 Acute pancreatitis with fat necrosis was first described by Balser in 1879, but it was not until 1889, when Fitz published his classical papers, that the attention of the medical world was really aroused and inflammatory dis- eases of the pancreas came to be carefully studied at the bedside and in the laboratory. Much experimental work has been since devoted to the investigation of acute pan- creatitis, for opportunities of studying the disease clinically, although more frequent than has been supposed, are still not common. Our present knowledge of the chronic inflammatory affections of the pancreas, however, while dependent to a cer- tain extent upon the re- sults of experiment per- formed in animals, is mainly due to clinical re- search and observations made, in the course of op- erations upon the biliary tract. Experiment has shown that the injection of a variety of substances into the pancreas, either directly into the parenchyma or through the duct of Wirsung, gives rise to severe, and often rapidly fatal, inflammation, the severity of the lesion depending upon the nature and amount of the substance injected. The most varied and successful experiments have been carried out by Flexner, who found that dilute hydrochloric, nitric, sulphuric, or chromic acid, solutions of caustic Fig. 59. — Acute pancreatitis (Univ Museum, 3194 B). hemorrhagic Coll. Hosp. 142 The Pancreas: Its Surgery and Pathology alkalies, formalin, or suspensions of bacillus pyocyaneus or of bacillus diphtheriae, when injected into the pancre- atic duct of dogs gave rise to acute inflammation of the gland, which, in rapidly fatal cases, was commonly accom- panied by hsemorrhage, fat necrosis, and glycosuria. In those instances in which the animal survived for some time necrosis of portions of the gland, abscess formation, and, occasionally, chronic interstitial inflammation were met with. Subsequently Flexner and Pearce showed that the introduction of artificial gastric juice into the duct gave rise to similar changes, but that injections of sterile blood, while they occasioned an increase of fibrous tissue, did not produce acute inflammatory changes. Previous to these experiments by Flexner and Pearce, acute hsemorrhagic pancreatitis had been produced in dogs, by Hlava, from injections of artificial gastric juice, and he suggested that the cause of acute pancreatitis in man lay in the passage of hyperacid gastric juice into the pancreatic duct through anti-peristaltic action in the intestine. There is, however, no evidence which would lend support to such a view, and it cannot be accepted as a likely explanation. Hlava and Carnot, as well as Flexner, have induced acute pancreatitis by injections of a variety of bacteria, or, in some instances, of their toxines, into the pancreatic duct, and, since a large number of different organisms have been isolated from the pancreas in cases of acute pancre- atitis, it has been suggested that the condition is depen- dent upon bacterial invasion from the intestine. It is now agreed by almost all writers upon the subject that the bacteria isolated in these cases have no etiological' connection with the lesion and are only present through secondary invasion of the injured tissues. Hess has produced necrosis, with hemorrhage and fat necrosis, by injections of olive oil into the pancreatic duct, and has advanced the hypothesis that this effect Pathology 143 was brought about by the products into which the oil was split by the steapsin of the pancreatic juice. To verify this he studied the effects induced by injections of fatty acids, soda-soap solutions, and glycerine into the duct. Oleic acid, and 4 per cent, soda-soap solution, were both found to produce the same result as the oil, but glycerine failed to give rise to any acute inflammatory changes. In consequence of the results obtained in these experiments Hess suggested that regurgitation of fatty substances from the intestine, favoured by widening and injury of the duodenal orifice of the common duct by the passage of a gall-stone, was a possible cause of human pancreatitis. He has also suggested that poisoning with soap might be the cause of death and of the symptoms of intoxication that precede it, for Munk and Freidenthal have shown that the injection of o.i gram of soap per kilo of body-weight into the vessels of an animal brings about collapse and death. More recently Guleke has succeeded in producing acute pancreatic necrosis in 20 out of 27 dogs by ligaturing the pancreatic duct close to the duodenum, and injecting 5 per cent, of oil on the pancreatic side of the ligature. In the majority death took place six to twenty hours after the operation, but in three, where extensive necrosis was present, the fatal termination was delayed three to six days. In seven of the animals no effect was observed for seven to ten days. They then had loss of appetite, became emaciated, had fatty stools, and died seventeen to twenty-one days after the operation, excepting one strong young animal, which survived. Post-mortem the pancreas was found to present characters closely simulating those seen in man in chronic pancreatitis. Guleke has also induced acute necrosis, terminating in death in twenty to thirty hours, by injecting oil into the arteries supplying the pancreas, thus producing artificial infarcts. 144 The Pancreas: Its Surgery and Pathology The association of diseases of the pancreas with morbid conditions of the biliary passages has been pointed out by a number of observers, and since in most individuals the common bile-duct and pancreatic duct unite to form a common channel before entering the duodenum, and, in many persons (62 per cent., Helly), the common duct is embedded in the tissue of the pancreas for a part of its course, such a connection is probable on anatomical grounds. The possibility that a gall-stone lodging in the diverticulum of Vater might produce conditions favour- able to the passage of mi- cro-organisms into the pan- creas was suggested by Lancereaux, but that such an event might also bring about the penetration of bile into the pancreatic duct was first clearly dem- onstrated by Opie. In an autopsy on a case of acute pancreatitis under Halstead's care Opie found a small gall-stone impacted the duodenal orifice of Fig. 60. 60 and ■ Figs Diagram to show how a small gall-stone may obstruct the pa- pilla, and, if the ampulla of Vater be very large, may convert the common bile-duct and duct of Wirsung into one canal, thus pre- disposing to acute pancreatitis. Fig. 6 1 , Diagram to show a method of termination of the ducts which will not predispose to pancreatitis (Opie). m the ampulla of Vater, which, while too large to pass into the duodenum, was yet too small to fill the diverticulum and close the opening of the pancreatic duct. The bile and pancreatic ducts were thus converted into a continuous channel, and that the contents of the former had passed into the duct of Wirsung was shown by its walls being deeply stained with bile. Investigat- ing the literature of the subject Opie collected thirty- nine cases of acute pancreatitis associated with gall-stones, situated either in the gall-bladder or in the bile-ducts. Pathology 145 and in eight of these he found that there was a calculus in the diverticulum of Vater. It appeared possible, therefore, that the entrance of bile into the pancreatic duct was the cause of the pancreatitis in a considerable proportion of cases. In attempting to verify this ex- perimentally, Opie showed that the injection of 5 c.c. of bile into the pancreatic duct of dogs set up acute inflammatory changes, which in some instances were fatal within twenty-four hours. While claiming that bile diverted into the pancreatic duct by a biHary calculus had thus been shown, both clin- ically and experimentally, to be capable of producing acute pancreatitis, Opie pointed out that it could not be demon- strated that all cases of acute pancreatitis w^ere dependent upon this cause. Such an effect can take place only when the gall-stone is very small, and the anatomical conditions of the duodenal orifice and of the diverticulum of Vater are favourable. Measurement of the diverticulum of Vater by Opie proved that in about 30 per cent, of cases a small calculus could probably lodge in the opening of the diver- ticulum and yet only partially fill the cavity, and, as in one out of ten individuals the bile-duct joins the smaller pancreatic duct, while the larger duct of Santorini enters the duodenum at the site of the lesser papilla, the neces- sary anatomical conditions are present in but a small proportion of cases, and the rarity of acute pancreatitis from this cause, when compared with the relative fre- quency of cholelithiasis, is not difficult to explain. It has been shown by more than one observer that the lower end of the biliary passage in the dog is normally the habitat of pyogenic bacteria, and it has been suggested by Trevor that the experimental injection of bile, and other substances, into the pancreatic duct may produce its effect by lowering the resistance of the walls of the duct, thus allowing the entrance of septic organisms. Recent experiments by Flexner, however, tend to prove 146 The Pancreas: Its Surgery and Pathology that the inflammatory changes are directly due to the action of the bile salts upon the pancreatic cells, and it is probable that other irritating substances may act in a similar manner. The experimental evidence already quoted has demon- strated that the injection of various substances into the pancreatic duct, or tissue of the pancreas, gives rise to Fig. 62. — Acute haemorrhagic necrosis (above). Zone of leucocytes and red cells, acini, swollen and cloudy (below) (Deaver and MuUer). acute changes which, in many instances, are accompanied by an effusion of blood, thus giving rise to what is clini- cally termed " hcsmorrhagic pancreatitis.'' In some in- stances, however, where the animal has survived for a more or less lengthy period we have seen that purulent changes, or necrosis of portions of the gland substance, have been found. These may be regarded as correspond- Pathology 147 ing to suppurative and gangrenous pancreatitis respec- tively as met with in man. It would therefore appear that hemorrhagic, purulent, and gangrenous pancreatitis, in spite of the differences in their morbid appearances, are probably but phases in the same process, and that, as Opie has remarked, gangrenous pancreatitis is but a late stage of the hsemorrhagic form. The pancreas in hcemorrhagic pancreatitis is enlarged ; its interstitial tissue, as well as the tissues in its neigh- bourhood, is infiltrated with blood. Microscopically the parenchyma is necrotic and infiltrated with cellular Fig. 63. — Abscess of the pancreas (Roval College of Surgeons Museum 2832).' and fibrinous exudates. Numerous foci of disseminated fat necrosis are always found in the omentum and sub- peritoneal tissue. In suppurative pancreatitis the gland contains one or more abscess cavities of varying size. It may be enlarged, from the accompanying inflammatory changes, and the surrounding tissue may be indurated and adherent. The history of suppurative pancreatitis is always much longer than that of the hsemorrhagic form, as the results of animal experiments would suggest, and may extend to weeks or even months. Gangrenous pancreatitis rarely proves fatal until some 148 The Pancreas: Its Surgery and Pathology little time after the first onset of the symptoms. The organ after death is dry and dark, or even black, and is enlarged and friable. If the process extends to the sur- rounding tissues, and affects the lesser peritoneal sac, this may be converted into an abscess cavity containing pus and necrotic material, in which the remains of the gangrenous pancreatic tissue lie. In about half the recorded cases there has been evidence of previous hemor- rhage in the altered gland, which points to its frequently being the result of hasmorrhagic pancreatitis in man, as in animals. Disseminated fat necrosis is constantly found in gangrenous pancreatitis, but is uncommon in the suppurative form. In the past considerable importance has been attributed to the action of trypsin in the production of necrosis and other pathological conditions in the pancreas, but, as the pancreatic secretion is now known to have but slight proteolytic powers until it has been activated by the enterokinase of the succus entericus, and there exist in the blood anti-bodies to both trypsin and enterokinase, this explanation cannot now be considered as of much weight. Trypsin poisoning, however, is considered by Guleke to be the true cause of death in acute necrosis of the pancreas. He found that if the pancreas from one dog is introduced into the abdominal cavity of another, the same clinical symptoms and intra-abdominal picture, excepting for the local condition of the dog's own pan- creas, are produced as are seen in cases of acute necrosis. He also found that constitutional symptoms resembling those seen in acute necrosis were induced by intravenous, intraperitoneal, or subcutaneous injections of trypsin, but that if the animal were immunised against trypsin by gradually increasing doses, the introduction of an extir- pated pancreas into the abdominal cavity was not so rapidly fatal as in unimmunised animals. A question that has been much debated is the rela- Pathology 149 Hon of pancreatic hcemorrhage to acute pancreatitis. Two possibilities present themselves: (i) that the haemorrhage is a consequence of the inflammation; (2) that the haemorrhage is the primary factor and the inflammatory changes secondary phenomena. Fitz, Orth, Birch-Hirsch- feld, Zeigler, Korte, and most modern writers on the subject hold that the inflammation precedes the haemor- rhage, and the first named has in consequence designated the disease " hasmorrhagic pancreatitis." Dieckhoff, Seitz, and Hawkins maintain, on the other hand, that the haemorrhage precedes the inflammation, which is, in fact, caused by a bacterial infection of the haemorrhagic effu- sion. Seitz considers that in the cases recorded by Fitz and others as examples of hccmorrhagic pancreatitis, conclusive evidence in proof of their conclusions was not offered in a single instance, and he further contends that if the haemorrhage were secondary to the inflammatory changes cases would from time to time occur in which there was a rapidly fatal inflammation without haemor- rhage. Cayley has since reported a case, presenting symptoms of acute pancreatitis, which was fatal on the fourth day, and on post-mortem examination the pan- creas was said to show evidence of acute inflammatory changes. There was neither haemorrhage nor suppura- tion in the gland, but a general infiltration with blood- coloured serum in and around it. From this case Cayley argues that hsemorrhage is not an essential feature of acute pancreatitis. A similar case which proved fatal forty-eight hours after the first onset of the symptoms has also been recorded by Kennan. The pancreas in his case was markedly enlarged from inflammatory changes, the common bile-duct and the gall-bladder con- tained numerous gall-stones, and a calculus the size of a pea was found in the duodenum. A study of the reported cases of "haemorrhagic pan- creatitis," and of those we have had the opportunity of 150 The Pancreas: Its Surgery and Pathology ourselves observing, suggests that both views may be correct in different cases, for although a primary pan- creatitis may be accompanied by haemorrhage, this origin is not the only one, and there are many cases in which haemorrhage precedes and, in fact, is a cause of inflam- mation. This is due, first, to the great tendency of the gland to disruption, because of its soft structure, when haemorrhage does occur; secondly, to the communication of the gland with the intestine, which renders the access of putrefactive organisms likely; thirdly, to the great tendency of the damaged gland and effusion to decom- pose as soon as organisms gain access to them. This view of the subject has at least the merit of simplicity, and br ngs "haemorrhagic" pancreatitis into line with other well-known inflammations. Suppurative pancreatitis and abscess of the pancreas are sometimes met with as the result of extension from neighbouring organs, and more particularly from the stomach. Dieckhoff has reported a case of secondary suppuration associated with cancer of the duodenum, and Hale White speaks of an instance of abscess of the head of the gland in a patient who died of a malignant growth of the sigmoid flexure. Pycemic abscesses of the pancreas are rare, but have been met with in cases of pyaemia and puerperal fever. Lancereaux refers to a case of abscess of the pancreas due to the pneumococcus. The]_injection of foreign substances into the parenchyma and ducts of the pancreas in some instances, instead of giving rise to the acute changes already described, causes local or general induration of the gland, accompanied by an overgrowth of the interstitial tissue. Korte found that injections of oil of turpentine caused very intense interstitial changes, and Oser showed that indurative pancreatitis followed the injection of alcohol and zymine into the parenchyma. Flexner produced sclerosis by Pathology 151 injecting 3 per cent, agar into the pancreatic ducts, and in seven out of twenty-seven dogs, Guleke, as we have seen, induced chronic pancreatitis by injecting oil into the ligatured pancreatic duct. Two animals were also injected by the last-named observer with blood drawn from the femoral vein ; both showed signs of pancreatitis ; one recovered, but the other died of chronic pancreatitis in three weeks. Korte and Senn produced indurative pancreatitis by injury of the pancreas, and Sandmeyer has found that sclerotic changes take place in the portions of pancreatic tissue left after partial extirpation. Evi- dences of chronic inflammatory changes also develop early, according to Flexner, around acute lesions in dogs. The overgrowth of connective tissue caused by oil of turpentine is said by Korte to be increased by crushing, tearing, or cutting the gland. The part played by micro-organisms in the production of chronic pancreatitis has been investigated by Korte and Carnot. The former demonstrated that injections of pure cultures of bacillus coli caused more or less extensive interstitial changes, and that f cecal matter gave rise to similar results, Carnot, by an ingenious device, produced sclerosis of the pancreas from an ascending infection from the duodenum. A thread was fixed in the pancreatic duct and carried through its orifice into the duodenum, where it was allowed to hang free, and, when the animal was subsequently killed, the walls of the duct were found to be infiltrated with leucocytes and thickened, while there was a well-marked overgrowth of the interstitial tissue in the gland parenchyma. The ascending infection from the duodenum thus dem- onstrated by Carnot as a possible cause of chronic pan- creatitis is no doubt operative in man under certain circumstances, since a history of dyspepsia and intestinal derangement, with or without vomiting, is not uncommon in this disease. The changes in the pancreatic functions 152 The Pancreas: Its Surgery and Pathology thus brought about are probably responsible for the continuance of many cases of chronic dyspepsia in which the relation is not usually recognized, for, while the changes in the pancreas may be initiated by pathological condi- tions in the intestine, the diseased pancreas, not being capable of properly performing its functions, will accen- tuate and prolong any digestive disturbances that may be present. The entrance of micro-organisms into the duct of Wir- sung is normally prevented by the flow of the secretion and the valve-like folds in the walls of the diverti- culum of Vater, but the duct of Santorini is not thus protected, for, according to Desjardins, the secretion may there flow indifferently towards the intestine or from the intestine towards the gland. Organisms carried in from the duodenum by a reverse current will be conveyed through the substance of the organ to the point where the ducts of Santorini and Wirsung are connected, and there, meeting with the direct current in the main duct, will be carried into the intestine again. These circumstances would naturally favour the infection of the gland, and particularly the head, when from any cause the virulence of the intestinal organisms is increased, or the resistance of the pancreatic tissue is lowered. The area of the pan- creas enclosed between the duodenum to the right, the duct of Santorini above, and the duct of Wirsung below, has been termed the " triangle of infection of the pancreas, ' '■ and represents the most frequent site of inflammatory changes in the gland. Chronic pancreatitis is produced experimentally with the greatest degree of certainty, and is found most com- monly in man, as a consequence of obstruction of the ducts. Pawlow has produced interstitial pancreatitis in rabbits by ligature of the pancreatic duct; Langendorff has obtained similar results in pigeons ; Schulze, working with guinea-pigs, Ssobolew with rabbits, dogs, and cats, Pathology 153 Opie and other observers with cats and dogs, have also produced an overgrowth of connective tissue in the pan- creas by a similar method. The pathogenesis of the condi- tion is not, however, quite clear. Carnot has suggested that the retained secretion has a toxic effect upon the parenchyma of the gland, and, since the obstruction of the flow will favour the entrance of micro-organisms from the duodenum, chronic inflammatory changes are set up which result in disappearance of the secreting cells and an increase of the inter- stitial connective tis- sue. He has also sug- gested that the re- flex nervous stimuli which were believed to give rise to the pancreatic secretion were no longer able to excite normal func- tional activity when the ducts were ob- structed, so that the cells might atrophy as muscle fibres do after section of their motor nerve. The discovery of secretin, while it has deprived this hypothe- sis of the basis on which it was founded, has not altogether disposed of it, for it is conceivable that the constant stimulation of the cells by secretin may eventually give rise to atrophy when they are working against a pressure that they cannot overcome, in a similar manner to that in which the kidney undergoes changes in obstruction of the ureter. Clinical observation has shown that, although chronic pancreatitis may arise from obstruction of the duct due Fig. 64. — Chronic interstitial pan- creatitis in a cat following ligation of the pancreatic ducts (Opie). 154 The Pancreas: Its Surgery and Pathology to the pressure of tumours, stenosis of the duodenal orifice following ulceration, growth in the duodenal pap- illa or ampulla of Vater, and the presence of impacted pancreatic calculi, intestinal worms, or portions of hy- datid membrane in the duct, it is most commonly asso- ciated with cholelithiasis. The great practical importance of the association, and the frequency with which inflam- matory enlargement of the head of the gland accompanies gall-stone trouble, were first brought to the notice of the profession in a lec- ture delivered by one of us at the London Polyclinic in July, 1900, and it was then shown that surgical treat- ment is capable of affording complete relief in nearly all cases. It has since been pointed out that Riedel had published, in 1896, an account of three cases in which he drew attention to the relation of chronic pancreatitis to cholelithiasis, but we were unacquainted with his work at the time this lecture was given, and his observations do not appear to have attracted notice until after that date. The first case of chronic pancreatitis actually operated on was by one of us in June, 1890, the patient being alive and well in 1905. In this instance the pancreas was enlarged and hard, and malignant Fig. 65. — Cancer of the duodenal pap- illa, with dilatation of duct of Wirsung and chronic pancreatitis (St. George's Hosp. Museum, 113 K). Pathology 155 disease was, at the time, suspected, but her ultimate com- plete recovery suggested the true explanation. In April, 1892, however, a case was operated on in which the con- dition was proved by microscopical examination, and this, which was investigated a year before Riedel's first case, in 1893, is, so far as we can find, the earliest instance where chronic pancreatitis was conclusively demon- strated. A large number of cases are now on record in which various surgeons have observed induration of the head of the pan- "creas associated with gall- stones. The published cases cannot, however, be taken as truly represent- ing the frequency of the condition, for, in our ex- perience, pancreatitis is met with in about 60 per cent, of cases in which gall-stones are found in the common bile-duct at operation. The reason for the asso- ciation of the two condi- tions is not difficult to understand, when the anatomy of the parts is considered. Under ordinary cir- cumstances, when a gall-stone passes along the common bile-duct and reaches the ampulla of Vater it will not only occlude the bile-passage, but also the chief excretory duct of the pancreas, the secretion of which will be re- tained. An infection of the retained secretion, of tjie walls of the ducts, and of the parenchyma of the gland, is then likely to occur, and this will continue so long as the obstruction persists. Fig. 66. — Gall-stone in the com- mon bile-duct surrounded by the head of the pancreas (St. Thomas' Hospital Museum, 1380). 156 The Pancreas: Its Surgery and Pathology How far the pancreatic lesion in these cases is to be attributed to the irritating action of the retained secre- tion, and how far to the associated bacterial infection, it is difficult to say; but it is probable that, while the former damages the tissues of the gland and renders them susceptible to infection, the process is chiefly due to the action of micro-organisms. Even when the block- ing of the ducts is complete and- no direct communication between the micro-organisms in the duodenum and the stagnant secretion is possible, the inflamed walls of the duct present a ready path by which infection may travel from the intestine, and that this does occur in the bile passage has been proved by aseptic ligature of the com- mon duct. Absolutely complete blocking of the duct is, however, very uncommon, except in cancer of the head of the pancreas, for, as we shall show later, bile-pigment can be found chemically in the fasces in nearly all cases, even when the stools appear free from colour to the eye. Another route by which infection may reach the duct of Wirsung and biliary passages has been suggested by Desjardins. We have already pointed out how, in his opinion, organisms normally find their way from the duodenum along the duct of Santorini into the main pan- creatic duct, and thence into the ampulla of Vater and back to the duodenum again, and it is by this route, he believes, that infection of the biliary passages and pan- creas takes place when there is obstruction from gall- stones. The first effect produced by the lodging of a biliary calculus in the ampulla of Vater, or lower part of the common bile-duct, will be that the septic organisms, arriving by way of the duct of Santorini and thence passing into the duct of Wirsung, will be unable to travel into the duodenum, and, being arrested in the biliary passages and pancreatic ducts, will there set up inflammatory changes. As the pancreatic ducts are, however, more or less immune, from the constant presence in them of intestinal organisms, Pathology 157 the chnical symptoms will, in the first place, be referred to the bile-passages, the pancreatic ducts and the head of the pancreas not being markedly affected by the process until they are attacked by bacteria whose virulence has been increased by growing in the morbid secretion and diseased tissues of the bile-ducts. Inoculation and re- inoculation from the pancreas to the biliary passages, and back from the bile-ducts to the pancreas, will then occur, so that the pancreatic lesion will continually pro- gress so long as the obstruction is unrelieved. This hypothesis has been accepted by Quenu and Duval. While not denying that it is in some instances possible, we would point out that, according to the observations of Opie upon the ducts, it is unlikely that infection of the retained secretions and walls of the ducts can take place in the manner suggested in at least 31 per cent, of cases, for Opie found that in 21 per cent, of the bodies he exam- ined the duct of Santorini was impervious, and in 10 per cent., where a through channel existed, the duct of Wir- sxing did not anastomose with the duct of Santorini. In a valuable paper on "The Constituent of Bile Caus- ing Pancreatitis" Flexner has given an account of a num- ber of experiments which throw considerable light upon the relation of cholelithiasis to both chronic and acute pancreatitis. He found that solutions of purified bile salts when injected into the pancreatic duct of animals gave rise to acute fatal haemorrhagic and gangrenous inflammation of the gland, with fat necrosis, but that the mucigenous residue from dog-bile, precipitated out by alcohol, when similarly injected produced no lesion in seventeen days, except that there was slight sclerosis, probably from ligature of the duct. It was therefore evident that the effect of the injection of bile was due to the bile salts. On mixing the bile salts with the mucige- nous residue, the lesion produced by injection was found to be of a less acute and destructive character than that 158 The Pancreas: Its Surgery and Pathology following the injection of bile salts alone. A similar result was obtained when agar and gelatine were substi- tuted for the mucoid material. The action of the bile salts appeared therefore to be restrained by mixing it with colloid substances. Experiments with gelatine and bile salts showed that if the colloid was readily attacked by the pancreatic juice, so that the salts were brought rapidly into contact with the gland tissue, a moderately severe lesion resulted; whereas, with a substance such as agar, which is little, if at all, altered by the pancreatic secretion, the effect of the bile salts was exerted so slowly that all gross injury of the pancreatic substance was avoided. Increasing the colloid strength of bile, or of solutions of bile salts, by the addition of mucin, nucleo- proteid, or even by diluting with normal saline, was found to modify the intensity of the lesion in a similar way to gelatine and agar, so that the pancreas might altogether escape injury from a quantity of bile salt which would otherwise have caused a severe and rapidly fatal condi- tion. Flexner concludes from these experiments that when the composition of the bile is modified by a diminution of its salts or an increase of colloid material, its pas- sage into the pancreatic duct is likely to set up chronic pancreatitis, but when fresh unaltered bile gains entry into the duct of Wirsung, it sets up acute changes. He points out that in obstruction of the biliary passages there is a loss of diffusible salts and an increase of colloid mate- rial, and, further, that inflammation of the passages causes an accumulation of albuminous products, so that by both means the composition of the bile is altered in a direction which, according to his experiments, would tend to favour the production of chronic changes rather than acute and fulminating lesions of the gland. As we have already said, it is difficult to know how far chronic changes in the pancreas in common duct obstruction Pathology 1 59 are due to micro-organisms and how far to mechanical and purely chemical causes, but it is clear from Flexner's experiments that the influence of the bile must be taken into account. Chronic pancreatitis is often associated with very small stones, so that it is possible that bile which has been modified by inflammatory changes may have been diverted into the pancreatic duct and set up chronic inflammation in the way Flexner suggests. Even with large calculi the obstruction is rarely so complete that bile cannot find its way into the intestine, and possi- bly also into the duct of Wirsung, and there, by its toxic action, it may predispose the tissues to the action of bacteria. When the pancreatic duct is occluded, secre- tion ceases at a pressure of only a few centimetres of water, owing probably to the ease with which any fluid formed by the gland cells escapes through the alveoli into the surrounding lymph spaces, so that it is not necessary to suppose that any great amount of force is required to carry bile or other fluids into the ducts, and thence into the interstices of the parenchyma. The diffusion which naturally occurs in a stagnant fluid has also to be consid- ered in this connection. That the entry of bile is not the only cause of the interstitial changes that are commonly found in obstruction of the ducts is shown, however, by the effects of ligature, or of introducing such substances as agar or turpentine, and some part of the process must no doubt be attributed to the altered pancreatic secretion itself. Morbid influences may not only reach the pancreas by way of the ducts, but also through the blood-vessels and lymph stream. Pancreatitis in general infectious dis- eases is not common, and we have already referred to the comparative rarity of abscess of the pancreas in general pyaemia. The pancreatitis which is occasionally met with as a sequel of typhoid fever is possibly due to a specific infection travelling up the ducts, but it is not i6o The Pancreas: Its Surgery and Pathology unHkely that it may arise from the infection of the blood which is now known to be always present in that disease. In Moynihan's case, which was operated on a year and a half after the attack of fever, typhoid bacilli were iso- lated from the bile, and the patient's blood gave the Gniber-Widal reaction for typhoid fever. Three weeks later typhoid bacilli were still present in the bile, but at the end of five weeks it was sterile. This observation, while it suggests that the source of the pancreatic condi- tion lay in the infected state of the bile, due to the passage of typhoid bacilli up the bile-ducts from the intestine, cannot be accepted as conclusively settling the point, for it is well known that the urine may contain typhoid organisms months, and even years, after recovery from, an attack, although in this instance the infection undoubt- edly reaches the kidneys, and through them the urine, by way of the blood. Influenza and other zymotic diseases are also occas- ionally followed by chronic inflammation of the pancreas. In these the infection is probably carried by the blood, although in the gastro-intestinal form of influenza direct infection of the pancreas through its ducts may take place. An attack of mumps is in some instances complicated by pancreatitis, and although the unknown causal agent is probably carried from one to the other by the blood- stream, the connection is as obscure as that which exists between parotitis and orchitis. The influence of alcohol in the production of cirrhosis of the liver is still a debatable point, and similarly its relation to chronic pancreatitis has not been settled. In some cases a history of alcoholism can be obtained, but this is not common. It is probable that alcohol is not itself a direct determining cause, but that, indirectly, by the influence it exerts upon the circulation, and by a production of a catarrh of the duodenum, it may give rise to pancreatitis. Pathology i6i The chronic infections, syphilis and tubercle, affect the pancreas through its blood and lymph supply. Both give rise to changes in the interstitial tissue. The syphi- litic lesions may be divided into those met with in con- genital syphilis, and those occurring in the acquired infection. Attention was first drawn to the frequency with which the pancreas is affected in congenital syphihs by Birch- Hirschfeld in 1875, and, although the investigations of Schlesinger, and later observa- . .r/-r''^'^'3«^'^'- -v., tions by Birch- Hirschfeld him- ;• . " .. s self , have shown '•. .,,.*■,;•;-■.;.' that it is not so common as he had at first sup- posed, it is by no means un- common, and would appear to be present in about 22 or 23 per cent, of all cases of syphilis in new-born in- fants. The con- dition, like many other syphilitic lesions, is due to an overgrowth of the interstitial tissue, which, according to Schlesinger, orig- inates about the blood-vessels. The inflammatory new- growth affects both the interlobular and interacinar tissue, and occasionally spreads between the acinar cells, which atrophy and disappear without presenting any evi- dences of degeneration. In the two cases of congenital syphilitic pancreatitis examined by Opie, numerous Fig. 67. — Congenital syphilitic pancreatitis, with almost complete destruction of the secreting acini and persistence of the islands of Langerhans (X 50) (RoUeston's case). 1 62 The Pancreas: Its Surgery and Pathology islands of Langerhans were present in the thickened stroma, and some were found to be in connection with the secreting structure of the gland, although the lumen of the duct could be traced no further than the periphery of the island. Schlesinger has also pointed out that the islands of Langerhans are neither invaded by the new- growth of interstitial tissue nor implicated in the atrophy which affects the cells of the acini. Opie mentions that the parenchyma in his cases pre- sented the appearance observed about the fifth month of development, save that the islands of Langerhans were more marked features in the syphilitic glands. An explanation of the similarity between the undeveloped and syphilitic organ is afforded by supposing that the development of the individual cell is not retarded, and that the changes in the parenchyma result, not so much from its destruction, as from interference with its growth. The islands of Langerhans being the result of an early cell-differentiation, and lying more or less in the centre of the masses of secreting cells, where they are protected from the early results of the overgrowth of connective tissue, develop and remain unaffected by the interstitial changes until a late stage of the disease. Birch-Hirsch- feld believed that congenital syphilitic pancreatitis affected the organ during the last months of foetal life, but Schlesinger concludes from a study of his own cases, and those of Mraczek and Miiller, that it may be affected as early or as late as other organs. Syphilitic lesions in acquired syphilis are much rarer than in the congenital form of the disease, although Hansemann, Kasahara, and other writers have contended that it is the most common cause of chronic pancreatitis. A few cases of indurative pancreatitis, due to acquired syphilis, have been recorded, but the condition most frequently met with is the penetration of the parenchyma by irregular bands of scar-like tissue and gummata. Pathology 163 Occasionally, as in the case reported by Drozda, the pancreas may be converted into a mass of indurated tissue in which only remains of the gland substance can be found. Betham Robinson has reported a case of obstructive jaundice due to gummatous infiltration of the head of the pancreas, in which cholecystcolostomy was successfully performed. In a few cases of congenital syphilis minute and rarely large gum- mata have been noticed. Tuberculosis of the pancreas arises practically always in connection with tuberculosis in other organs, the blood-vessels furnishing the channel by which the bacilli are distributed to the gland in the majority of in- stances. Primary tuberculosis of the pancreas was probably present in a case of Senn's described by Mayo, but it is the only one of which we can find any record. Multiple small tuberculous deposits may be found irregularly scattered through the sub- stance of the gland, or single large masses, which may caseate and form cavities that open into adjacent or- gans, such as the stomach, may be met with. It is probable that the single masses originate from the lymph glands buried in the substance of the organ, for such a mass was successfully removed by operation from the head of the pancreas by Sendler, and on microscopical examination was found to be a tuberculous lymph gland. According to Carnot, diffuse interstitial pancreatitis is more commonly associated with tuberculosis of other organs than is the specific lesion itself. Cases of this Fig. 68.— Tuber- culosis of the pan- creas following tu- berculous meningitis, showing deposits of tubercle and a small abscess cavity (St. Bartholomew's Hosp. Museum, 2272 A). 1 64 The Pancreas: Its Surgery and Pathology description have been reported by Carnot, Ancelet, Vulpian, Arnozan, Morache, and Opie. There is usually a moderate degree of chronic inflammation, causing an increase of the connective tissue normally present around vessels and ducts and between the lobules. By injecting considerable quantities of a suspension of tubercle bacilli into the ducts and parenchyma of the pancreas in dogs Carnot was able to produce caseous abscesses and inflam- matory changes, but the lesions showed none of the specific characters of tuberculosis, and tubercle bacilli were not found in the tissues. Inferring from these results that the changes noticed might be due to the action of the toxines contained in the organism, he injected tuberculin, prepared from dead bacilli, into the paren- chyma of the gland, and obtained in one instance loca- lised sclerosis. That chronic pancreatitis may be caused by chemical products elaborated in a tuberculous lesion is also suggested by cases described by Carnot and Arno- zan. In the former the splenic extremity of the gland, in contact with a tuberculous kidney, was alone afi'ected, and in the latter chronic pancreatitis accompanied tuber- culous peritonitis. Alterations in the blood supply of the pancreas are another cause of chronic interstitial changes. General arterial sclerosis and endarteritis, although more commonly associated with atrophy and fatty degeneration, some- times give rise to chronic interstitial pancreatitis. It is possible that the moderate increase of fibrous tissue found microscopically in a certain number of patients over forty years of age (lo per cent, of Bosanquet's cases) may be due to this cause. Fleiner suggests that the condition is similar to that met with in contracted kidney, and in the liver, brain, and heart, as a result of endarteritis obliterans. Both he and Hoppe-Seyler think that the arterial disease causes nutritive changes in the paren- chyma, which degenerates and is replaced by fibrous Pathology i6 D tissue. An increase of the connective tissue of the gland, especially that connected with the veins and lymphatic vessels, has been described by Lepine, Abia, Lemoine, and Lannois. In these cases the lobules were separated by strong trabeculse of connective tissue which penetrated between the individual cells. In spite of the marked microscopical changes found, it is noteworthy that no abnormality of the pancreas was seen on naked-eye examination. Long-standing difficulty of venous flow, due to chronic disease of the heart, liver, lungs, etc., may cause induration and fibrosis of the pancreas, as of other organs. Thrombosis, or blocking of the portal vein by growth, has also been met with as a cause of chronic changes in the pancreas. Basing his conclusion on the post-mortem records of Guy's Hospital, Hale White is of opinion that disturbances of circulation are much com- moner than other causes of cirrhosis, congestion, or hardening of the pancreas. Opie, however, comes to the conclusion that chronic passive congestion is an unimpor- tant factor in the production of chronic pancreatitis, and in our experience circulatory disturbances are uncommon clinical causes of pancreatic troubles. Nearly all modern observers are agreed that the pan- creas is frequently affected by an overgrowth of connective tissue in cirrhosis of the liver. In cases associated with portal cirrhosis the size of the pancreas varies. It is generally enlarged as a whole, but should atrophy recog- nisable by the naked eye be present, the body and tail of the gland are the parts chiefly affected. Klippel and Lefas found that the size and consistency of the pancreas are not related in any way to the condition present in the liver. As a rule, the liver is more seriously affected, but in some cases the disease of the pancreas is in a more advanced stage. In every case, however, the fibrous tissue in the pancreas was fully formed and poor in nuclei, even when the newly formed fibrous tissue in 1 66 The Pancreas: Its Surgery and Pathology the liver was of a semi-adult type. They therefore con- clude that cirrhosis of the liver and pancreas are due to the same etiological factors, but that the pancreatic condition is independent of, and not secondary to, the lesion in the liver. The increase of fibrous tissue in. the pancreas is perilobular, intralobular, or partly periaci- nous, but is usually chiefly intralobular. There is occa- sionally interlobular oedema, and scattered areas of small- celled infiltration are met with. The gland cells show fatty and pigmentary changes, but the islands of Langer- hans are unaffected. Although the pancreas in biliary cirrhosis is not gener- ally increased in size or weight, it is often indurated, and may be united to neighbouring organs by adhesions. The fibrosis is of an embryonic type, and appears to spread from the ducts. The acinar cells show signs of fatty degeneration, and there is some proliferation of the cells lining the ducts. Exceptionally there may be enlarge- ment of the pancreas with hypertrophic cirrhosis of the liver, when there is also extreme enlargement of the spleen. Chronic interstitial pancreatitis is met with in kcEmo- chromatosis. In this condition the pancreas is enlarged, firm, and pigmented, there is generally hypertrophic cirrhosis of the liver, and, in the majority of cases, bronz- ing of the skin also occurs. The islands of Langerhans are gradually altered or destroyed, and diabetes {diabetes bronze) usually supervenes in the later stages. The etio- logical factors of this disease are not known, but it is evi- dent that they simultaneously produce change in the liver and pancreas. Occasionally chronic inflammation of the pancreas may be due to the extension of inflammatory processes from neighbouring organs. The most common origin is from a gastric ulcer adherent to the head of the gland. Ulcers of the duodenum, and ulcerating malignant growths Pathology 167 of the pylorus, have also been responsible for the condi- tion in our experience. Cases in which secondary inflam- mation has been caused by ulcerating growths in other organs, by pre- vertebral inflammatory processes, and by aneurysm of the aorta or coeliac artery, have been re- corded. From what has been already said on the microscopical appearances seen in chronic pancreatitis arising from various causes, it will have been gathered that the histo- logical changes are not always of the same type. A classification of the various forms based upon etiological data is, in the present imperfect state of our knowledge, unsatisfactory, and similarly attempts to refer the origin of the fibrous overgrow^th to the ducts, blood-vessels, and lymph channels are so speculative as to be unreliable in practice. The best working classification is that first clearly outlined by Opie. This observer distinguishes two main types of chronic interstitial inflammation, which can be distinguished microscopically, and present more or less different characters to the naked eye. In the first, or ''interlobular'' type, the increase of connective tissue, although never accurately confined to one locality, is most conspicuous between the lobules, and affects little, if at all, the intralobular and interacinous trabeculae. The normally obscure lobulation of the gland becomes more conspicuous, and wide bands of sclerotic tissue separate groups of lobtdes. The progress of the lesion is apparently inward from the periphery of the lobules, which are invaded to a greater or less degree by the newly formed fibrous tissue. Often entire lobules are seen in the process of disintegration and replacement. The islands of Langerhans are not affected until a late stage of the disease and diabetes does not occur, except when the fibrosis is very advanced. Macroscopically the gland is ha.rd and dense, and has a nodular or granular surface when the lesion is well marked. On section the Fig. 69. — Microphotographs of the pancreas in six cases of chronic pancreatitis under our care, showing the stages of the process: (a) Ca- tarrhal pancreatitis; (6) shght interlobular pancreatitis; (c) more ad- vanced interlobular fibrosis ; (d) advanced interlobular pancreatitis ; (e) very advanced fibrosis; (/) cirrhosis of the pancreas from a case of diabetes (X ca 42). 168 Pathology 169 loose areolar tissue, normally present between the secon- dary and tertiary lobules, is found to have been replaced by sclerotic bands, so that the cut surface has a compact and homogeneous appearance. In the second, or " inter acinar'' type, the new-formation of fibrous tissue takes place j)rimarily within the lobules, is of a diffuse character, and forms an irregular network of fibrous strands of varying thickness which encloses the gland acini in its meshes. The interlobular tissue may be only slightly altered. The islands of Langerhans are early affected, and diabetes is a com- mon accompani- ment of the condi- tion. The organ is usually smooth on the surface, and in section is found to be tough rather than hard. Well- marked microscop- ical changes may be present, in either form, however, without there being any noticeable macroscopical lesion. Calculi. — An extreme degree of sclerosis is often found to be associated with the presence of pancreatic calculi in the ducts, and it is usually assumed that the pancreatitis is a result of the blocking and irritation of the ducts by the calculi. Although this is no doubt true to a certain ex- tent, and the very marked fibrosis found post-mortem in such cases is largely due to the presence of the calculi, it is probable that the concretions themselves arise as a conse- quence of morbid changes in the pancreatic secretion con- Fig. 70.— Interstitial pancreatitis in the neighbourhood of an adherent gastric ulcer (X 40). lyo The Pancreas: Its Surgery and Pathology nected with inflammatory changes in the glands and ducts. By ligaturing the duct of Wirsung, Pende was able to in- duce the formation of pancreatic calculi in a considerable proportion of the rabbits on which he operated. The concretions were small, and none were discovered until a minimum interval of twenty-eight days had elapsed. They consisted of a deposit of calcium carbonate in an Fig. 71. — Chronic interstitial pancreatitis of interacinar type, show- ing the invasion of an island of Langerhans by the inflammatory pro- cess (Opie). organic matrix, and contained no appreciable amount of phosphates, thus agreeing with the composition of many pancreatic calculi obtained from man. These are often found to contain 50 per cent, or more of calcium carbonate with traces of magnesium, some organic matter, mainly of a proteid nature, and of phosphates a varying amount or none at all. According to this observer, therefore, Pathology 171 simple obstruction of the pancreatic duct is sufficient to produce pancreatic lithiasis. Desquamation of the epithe- lium of the ducts is the primary effect; the cellular ele- ments then lend themselves to the formation of a fiVjrillar network which forms a nucleus for the precipitation of calcium carbonate. This salt, which is absent from the normal secretion, appears as a result of the chronic ir- ritation due to the stasis, which also leads to a reac- tion in the pericanalicular, interacinar, and interlobular connective tissue of the gland. Thiroloix has also pro- duced lithiasis in the pancreas of a dog experimentally » Fig. 72. — Pancreas from a case of pancreatic calculi, laid open to show the calculi lying in the dilated duct of Wirsung and the attendant atrophy of the gland due to the associated inflammatory changes, from a case of diabetes (Leeds Museum, E E 202). by injecting a mixture of soot and carbolized liquid vaseline into the duct of Wirsung, after the duct of San- torini had been tied. Post-mortem the pancreas was found to be sclerosed, and in the tail a large cystic cavity, containing clear, watery fluid, and surrounded by chronic inflammatory tissue, had developed. In the cyst, and in the duct also, small, hard, irregular concretions had formed. The view that gall-stone formation is due to the influence of micro-organisms has been gaining ground since Bernheim directed attention to the connection between typhoid fever and cholelithiasis in 1880, and Galippe found bacteria in. the interior of biliary calculi 172 The Pancreas: Its Surgery and Pathology *%>■ in 1886. This has naturally led some, and particularly Nimier, to attribute to micro-organisms a causative influence in the production of pancreatic calculi. The coexistence of biliary and pancreatic calculi in a case reported by Kinnicutt was thought by this observer to point to a common cause in the shape of an infection travelling up the biliary passages and pancreatic duct. Galippe on examining a stone found numerous bacteria, and Guidicean- dra discovered an organism very similar to, if not identical with, bacillus coli com- munis in two pancreatic cal- culi. The con- cretions experi- mentally pro- duced by Pende were, however, sterile, and the secretion which had been re- tained in the duct and its tri- butaries by the occluding ligature gave no growth on culture. Cysts. — ^Experimental work on the pancreas has not thrown much light upon the etiology of the cysts occurring in that organ. Our present knowledge has been obtained chiefly on the operating table and at the bedside, and to a less extent in the post-mortem room. Senn found that ligature of the pancreatic ducts in animals caused only a moderate dilatation beyond the point of constriction, but no true cyst formation. He points out that possibly ^'^S- 73- — Section of the pancreas shown in Fig. 72, demonstrating complete replacement of the glandular parenchyma by fibrous tissue (X 50)- Pathology 173 chronic, or intermittent, obstruction might result in the production of cyst, just as ligature of a ureter, or acute obstruction, leads to atrophy of the kidney, while chronic obstruction, or obstruction of an intermittent character, tends to the development of hydronephrosis. It is prob- able, therefore, that the simple so-called retention cysts of the pancreas do not result solely from a hindrance to the outflow of the secretion, but that some other factor is also involved. Heinricus sug- gests that there is some change in the pancreatic juice, prob- ably arising from its admixture with pathological non-absorb- able products, and lessened absorptive power on the part of the vessels. The experiment of Thiroloix, already referred to, lends support to the first part of the suggestion, for in that instance a cyst of the pancreas resulted from ligature of the ducts and alteration of their contents by the injection of carbolized vaseline and soot. It will also be remembered that Thiroloix tied both the duct of Wirsung and the duct of San- torini. This is an important precaution to take, for in some instances the latter may act as a safety-valve when the main duct is obstructed. Examination of the pancreas after death in this experi- ment showed that it was very hard and deeply sclerosed from chronic inflammatory changes. Chronic pancreatitis is also present in many cases of simple cyst of the pancreas in the human subject, and there is no doubt, both on ex- perimental and clinical grounds, that it is a frequent, •VfFJVff Fig. 74. — Pancreas with small retention cysts (Leeds Path. Museum, E E 203). 1 74 The Pancreas: Its Surgery and Pathology and probably the most common, cause of the condition. Contraction of the newly formed fibrous tissue may cause constriction of the ducts in places, while in other parts they may be pulled upon and dilated; the pancreatic secretion will then tend to collect in the dilated portions and undergo chemical changes by which its physical characters are altered, and its absorption interfered with, as Heinricus suggests. Pancreatic calculi, as Thiroloix's experiment showed, may originate under conditions similar to those that give rise to pancreatic cysts, and both are intimately asso- ciated with chronic pancreatitis . It is probable , therefore , that when cysts and calculi occur together, as is sometimes the case, the one is not the cause of the other, as is gener- ally assumed, but that both originate from the same path- ological process. Blocking of the excretory duct by a calculus is, however, likely to bring about more rapid distension of any cystic cavity that may be present and so increase its size as to make it clinically recognisable. A gall-stone impacted in the ampulla of Vater has been quoted as the cause of retention cysts of the pancreas in some cases, and although the blocking of the pancreatic duct in this way may bring about the distension of a small pre-existing cyst in a similar manner to a pancreatic calculus in the duct itself, it is not likely that the obstruc- tion is the primary cause of the cyst formation. Gall- stones and chronic pancreatitis are very frequently associated, and in some instances the two conditions may be due to the same infective process, so that in this in- stance also the calculus is possibly but the secondary cause that brings into clinical prominence cysts previously formed as the result of chronic inflammatory changes in the gland. Repair following traumatism, duodenal ulcers, tumours in the bile-passages or duodenum, swollen lymphatic glands, and even intestinal parasites in the ' pancreatic Pathology /o duct are all liable to be accompanied by chronic pan- creatitis, and have each been met with in association with cysts of the pancreas. Whether they are to be regarded as the primary causes of the cysts, or merely as secondary factors accentuating a pathological state arising from the inflammatory changes in the gland, it is difficult to say, but in our opinion the latter is the more probable explanation, at least in many instances. It has been contended by some writers that cysts of the pancreas may originate from extravasations of blood, either within or without the gland, and Hagenbach has distinguished between "hsematoma," in which bleeding occurs into a pre-existing cyst, and "apoplectic cysts," resulting from haemorrhage into softened, degenerate gland-substance. The distinction is, however, not recog- nised by most modern authorities, and the presence of a large amount of blood in a pancreatic cyst is now gener- ally regarded as merely the result of a more marked haem- orrhage into the cyst cavity than usual. Retention cysts may be single or multiple, unilocular or multilocular. Two cysts of almost equal size may be present simultaneously, or one cyst may be found with a number of smaller ones attached to its walls. Obstruc- tion of the main duct near its entrance into the duodenum may cause a rosary-like dilatation to which Virchow has given the name of "ranula pancreatica." Dilatation of part of the main duct tends to give rise to a spherical or oblong swelling, while obstruction of the smaller ducts may result in the formation of the collection of minute cysts spoken of by Klebs as "acne pancreatica." As a cyst enlarges it encroaches upon and destroys the substance of the gland, or, growing away from the pancreas, it ma}^ become pedunculated. The size varies within very wide limits, from the tiny points of fluid met with in acne pancreatica, to enormous tumours holding fifteen or twenty litres of fluid. The wall of a simple 176 The Pancreas: Its Surgery and Pathology retention cyst is composed of dense fibrous tissue, poor in cells, and is generally from 3 to 4 mm. in thickness. The inner surface may be smooth, shining, and free from epithelium, or be covered with a layer of cylindrical cells resembling those lining the ducts. Indications of the formation of the larger cysts by the fusion of smaller cavities may be met with in the shape of projections of, or septa on, the inner surface, and portions of pancreatic tissue are not infrequently found embedded in their walls. The outer surface is often traversed by large distended blood-vessels. The contents of these cysts are of a fluid character, but vary considerably in their appearance and properties. The colour is generally dark reddish-brown, but may be yellow, greenish, milky, or even bright red, from recent haemorrhage. The fluid is usually viscid, generally more or less turbid, and of a specific gravity of 1.007 "to 1,028. It is generally alkaline in reaction, is rarely neutral, and in one instance, reported by Bozeman, was acid. Albumin, as might be expected from the very frequent presence of blood, is a constant constituent. Sugar has been met with in rare cases; 2.7 per cent, was found in a case of diabetes recorded by Bull. In Hoppe's case the fluid contained 0.12 per cent, of urea. Microscopical examination generally shows blood-cells, fat, and epithelial cells, often cholesterin, and rarely leucin and tyrosin (Tilger, Newton Pitt, and Jacobson). In many of the recorded cases one or more of the pancre- atic ferments have been detected, but they have not been invariably found, and have indeed been proved to be absent in cases of undoubted pancreatic cyst, confirmed by post-mortem examination. Proliferation cysts, or cystic neoplasms, of the pan- creas are very much rarer than the form just described. They may be either simple or malignant, although some cannot be relegated to one or the other category on Pathology 177 histological grounds alone. The simple proliferation cyst, or "cystadenoma," is usually multilocular and has a lining of columnar epithelium, which is sometimes seen to dip down into the wall of the cyst in the form of a gland, and often covers polypoid masses projecting into the cavity of the cyst. The malignant form, or "cystic epithelioma," occurs as a series of small cysts, or a poly- cystic mass, showing patches of carcinomatous material in the walls. The cells of the solid portions of growth are arranged in irregular groujjs, devoid of any true glandular order, and are large polyhedral and often multinuclear. In Hartmann's case the tumour is re- ported to have contained 200 grams of chocolate-coloured fluid. Metastatic deposits may be found in the liver, pancreatic glands, duodenum, and in other situations. The formation of cysts in cancer of the pancreas has been described by Roux. It has been suggested that tumours arising in the neighbourhood of the pancreas, and closely resembling multilocular proliferation cysts of that organ, may arise from remnants of the Wolffian body and be mistaken at operation for pancreatic growths. Alonprofit has pub- lished a case in which a large cystic tumour was so firmly attached to the spleen and the tail of the pancreas that it was necessary to remove both to complete the opera- tion. From the microscopical characters and situation of the growth it was concluded that it had originated from remains of the Wolffian body in the posterior layer of the mesocolon. Dunning has also reported a somewhat similar case. Invasion of the substance of the pancreas by such growths, or by similar neoplasms in the left suprarenal capsule, may readily be mistaken at operation for a growth of the pancreas, and the presence of adhe- rent remains of the pancreatic tissue may tend to confirm the error on microscopical examination. Hydatid cyst and congenital cystic disease of the pan- 1 78 The Pancreas: Its Surgery and Pathology creas are both exceedingly rare. They differ in no essen- tial particular from similar lesions met with in the liver, kidneys, and elsewhere. Pseudo-cysts, as Korte has proposed to call them, constitute a large proportion of the cases reported as pancreatic cysts. They are fluid tumours found in more or less close proximity to the pancreas, but not originating Liver Pancreas Lesser peritoneal sac Stomach Colon Small intestine Fig. 75. — Diagram to show the method of origin of a pseudo-cyst of the pancreas. in the substance of the gland. The most frequent form is that described by Jordon Lloyd, in which an efliusion takes place into the lesser peritoneal cavity, mainly as the result of injury of the pancreas. The escape of blood, followed by pancreatic juice, into the lesser cavity of the perito- neum sets up a mild form of peritonitis which may close the foramen of Winslow and produce a tumour, which, Pathology 179 during life, it is impossible to distinguish from a true pancreatic cyst. The fact that a cavity within the ab- domen contains a fluid possessing digestive powers is no proof that it is a retention cyst, but merely that it is probably connected with the pancreas. Other forms of pseudo-cyst will be considered in connection with the diagnosis of swellings of the pancreas, but it may be mentioned here that the greater proportion are met with in males, who are more exposed to injury, whereas the majority of true cysts appear to occur in women. The relation of trauma to cysts of the pancreas has been debated by Korte, Tilger, Moynihan, and others. Korte has described two classes of traumatic cysts of the pancreas, one in which, after long-continued discom- fort in the epigastrium, a tumour gradually develops, and the other where, within a short time of the injury, a tumour of considerable size has formed. In the former it is possible that the injury leads to a chronic interstitial inflammation, such as is known to follow experimental injury of the pancreas in dogs, and that this gradually gives rise to cyst-formation in the manner already des- cribed. In the latter class of cases, where the tumour forms rapidly, and increasing up to a certain point, then remains stationary, the effusion is probably poured out into a pre-existing cavity, such as the omental bursa, and is a form of pseudo-cyst. Tumours. — The solid tumours met with in the pancreas are carcinoma, sarcoma, adenoma, and lymphoma. It was formerly taught that carcinoma is the most frequent of all diseases of the pancreas, and although it is the most common new growth, it is certainly not the commonest lesion. The mistake has arisen from too great reliance being placed upon naked-eye observation in the post- mortem room, and the failure of surgeons in the past to recognize that many swellings of the head of the pancreas associated with jaundice are merely inflammatory. The i8o The Pancreas: Its Surgery and Pathology importance of the latter point was strongly insisted upon by one of us in 1900, and has since been confirmed by a number of independent observers. Fig. 76. — Spheroidal- celled carcinoma of the pan- creas (X 50). ^ Fig. 77. — Columnar -celled carcinoma of the pancreas (X 50). Fig. 78. — Columnar-celled carcinoma of the pancreas (X 150). *^*^^^ Fig. 79. — Columnar - celled carcinoma of the pancreas under- going colloid change, from a case of diabetes (X 40). Primary carcinoma of the pancreas may begin in the glandular epithelium, or in the cells lining the excretory Pathology i»i ducts. In the former case it is of the spheroidal, and in the latter is generally said to be of the columnar, ty^je. Letulle, however, maintains that primary carcinoma of the duct of Wirsung is spheroidal and not columnar celled. Spheroidal-celled carcinoma is, at any rate, -much the more common, and is usually of the scirrhous variety. Encephaloid tumours are sometimes met with, and rarely a colloid carcinoma, resulting from degenerative changes in a columnar-celled growth, has been encountered. Hillier and Goodall have distin- guished a variety of carcinoma characterised by great irregu- larity in the size and shape of the cells, which they believe arises in the island of Langerhans. The most frequent site of the lesion is in the head of the gland, some 62 per cent, of the recorded cases being in that position. In about 5.5 per cent, the tail of the organ was most affected, in 3.5 per cent, the body, and in 29 per cent, there was a diffuse growth involving, more or less, the whole of the pancreas. The duct of Wirsung is compressed by the growth in nearly all instances. Courvoisier found it obliterated in 55 out of 66 cases, and, according to Boldt, it is dilated beyond the point of stricture in one-third of all cases of malignant disease of the head of the gland. Growth in this situation almost always causes gradually increasing, painless jaundice with en- largement of the gall-bladder, points of some importance in the diagnosis of the condition from obstruction of the duct due to gall-stones, in which, although there may be equally deep jaundice, there is commonly a history of Fig. 80. — Cancer of the head of the pancreas (St. George's Hospital Museum, 201 B). 1 82 The Pancreas: Its Surgery and Pathology pain and the gall-bladder is small and shrunken. The Fig. 8 1 . — Cancer of the head of the pancreas showing dilatation of the duct of Wirsung (St. Thomas' Hosp. Museum, 1415). Fig. 82. — Carcinoma of the body of the pancreas (Royal Coll. of Surg. Museum, 3835). explanation of the different behaviour of the gall- bladder in the two conditions appears to be that, as Pathology 183 the result of gall-stone irritation, it frequently becomes diminished in size and adherent, so that when the common duct is subsequently blocked, it is unable to expand, whereas blocking of the duct by a tumour at once causes distension of the gall-bladdet, for it has not been altered by previous inflammatory changes. It has also to be remembered that, although the obstruction due to a large gall-stone may pos- sibly be absolute at first, it quickly ceases to be so, and small quantities of bile find their way into the intestine, but that in malignant disease of the head of the pancreas the obstruction gradually in- creases, and eventually be- comes absolute, so that not a trace of stercobilin can be found in the faeces. The backward pressure in the ducts in these cases, while it prevents the excretion of bile, does not interfere with its formation, and it is conse- quently absorbed by the lymphatics and gives rise to jaundice. The pressure in the ducts also prevents the bile reaching the gall-bladder and it is consequently found to be only filled with mucus. The relations and size of some of these growths explain, the compression and perforation of the duodenum, stom- ach, colon, ureter, portal vein, aorta, vena cava, splenic artery and vein, and superior mesenteric vein that some- times take place. Fig. 83. — Cancer of the head of the pancreas producing dilatation of the common bile- duct and gall-bladder (Univ. Coll. Hosp. Museum, 3198). 184 The Pancreas: Its Surgery and Pathology Secondary deposits occur most frequently in the liver, but may be found anywhere, or indeed everywhere, for Oser has reported general carcinomatosis from a pancrea- tic growth, although the primary origin of the growth is perhaps rather a matter of surmise in such cases. As in carcinoma elsewhere, carcinoma of the pancreas is most frequent after the fortieth year. Bohn has, however, reported the occurrence of the disease in a child of seven months, Kuhn in one of two years, and Dutil in a patient fourteen years of age. In many cases of primary carcinoma there is a coexist- ing fibrosis of the gland, and, although it is prob- able that in many in- stances the overgrowth of fibrous tissue results from the inflammatory changes set up by the spread of the tumour, it is possible that, in some, the fibrosis may have been the primary condition, and that the carcinoma may have originated in groups of cells isolated by the fibrous tissue, in much the same way as primary cancer of the liver appears to arise from groups of cells similarly isolated in cirrhosis of that organ. The frequent asso- ciation of chronic pancreatitis with cholelithiasis has already been insisted upon, and it is not unlikely that gall- stone trouble may thus be a cause of cancer of the pan- creas. The very high proportion of cases in which the primary growth is situated in the head of the gland tends to favour the view that there may be some such association. Fig. 84. — Colloid carcinoma of the pancreas (St. George's Hosp. Museum, 201 D). Pathology 185 Primary sarcoma is very rare. Segre met with only two cases in 11,492 post-mortems, anr] Hale White men- tions only one, of undoubted primary sarcoma, in 6708 autopsies at Guy's Hospital. In most instances, where a histological examination has been made, the growth has been described as a small round-celled sarcoma or a lymphosarcoma, but it is doubtful whether some of these can be regarded as truly primary growths of the pancreas, for although the pancreas was deeply involved, the lymphatic glands, duodenum, or other structures have also been affected. A very large spindle- celled sarcoma of the pancreas is preserved in University College Hospital Museum (No. 3200), and mixed-celled sarcomas have been described by Healey and by Kakels. The pancreatic tu- mour in the former case con- sisted chiefly of round cells, but in some parts groups of mixed cells were seen, while the sec- ondary growths, which were present in the liver, were chiefly of the large spindle-celled vari- ety. Kakels' case was a very vascular mixed-celled sarcoma in the tail of the gland, and, according to him, is only the third authentic case of a primary sarcoma in that situation. Kronlein and Lubarsch have each reported a case of angiosarcoma. A sarcoma of the pancreas, in which typical epithelial pro- liferations were found in the growing parts, has been described by Michelsohn, and a similar case of "sarco- carcinoma" has been reported by Baudach. Briggs re- moved an old hydatid cyst from the pancreas of a woman the walls of which, on microscopical examination, were Fig. 85. — Sarcoma of the pancreas (St. George's Hospital Museum, 201 E). i86 The Pancreas: Its Surgery and Pathology found to have undergone sarcomatous degeneration. This case is of some pathological interest in view of the recent observations of Borrel on the supposed relation of can- cerous tumours to helminthiasis. The pancreas is said by Oser to be the seat of secondary deposits in more than lo per cent, of all cases of primary carcinoma of the stomach, and, as metastasis also takes place from organs elsewhere, such as the rectum, sigmoid flexure, oesophagus, ovary, and breast, it is not uncommon. The majority of authors state that secondary carcinoma is more frequently met with than primary growth, but Hale White found only twenty-four cases with secondary deposits, as against thirty-one with a primary growth, in the Guy's Hospital post-mortem records from 1884 to 1897. Hale White, from an analysis of these records, confirms the statement of Lancereaux that the stomach is the organ from which the growth most frequently extends directly to the pancreas. Olivier and Dieckhoff are disposed to doubt some cases described as primary pancreatic, and think that their microscopical characters suggest that they may have really originated in the glands of the duodenum. Orth, however, points out that the transition of the atypical growth of intestinal gland acini into cancerous alveoli is not always easy to establish in primary tumours, and that columnar-celled carcinoma may originate in excretory ducts of the pancreas as well as in the duodenal glands. Secondary deposits of carci- noma may be found in any part of the gland, but whenever present in the pancreas are also to be found at the same time in many other organs of the body. Secondary sarcoma of the pancreas is not uncommon. It occurs most frequently as a lymphosarcoma arising from the abdominal lymph glands, mediastinum, or duodenum. A number of cases of melanotic sarcoma have also been described. In these the primary growth is most commonly situated in the eye. Pathology 187 Adenoma of the pancreas may originate from the duct- epitheHum, the gland acini, the islands of Langerhans, or from suprarenal rests. Examples of this condition have been described by Thierfelder, Biondi, Cesaris-Demel, Neve, and Nicholls, but in at least one of them the diag- nosis is open to question. Thierfelder's case was a man who died of general tuberculosis ; a definitely encapsuled but easily shelled-out tumour was found in the head of the pancreas. Biondi excised a " fibro-adenoma" from the head of the gland. Cesaris-Demel found a growth, the structure of which was similar to that of the pan- creas, but its interstitial tissue, as well as that of the gland substance, was thickened. He suggests that the cirrhosis of the pancreas, developed upon a syphilitic basis, incited the formation of the tumour. Neve de- scribes a case in which there was a glandular tumour in the region of the pancreas adherent to the duodenum and compressing the common duct. In Nicholls' case a small, encapsuled, round, somewhat flattened nodule was present on the anterior surface of the pancreas which, on microscop- ical examination, was found to con- sist of a stroma of connective tissue, arranged in the form of imperfect and irregular alveoli, that contained cells of a glandular type forming masses and wavy bands. As compared with the acinous cells, those of the tumour were smaller, their nuclei were rela- tively larger, and their cytoplasm was looser in texture and stained more faintly and irregularly. Nicholls concluded, from its staining reactions and structure, that the start- ing-point of the tumour was in an island of Langerhans. Fig. 86. — Deposits of melanotic sarcoma in the pancreas with haemorrhage into the gland and fat necrosis (St. Thomas' Hospi- tal Museum, 14 16). 1 88 The Pancreas: Its Surgery and Pathology Lymphadenoma of the pancreas is excessively rare; but two cases are referred to by Lancereaux, and Hale White speaks of the condition as having been met with at Guy's Hospital once in fourteen years, in a patient who died from Hodgkin's disease. Literature Baillie: "Morbid Anat.," 1833, P- 221. Balser: Virchow's Archiv, 1882, xc, 520. Baudael: Dissertation, Freiburg, 1885. Biondi: Ref. Med., 1896. Birch-Hirschfeld : Gerhardt's " Handbuch d. Kinderkrank., " 1880. Arch. d. Heilkunde, 1875, xvi, 174. Bohn: Jahrbuch f. Kinderheilkunde, 1885, xxiii. Boldt: Dissertation, Berlin, 1882. Borrell: Brit. Med. Journ., August 18, 1906, p. 392. Bosanquet: Lancet, 1905, i, 977. Bozeman: New York Med. Rec, 1882, p. 46. Bramwell, Byrom: Scottish Med. and Surg. Journ., April, 1904. Briggs: St. Louis Med. and Chir. Journ., 1890, p. 154. Bull: New York Med. Journ., 1887, p. 376. Carnot: Thesis, Paris, 1898. Cayley: Brit. Med. Journ., 1896, ii, i. Cesaris-Deruch: Arch. p. le Scienze Med., 1895, ^ix. Chiani: Zeit. f. Heilkunde, 1896, xvii, 69. Courvoisier: Beitrage z. Chir. der Gallenwege, 1890. Desjardins: Thesis, Paris, 1905. Dieckhoff: Festschrift f. Thierfelden, Leipzig, 1895. Drozda: Wiener med. Presse, 1880, xxi, 993. Dunning: Amer. Journ. of Obst., Jan., 1905, p. 161. Dutil: Gas. med. de Paris, 1888, No. 38. Filger: Virchow's Arch., cxxxvii, 348. Fitz: New York Medical Record, 1889, xxxv, 197, 225, 253. Fleiner: Berliner klin. Woch., 1894, xxxi, 34. Flexner: Journ. of Exp. Med., viii, 167, Jan., 1906. University Med. Magazine, xii, 780. Flexner and Pearce: Univ. of Penn. Med. Bull., xiv, 193. Friedreich: "Diseases of the Pancreas, " von Ziemssen's Pract. of Med., 1878, viii, SSI. Guidiceandra: II Policlinico, 1896, pp. 33 and 126. Guleke: Arch. f. klin. Chir., Ixxxvii, Hft. 4. Hagenbach: Deutsche Zeit. f. Chir., 1887, xxvii, no. Hansemann: Verhandl. der Deutsch. path. Gesellsch., 1902, iv, 187. Hartrnann: Cong. Franc, de Chir., 1891. Hawkins: Lancet, 1893, ii, 3s8. Healey: Journ. of the Royal Army Med. Corps, iv, 362. Heinricus: Congres. de Chirurg. du Nord, ii, i, 1896. Hennigs: Dissertation, 1880. Hess: Miinch. med. Wochenschr., 1903, xliv, 1905. Hillier and Goodall: Archives of the Middlesex Hospital, 1904, xi, i. Hlava: Bull, internat. de I'Acad. des Sciences de Boheme, 1898. Hoppe: Virchow's Arch., i8s7, ^i- 96- Hoppe-Seyler: Deutsch. Arch. f. klin. Med., 1893, iii, 171. Pathology 189 Kakels: Amer. Journ. of Med. Sci., cxxiii, 471. Kasahara: Virchow's Archiv, 1896, cxliii, iii. Kennan: Brit. Med. Journ., 1896, ii, 1442. Kinnicutt: Amer. Journ. of Med. Sci., Dec, 1902, p. 948. Klebs: " Handbuch der path. Anat.," 1870, ii, 533. Klippel and Lefas: Revue dc Medecine, 1903, No. i, p. 23. Korte: Berliner Klinik, 1896, No. 102. Korte: Deutsche Chirurgie, Stuttgart, 1898. Kiihn: Berliner klin. Wochenschr., 1887, S. 628. Lancereaux: "Traite des malades da foie et du pancreas," 1899. Lemoine and Lannois: Arch, de Med. Exper., 1891, iii, 7,^. LetuUe: La Presse Med., 1906, p. 256. Lloyd, Jordon: Brit. Med. Journ., November, 1892. Moynihan: Lancet, June 6, 1903. Maj^o: "Outlines of Human Pathology." Mering, von, and Minkowski: Arch. f. exp. Path. u. Phar., 1890, xxvi, 371. Semaine Med., 22 Mai, 1889. Michelsohn: Dissertation, Wurzburg, 1894. Monprofit: Gaz. m^d. de Paris, March 12, 1904. Neve: Indian Med. Record, 1892, p. 208. Newton Pitt and Jacobson: Lancet, 1891, i, 13 15. Nicholls: Journ. of Med. Research, viii, 385. Nimier: Reva de Med., 1894, x, 9. Opie: " Diseases of the Pancreas," 1903. Orth: " Lehrbuch der path. Anat.," 1887. Oser: "Diseases of Pancreas," Nothnagel's "Encyclop. of Med." (Engl, trans.), 1903. Pawlow: Ref. Jahresbericht u. d. Fortsch. d. Anat. u. Phys., 1890, xix. Pende: II Policlinico, 1905, p. 122. Percival: Transact. Associat. King's College of Ireland, 18 18, p. 128. Portel: Anat. Med., 1804, p. 353. Quenu and Duval: Rev. de Chirurgie, Oct. 10, 1905. Rentoul: Brit. Med. Journ., Dec. 24, 1904, p. 1894. Riedel: Berliner klin. Woch., i896,xxxiii, 32. Robinson, Betham: Brit. Med. Journ., 1900, p. 1004. Robson, Mayo: Lancet, 1900, ii, 235. Rokitansky: Lehrbuch der path. Anat., 1863, iii, 313. Roux: Thesis, Paris, 189 1, part 3. Sandmeyer: Zeitschr. f. Biol., 1895, xxxi, 12. Saunby: Lancet, 1890, ii, 383. Schleisinger : Virchow's Arch., 1898, cliv, 501. Schulze: Arch. f. mik. Anat., 1900, Ivi, 491. Segre: Ann. univers. della med. e chir., cclxxxiii. Seitz: Zeit. f. klin. Med., 1892, xx, 203, 311. Sendler: Deut. Zeit. f. Chir., 1896. Senn: Amer. Journ. of Med. Sci., 1885, p. 37. Trans, of the Amer. Med. Assoc, 1886. Ssobolew: Centralbl. f. allg. Path. u. path. Anat., 1900, xi, 202. Thiroloix: Arch. f. Physiol., 1892, S. 716. Trevor: Practitioner, April, 1904, p. 574. Tulpius: Observat. Med., 1872, p. 328. Virchow: Die krankhaften Geschwulste, 1863, i, 276. White, Hale: Guy's Hospital Reports, liv, 17. Williamson: " Diabetes Mellitus, " 1808. CHAPTER IX FAT NECROSIS The term "fat-tissue necrosis" was introduced by Langerhans to describe the small, opaque, yellowish- white areas described by Balser as occurring in the inter- Fig. 87. — ^Areas of fat necrosis in the mesenteric and omental fat and in the abdominal wall in a case of acute hsemorrhagfic pancreatitis (Fison). acinous tissue of the pancreas, and more rarely in the surrounding fat, of many bodies taken indiscriminately in the post-mortem room. Balser had observed the lesion in five out of twenty-five bodies he examined, and 190 Fat Necrosis 191 in two found that the process not only involved the fat about the pancreas, but was also present in scattered foci at a considerable distance from the gland. He believed that occasionally the areas might become confluent and cause death, either by their extent and the simultaneous sequestration of large portions of the abdominal fat, or from haemorrhage, indications of which he found in the adjacent tissue, especially when the changes were exten- sive. Balser made microscopical preparations from the affected areas and adjacent tissues, and came to the con- clusion that the lesion was due to an increase of the fat cells. Chiari confirmed the observa- tions of Balser as to the oc- currence of the condition, and stated that he had found it in five cases of severe disease of the pancreas, but did not agree as to its nature, for in his opinion the lesion was due to fatty degeneration and simple necrosis. A correct explanation of the essential nature of the process was first afforded by the chemical and histological studies of Langerhans, who showed that the change of the fat cells into granular balls, and the appearance in the older foci of peculiar flakes of the size and form of ordi- nary fat cells, observed by Chiari, are due to splitting of the neutral fat of the cells into fatty acid and glycerine. The fatty acids are deposited as needle-like crystals within the cells, which have lost their nuclei and are necrotic, while the soluble glycerine is absorbed. Subse- quently union of the fatty acids with calcium gives rise Fig. 88. — Portion of the omental fat from the same case of acute pancreatitis, showing areas of fat necrosis (Fison). 192 The Pancreas: Its Surgery and Pathology to irregular and often globular masses of lime salts, which more or less preserve the outlines of the cells. Langerhans found that an entire lobule, or several neighbouring lobules, may form a dead mass which is separated from the living tissue by a proliferation of the fixed tissue cells, and that the dissecting inflammation is most conspicuous .' >.v-' -' " ' >« ,'/' , 'I.. ■ ..|....|....|..'..|.v.|....|....[....|....|^ I ^ corked, and inverted Fig. 96.— Schmidt-Stokes milk-tube. forty times, taking care that the whole of the solid material runs through at each turn. Each tube is then rotated between the hands, and allowed to stand for half an hour or more, in order that the solid residue may be collected into the lower bulb. Considerable care is neces- sary in carrying out this part of the process in some in- stances, or a perfectly clear supernatant layer of ether, free from solid particles, is not secured. With a pipette, exactly 20 c.c of the clear ethereal extract are drawn off from each tube and delivered into two C02-fiasks of known weight, the amount of ether left in the tubes being noted. The ether in the flasks is then evaporated, the residue dried on the -water-bath, and the flasks again weighed. From the amount of extract yielded by 20 c.c. of ether, and the quantity of ether left in the tubes, the total amount yielded by the weight of dried f^ces used may be calcu- lated, and from this the percentage in the stool determined. Chemical Pathology 213 The result from the A-tube gives the total fat in the faeces, including the neutral fats, free fatty acids, and combined fatty acids, or soaps, since the latter will have been decomposed by being boiled with the hydrochloric acid and thuc rendered soluble; that from the B-tube represents the neutral fats and fatty acids only, as the soaps will remain undissolved by the ether: the difference between the two will therefore give the proportion of saponified fat present. Other substances in the faeces soluble in ether, such as cholesterin, lecithin, cholic acid, and pig- ments, are included in the estimates, but as the quantity is small it does not appreciably affect the results. For convenience of reference we shall speak of the yield from the A-tube as "total fat," that from the B-tube as "neu- tral fat," and the difference between the two as "fatty acid." The solid residue from the B-tube can be used for the detection of stercobilin. For this purpose it is filtered off, extracted with acid alcohol, the extract neutralised with ammonia, and mixed with an equal quantity of 10 per cent, zinc acetate in alcohol. The precipitate that forms is removed by filtration, and the clear filtrate ex- amined with a lens, against a black background, for the green fluorescence that indicates the presence of sterco- bilin. The intensity of the colour varies with the amount of pigment, so that by always using approximately the same proportion of faeces and of the reagents any marked variation from the normal can be detected. We have examined over three hundred specimens of faeces by these methods, but taking a consecutive series of one hundred recent cases, in which it has been possible to determine the condition of the pancreas and biliary passages at operation or post-mortem, and comparing the results with those obtained in sixteen normal speci- mens, they may be classified as follows: tfi < O <^ H o < Q W H O Ph o O Q W Q o t/5 o w w H o CO >^ o o 00 0\ o to o M O •« o PI o " o m-" N +-• in<-. ^■H VO i! VO VD ^ H 00 00 VO -.12 v!; -D 00 t^ f^ ^^ ^^ ^^ ^^ ,_^ ^^ ,^ ^^ JH S? 6? 6? 6? fe? 6? 6? &s ?l „ ■* ri- -VO ^00 •^' 6? 6? 6? fe? 6§ » fe? £S feS \n o '-' o "^2 CO O VO o " o t- o o o fO+J w *j vo o t^ lO VO CO VO o> t VO vil ^ •* CO ^ ,_, ^-, ,^ ^^ ,_^ ^^ ^^ ^^ ■< 6? 6? 6? feS 6? feS ^ 6? « - „ »<3 ■* t^ ^5 6?" 6? fe§ " feS 6? S? 6? ^ fe?"' 00 o 00 o VO O " 2 Ov O 00 o 2 M O ;2;fe C^+J co*^ M *J M +J t^ o »o M VO CO Ov ,H - ■D s W •* -5 VO W ^~. ^_, ^^ ,^ ^_, ^.^ ^^ ^^ iJ 6? fe? 6? 6? 6? feS 6? s? ■f'- _<2 r^ ~ f^ „ o _ -t _vo o 00 fe? " 6? " fe? " 6?" fe? " ^ " feS-f 6? ^ o 20 O o ■*o lo O OO t^ o *-< O Tf *j VO *J t^*i N o n VO Ci VO CO f^ 00 f^ 00 VO ^ VO Ov -^ ^^ ^-^ ^^ ^^ ,_, ,_^ ,_^ ^^ ii «■ 6? £? S? 6? s? 6? 6? 6? O i;< 6? 00 6?°o fe?<» „ o 6? 00 _ "o S^° iC-o SO r- o VO O 1- o Ov o lo O f^d oS 00 ^ 00 -I-' 00 -M 00 ■" 00 ■" 00 ■" 00 *^ 00 i> M CO n Cv) 00 VO Tf VO o> o. a Ov o 00 Ov 00 v^ 1 ■ ■ -o o « ro « « t»l M O N lO « lO " O i^ r^ N Ct CO CO Cii d o 1^' 1 e J &■ a k 1 te" 2 1 g u &' e J 6= rt u ^ s u ^ e hj .■s^ & .|j2 & .■s^ S .5^ & •--2 g .5^ & o U $>n N VO O ifl WHO t^ o lO CO CN o M M r) VO 8z C! S 1 " : d £ m u £ ; S g- gs l§ i 3j " : a £ • t^ S g— . S ca^ £ g — £ g — £ cS_ £h'^ fchZ (ilH'lZ fchg £hS (inS (ShS £Hg Tj- CO M 'tM CO M H m a CO O COIT) M « Ol r< H M VO ^ o . * . •13 S . c3 g "ri c ■ g.s -J 1> cd a 13 c •0 S:^ •0 S^ -t3 ■g'rt ^1^ •§i3 o 00 v> VO t^ 00 N t VO !2; '^ N ^ ^ XI 13 y h o O S J o o. : n •a si 1 sa m o i.L ^ |.L ^ •-3 J ft tj " ^ o,-" u a «-a u c! i '"> '^ •ill 1 = 11 .1 c Si fl "n o 1— > '^ cci t— > s u O ^ :^ ^ l-i l-H l-H > > 214 Chemical Pathology 215 It will be seen that the percentage of "total fat" in the fcEces has, as a rule, been in excess of the normal in those cases in which there was reason to believe that there was a lesion of the pancreas. In one case of malignant disease of the head of the gland as much as 93 per cent, was found to be present, and in no instance has it fallen below 40 per cent., the average amount being 77 per cent. Chronic pancreatitis, associated with obstruction of the common bile-duct, appears in severe cases to interfere almost as much with the fat digestion as malignant disease, for 82 per cent, was found in one instance, where there was jaundice, and 76 per cent, in another, where there was no discolouration of the skin and bile -pigment was absent from the urine. That the high proportion of fat met with in some of these cases is not entirely due to the biliary obstruction is shown by the fact that as great an excess has been met with in others in which no obstruc- tion to the free flow of bile into the intestine was present. Simple biliary obstruction, not associated with pancreatic disease, may, however, cause a very considerable increase of fat in the stools, so that in those cases where there is both obstruction of the common bile-duct and disease of the pancreas both probably influence the result. Al- though severe or wide-spread inflammation of the pan- creas would appear to always give rise to a certain amount of steatorrhoea, we have repeatedly met with cases of pancreatitis in which the proportion of fat in the fseces was normal, or even subnormal, but these have been of a mild type and generally in an eaHy stage where the head of the gland only was involved. On comparing the proportions of "neutral fat" and "fatty acid," as indicating the degree to which the proc- ess of saponification has been carried in the intestine, we see that in simple pancreatitis, not associated with jaundice, the former was usually much in excess of the latter, whereas in cases of biliary obstruction, not accom- 2i6 The Pancreas: Its Surgery and Pathology panied by pancreatic changes, the reverse was generally found. The effect produced by the coexistence of pan- creatic disease and biliary obstruction would appear to depend upon the relative extent and standing of the two conditions, for whereas in malignant disease of the head of the pancreas the "neutral fat" has always been in excess of the "fatty acid," the latter in some cases of recent pancreatitis with gall-stone obstruction has been found to preponderate. It must be remembered, how- ever, that even the total absence of pancreatic juice and bile does not necessarily put an end to the fat-splitting process in the intestine, for, under the influence of organ- isms of the colon group, the conversion of fats into gly- cerine and fatty acids may go on energetically in the lower part of the small intestine, but since the absorption of these products will be interfered with by the absence of bile and the situation in which the process takes place, they will be excreted to a large extent in the faeces. The pres- ence of a higher proportion of saponified fat in the stools in some cases of malignant disease and serious pancrea- titis than might at first sight be expected is probably to be explained, at least in part, in this way. The different results obtained on examining the faeces in apparently similar cases is shown by a series of analy- ses we have recently had the opportunity of making in two instances of pancreatic disease associated with biliary fistulas. The second also illustrates the beneficial effects that may be produced by suitable treatment even in advanced and serious cases of pancreatitis. In the first case the biliary fistula formed after an operation for gall-stones undertaken by another surgeon, for whom an examination of the urine and faeces had been made for diagnostic purposes by one of us. The fasces were then found to be neutral in reaction, of a light brown colour, and to contain a fair amount of stercobilin. Chemical examination showed: Chemical Pathology 217 Organic matter 84.7% Total fat 4 1.9% f Neutral fat 26.4% \ Fatty acid i5-5% Organic matter not fat 42.8% Inorganic ash iS-3% On a second examination being made six months later, no stercobilin could be found in the fasces, they were alkaline in reaction, and of a greasy white appearance. Chemical analysis showed : Organic matter 85.8% Total fat 85.5% / Neutral fat 35-6% \ Fatty acid 49-9% Organic matter not fat o-3% Inorganic ash 14.2% An increase of 34.4 per cent, in the "fatty acid" as com- pared with 9.2 per cent, in the "neutral" fat. In the second case the patient had a biliary fistula when he came under our observation. The faeces were white and shining, acid in reaction, and contained only a faint trace of stercobilin. Microscopically crowds of fat globules, fat crystals, and undigested muscle fibres were found. Chemical examination gave the following re- sults : Organic matter 93-o% Total fat 72.6% / Neutral fat 69.7% \ Fatty acid 2.9% Organic matter not fat 20.4% Inorganic ash 7-o% A fortnight after this examination had been made he was operated on by one of us and a cholecytenterostomy performed. No gall-stones were found at the time of operation, but the pancreas was dense, hard, and rugged, and closely gripped the common bile-duct, which passed through it. Examination of the fasces a month later, when the patient had returned to a normal mixed diet, showed that they were of a light yellow, almost white, colour, acid in reaction, and contained many fat globules, 2i8 The Pancreas: Its Surgery and Pathology fat crystals, and some undigested muscle fibre. Chemi- cally the following results were obtained : Organic matter 93- 1% Total fat 68.2% / Neutral fat 65.7% \ Fatty acid 2.5 % Organic matter not fat 24.9% ■ Inorganic ash 6.9% The patient was then placed upon "pancreon," and it was found, when the faeces were examined five months subsequently, that they were still of a light colour, al- though they contained a normal amount of stercobilin, the reaction was acid, and microscopically a few fat glo- bules, fat crystals, and some muscle fibres were present. The chemical analysis gave : Organic matter 93.2% Total fat 40.2% /Neutral fat 26.1% \ Fatty acid i4-i% Organic matter not fat S3-o% Inorganic ash 6.8% Turning the bile into the intestine thus produced prac- tically no change in the fat content of the fseces, but when the deficiency of pancreatic juice was partly supplied by the administration of "pancreon" the neutral fat was diminished over 40 per cent, and the fatty acid increased II per cent., while the amount of unabsorbed fat in the stools was also very considerably diminished. In most of our investigations on the fseces in pancreatic disease the patients have been upon an ordinary mixed diet, and, for purposes of clinical diagnosis, we have found that this is quite sufficient, and that it is not necessary to delay the examination of the stools until a uniform fixed standard of diet has been established. It is neces- sary, however, that the character and amount of the food should be borne in mind when the results of the chemical analysis are considered, for some fats are more ' readily absorbed than others, and large quantities might Chemical Pathology 219 pass through the intestine unchanged even in normal persons. We have already mentioned that exxjerimental investigations on depancreatised animals have proved that a natural emulsion, such as milk, is more readily dealt with than fats in the solid form, and there is evidence which shows that the chemical constitution of the latter is also not without influence upon their susceptibility to digestive processes. As a general rule, it may be stated that the lower the melting-point of a fat employed as food, the more completely will it be absorbed; thus, olein is more readily utilised by the organism than pal- mitin or stearin, and food materials containing the for- mer are not so likely to appear unchanged in the faeces as those containing the latter. In a case of cancer of the head of the pancreas, with complete biliary obstruction, we found that when the patient was upon a mixed diet the dried fasces contained 58.7 per cent, of "total fat," 41.4 per cent, of "neutral fat," and 17.5 per cent, of "fatty acid"; on a milk diet, however, the "total fat" was reduced to 26.2 per cent., the "neutral fat" to 25.9 per cent., and the "fatty acid" to 0.3 per cent. The digestibility of fat, in the form of meat, is also influenced to a certain extent by the condition of the gastric secretion. We mentioned, when considering the physiology of pancreatic digestion, that collagen is not acted upon by pancreatic secretion; fat, therefore, which is enclosed in a mesh of connective tissue is liable to be protected from digestion in the intestine unless it has been previously acted upon in the stomach, so that defi- ciency or absence of hydrochloric acid in the stomach may lead to the appearance of an abnormal proportion of fat in the fasces. An excessively fatty diet may also increase the fat content to an unusual degree, both normally and in cases of disease of the pancreas. In this connection we may mention that we have found that in some apparently 220 The Pancreas: Its Surgery and Pathology healthy persons there appears to be an inability to digest more than a very limited amount of fat, and, as their powers in this direction can, at least in some instances, be improved by the administration of preparations of pan- creas, it is possible that the difficulty is due to a congenital or acquired deficiency of that organ. In addition to diseases of the pancreas, biliary obstruc- tion, defective gastric digestion, and excess of fat in the food, abnormal quantities of fat may be passed in the faeces from faulty absorption, due either to disease of the intestinal mucous membrane or to obstruction of the lymphatics. Such conditions are, however, compara- tively rare, and are chiefly met with in extreme intestinal tuberculosis, amyloid disease, sprue, etc. Salomon states that purely functional disturbances of fat digestion may occur, but there is as yet little to substantiate this. The recognition and differentiation of the steatorrhoea met with in these conditions from that due to pancreatic disease can only be arrived at by attention to other signs, for the steatorrhoea itself presents no special characters by which it can be recognised. Thus in a case of intestinal tuber- culosis we found that the "total fat" constituted 33.1 per cent, of the dry weight of the faeces, and that 21.4 per cent, of this was "neutral fat" and 11. 7 per cent, "fatty acid." In another patient suffering from the same disease the faeces were found to contain 61.3 per cent, of "total fat," 42.5 per cent, of "neutral fat," and 18.8 per cent, of "fatty acid"— figures which closely resemble those met with in steatorrhoea of pancreatic origin. We have also had the opportunity of examining the faeces from a considerable number of cases diagnosed as sprue, and in all have found a large excess of fat, which consisted chiefly of "neutral fat." Although there is no doubt that, in this disease, the steatorrhoea is in part due to defective absorption, from atrophy of the mucous membrane of the intestine, we have come to the conclu- Chemical Pathology 221 sion that, in some instances at least, the condition is contributed to by concurrent: disease of the pancreas. In one case of this description, in which we had reason to believe, from an examination of the urine and faeces, that the pancreas was diseased, material benefit followed an operation performed by one of us for the relief of pan- creatitis. Before the operation the faeces showed 55.6 per cent, of "total fat," of which 51.4 per cent, was ' ' neutral fat' ' and 4. 2 per cent. ' ' fatty acid. ' ' Six months after the operation, when the patient was put on a simi- lar diet, there was 43.0 per cent, of "total fat," of which 22,3 per cent, was "neutral fat" and 20.7 per cent, "fatty acid ' ' ; thus suggesting that the fat-splitting process was being more efficiently carried out, and that a somewhat larger proportion of fat was being absorbed. The utilisation of proteids after complete and partial extirpation of the pancreas has been investigated by Abelmann. He found that when the gland was com- pletely removed in dogs, only 44 per cent, of the albumin given as food was absorbed, and that when a portion of the organ was left behind, 54 per cent, of the proteid was made use of. Some part of this deficient absorption he ascribed to the presence of undigested fat in the intestinal contents. Administration of pigs' pancreas to the depan- creatised animals was found to increase the amount of utilised albumin to from 74 to 78 per cent. De Renzi and Cavazzani showed that after extirpation of the pancreas the amount of nitrogen in the faeces was increased, and Sandmeyer demonstrated that after partial extirpa- tion of the gland from 62 to 70 per cent, of the albumin of the food was unused. Clinically, Hirschfeld found that in certain cases of diabetes, possibly due to disease of the pancreas, as much as 31.8 per cent, of the nitrogen of the food reappeared in the faeces, and Weintraud states that in a case of chronic pancreatitis, in which the diag- nosis was confirmed post-mortem, 45.2 per cent, of the 2 22 The Pancreas: Its Surgery and Pathology ingested proteid was found in the stools. Miiller re- ported that in his cases the absorption of proteids was only slightly affected. The appearance of undigested muscle fibres in the stools has been described by numerous observers, both clinically and after partial or complete extirpation of the pancreas in animals. Fles, who was the first to draw attention to the value of this symptom in the diagnosis of pancreatic disease, states that in his case the muscle fibres disappeared after the administration of calf's pancreas. As stated in the table on page 214, our own investigations of the faeces have shown that undigested muscle fibres can be found more frequently in those cases where the functions of the pancreas are interfered with, than in those in which it is apparently normal, and that, since they were discovered in twenty out of twenty-four cases of cancer of the pancreas, but in only sixteen out of fifty-six cases of simple pancreatitis, their appearance in the stools, other things being equal, indicates a serious lesion of the gland. In some cases of pancreatic disease, undigested muscle can be detected in the stools with the naked eye, but, in the majority, they are only found on microscopical examination. It is impossible, however, to infer that the functions of the pancreas are disturbed from the appearance of muscle fibres in the faeces alone, for, excluding their presence from an excessive amount of meat having been taken in the diet, they may also be found in cases where, owing to increased peristalsis, or putrefactive changes, leading to secondary diarrhoea, they are hurried through the intestine before they have had time to be digested. Normally the stomach shares only to a slight extent, according to Schmidt, in the disso- lution of muscle, its chief action being the digestion of the connective tissue of the meat ; it is probable, however, that when the pancreatic juice is diminished or absent, gastric digestion may be continued lower down in the Chemical Pathology 223 intestine than is usually the case, and that consequently proteid digestion may not be as incomplete as might at first sight be expected. On the other hand, defective gastric secretion may lead to imperfect digestion of muscle, for the pancreatic juice being presented with more or less solid masses of fibres, bound together by connective tissue, can only attack them slowly from the surface, in- stead of dealing quickly with separated cells or groups of cells. Reduction or failure of the pancreatic secretion might be expected to lead to impaired digestion of starchy foods and the appearance of an excess of carbohydrate in the stools. The observations made by various inves- tigators on the faeces, however, have shown that only a small proportion, or none at all, of the carbohydrate taken in the food is excreted unchanged in cases where these conditions exist. According to Abelmann, 20 to 40 per cent, of the amylaceous material ingested reap- pears in the fasces in animals from which the pancreas has been extirpated, while Mtiller was unable to find any evidence that more carbohydrate was present in the stools of patients suffering from diseases of the pancreas than in those of normal individuals. Our own investigations on the faeces in pancreatic dis- ease tend to support the conclusions of Miiller, on the whole ; for, although we have found that in some instances a larger proportion of carbohydrate than is normally present in the stools of persons on a mixed diet could be detected, this was by no means constant, even in well- marked cases. The loss of weight and inability to accu- mulate fat, in spite of an abundant carbohydrate diet, points, however, to a diminished assimilation in excess of that indicated by the condition of the faeces, and it is probable that the figures given by analysis of the stools cannot be taken as a true index of the loss to the organism of carbohydrate material. The difference between the 224 The Pancreas: Its Surgery and Pathology amount assimilated and that present in the stools is probably to be explained by bacterial action, the starch of the food being slowly converted into maltose, and this in its turn being split up into lactic acid, acetic acid, alcohol, carbon dioxide, hydrogen, etc., by micro-organ- isms in the intestine. There is thus a loss of caloric potential which leads to inanition. The fact that many dyspeptics continue very thin, although they take an abundance of carbohydrate food, is possibly to be ac- counted for, as Herter suggests, by a diminution in the secretion of pancreatic juice, and the flatulence of which they complain may also be due to the consequent accumu- lation of carbon dioxide and other gases, while the drow- siness, with headache after meals, may arise partly from the absorption of alcohol and various organic acids. It seems likely that in disease of the pancreas all the three chief ferments usually suffer diminution together, but there is reason to think that, under some circum- stances, they are not diminished to an equal degree. In pancreatitis, due to obstruction or an ascending catarrh of the ducts, fat-splitting, proteolysis, and starch-conversion are no doubt equally affected, but it is said that in fever the ability to digest starches and fat may be much more impaired than the capacity of the pancreatic juice to act upon proteids, an observation which can only be explained on the assumption that trypsin under these circumstances is more abundantly secreted than the other ferments (Herter). The stools in cases of advanced pancreatic disease generally present very typical characters; they are fre- quent, bulky, soft, white, have usually an acid reaction and a peculiar odour. Their bulk is partly to be attrib- uted to the abnormal quantity of undigested material, particularly fat, passed through the bowel, and partly to the excessive fermentation which takes place in the lower part of the intestine. Their frequency is due in Chemical Pathology 225 part to their bulk, and is also no doubt contributed to by the excess of irritating by-products they contain. Considerable difference of opinion exists as to the cause of the white appearance of the fceces when the pancreatic secretion is much diminished or excluded from the intestine. Muller, as we have seen, attributes steat- orrhoea entirely to absence of bile, and, as it is well known that obstruction of the biliary passages gives rise to clay- coloured motions, he would refer the absence of colour also to that cause. Since many cases of pancreatic dis- ease in which the typical white stools exist are asso- ciated with more or less complete blocking of the common bile-duct by gall-stones, or growth in the head of the pancreas, the absence of bile-pigment is without question a frequent contributory factor, but that it is not the complete and invariable explanation there is abundant evidence to show. We have on several occasions met with cases of pancreatitis with white stools where there was no jaundice, and no evidence of biliary obstruction at operation, and in which a chemical examination of the fasces demonstrated a well-marked reaction for sterco- bilin. The case already quoted on page 217, in which, after cholecystenterostomy, the whole of the bile, which had previously been escaping by a fistula, was turned into the small intestine, demonstrated very clearly that the presence of the biliary secretion is not sufficient to ensure a return of the normal colour when the pancreatic juice is still absent, for the appearance of the faeces was practically unchanged by the operation. As far back as 1856 Claude Bernard, writing of dogs whose pancreas had been destroyed, stated that "it is remarkable that bile only colours the faeces a very bright yellow, whilst with the pancreatic juice the bile takes a very brown colour," thus suggesting that the pancreatic juice contributed indirectly to the colour of the faeces. Attention was drawn to this observation thirty-three years 15 2 26 The Pancreas: Its Surgery and Pathology later by T. J. Walker, in a paper read before the Royal Medical and Chimrgical Society, in which he described two cases of pancreatic disease where there were clay-coloured stools, although the liver and bile-passages were found to be normal post-mortem. He suggested that the white stools, to which he drew attention as indicative of disease of the pancreas, depended for their characteristic appear- ance upon the absence of the action of the pancreatic juice upon the bile-pigment they contained. This view has been supported by W. Gordon, who reported a case of pancreatic disease, with copious vomiting of green bile, in which the motions were sometimes clay-coloured and at other times cream or primrose-coloured, but never brown. Neither of these observers produced any evi- dence, beyond that afforded by mere inspection, that stercobilin was absent, or very much diminished, in the cases they report, nor did they take into account the enormous excess of fat present in the faeces in serious cases of pancreatic disease. It is to this large excess of fat that, in our opinion, the abnormal colour is chiefly due. The evidence on which this opinion is based may be summarised as follows : ( i ) Quantitative examination of the stools from a large number of our cases has shown that the colour varies directly with the percentage of fat present, the largest amount being found in those specimens which are white to the naked eye, and the least in those which approximate to the normal colour. (2) The glis- tening white appearance is most marked in those speci- mens which are found microscopically to contain large numbers of fatty acid crystals, in part probably for the same reason that snow and other substances of a finely crystalline character appear white in mass. (3) The white stools on being heated on the water-bath assume a dark brown colour. (4) Removal of the fat with ether leaves a residue of a dark brown colour, similar to that obtained from normal fseces. (5) Stercobilin can be Chemical Pathology 227 demonstrated chemically in all specimens not derived from patients in whom there is complete obstruction of the bile-passage by cancer of the head of the pancreas, gall-stones, etc., the amount being proportional to the quantity of non-fatty residue. In cases of pancreatic disease associated with incomplete obstruction of the biliary passages the amount of stercobilin varies with the degree of obstruction. (6) The white stools occasionally met with in tuberculosis of the intestine, and some other conditions, where there is defective absorption of fat, are similar microscopically and chemically to those seen in typical cases of pancreatic disease, although there is no obstruction to the free flow of bile, and the pancreas is not affected. These considerations point, we think, to an excess of fat in the stools being probably the most important element in the production of the white stools in seri- ous pancreatic disease. But they do not exclude other and contributory factors, and that such exist is sug- gested by the fact that when such stools are exposed to the air they are .sometimes seen to assume a darker colour on the surface. The acid reaction of the fseces in many cases is possibly associated with a modification of the flora of the intestine, and it appeared to us not im- probable that this might cause a partial or complete reduction of the stercobilin to a colourless compound which, on contact with the oxygen of the air, was slowly converted into the normal colouring-matter of the stools. Experimental proof of this was sought by taking a speci- men of normal, dark brown fseces, which had a faintly acid reaction, thoroughly mixing it with normal saline, so as to form a thin paste, and dividing it into two por- tions, which were placed in sterile test-tubes marked "A" and "B." The A-tube was plugged with wool and used as a control. To the B-tube was added, with a sterile platinum wire, a minute fragment of fseces from a typical 228 The Pancreas: Its Surgery and Pathology white pancreatic stool, and it was then plugged with wool. Both tubes were placed in the incubator and kept at 37° C. In twenty-four hours no change had taken place in the A-tube, but the lower part of the contents of the B-tube was distinctly lighter in colour than the upper portions, and than the control. The tubes were returned to the incubator and examined daily. The control and the upper part of the B-tube gradually became slightly darker, but the alteration in colour of the lower part of the latter previously noticed increased until the fourth day, when it was found to be of a light grey-brown appearance and presented a very marked contrast to the darker layers above. No further discharge of colour was observed, although the experiment was continued for several weeks. This result, incomplete as it was, pointed to the pres- ence in the pancreatic stool of organisms which, growing anaerobically, caused changes in the faecal pigment that resulted in partial decolourisation. On repeating the experiment with specimens grown under anaerobic con- ditions a similar change was obtained, only that in this instance practically the whole of the inoculated tube was affected. On spreading this light-coloured material on a dish, and exposing it to the air, it slowly darkened and assumed very much its original appearance. On repeat- ing the experiment with faeces of alkaline reaction no- alteration in colour could be produced. The very small proportion of fat in the particular speci- men of fseces first submitted to experiment (5.3 per cent.) is against the changes observed being due to some altera- tion in that constituent, but it will be observed that we were unable to obtain with this specimen the dead-white appearance met with in characteristic cases of pancreatic disease, possibly because of the small proportion of fat it contained. It appears probable, therefore, that the characteristic white appearance of the stools met with in serious cases of pancreatic disease, in which there is no Chemical Pathology 229 obstruction of the biliary passages, is due chiefly to the presence of an excess of fat, particularly to the crystalline fatty acids, but partly also to the reducing action of bacteria growing anaerobically in an acid medium. In pancreatic disease associated with biHary obstruction the absence of bile-pigment, or its diminished amount, is also no doubt a contributory factor. On referring to the table on page 2 14 it will be seen that in fifty-eight of the eighty cases in which there was evi- dence of disease of the pancreas the fresh fseces had an acid reaction, while in sixteen they were neutral or ampho- teric to litmus, and in six distinctly alkaline. The presence or absence of jaundice appears to exert little or no effect upon the reaction of the stools when the pancreas is diseased, for the proportion in which they were acid is about the same in jaundiced and in non- jaundiced patients. In simple jaundice, unaccompanied by disease of the pancreas, and in cases where there were calculi in the biliary passages but no bile-pigment in the urine, on the other hand, the stools have generally been alkaline in reaction. This is probably to be attributed to the pres- ence in pancreatic disease of free fatty acids, whereas in the non-pancreatic cases the excess of fat is due to com- bined fatty acids or soaps. The peculiar sour smell of the white stools in typical cases of diseases of the pancreas is also due, in all probability, to the higher free fatty acids they contain. Strasburger found, as a rule, a strik- ing diminution in the amount of bacteria contained in icteric stools, in spite of the generally accepted view that bile possesses antiseptic properties. It is probable, therefore, that there is a lessened rather than an increased degree of putrefactive change in the fatty stools met with in such cases. Where, however, the amount of proteid residue is at the same time increased, as the result of pan- creatic and intestinal affections, putrefaction may occur and the fasces become alkaline. Under these conditions 230 The Pancreas: Its Surgery and Pathology the acid reaction due to the fatty acids may be masked and the feces be neutral, amphoteric, or even alkaline to litmus. The association of general enteritis, including chronic colitis, with disease of the pancreas may thus account for the alkaline reaction of the stools met with in a few cases. The hlood changes that result from diseases of the pancreas have not, as yet, received much attention from investigators, but that they are important and interesting is shown by the profound alterations in both its morpho- logical and chemical characters that are met with in serious and advanced lesions, and to a less extent in milder types of disease. The hsemorrhagic tendency in pancrea- titis, to which attention was drawn by one of us in the Hunterian Lectures for 1904, is an indication of the altered blood state induced by pancreatic lesions, and the changes in the urine to which we shall presently refer, also point in the same direction. Besides hccmorrhagic pancreatitis, and the form of local haemorrhage to which we have referred under the term "pancreatic apoplexy," there is in many pancreatic affections a tendency to general haemorrhage, and pa- tients often complain that they very readily bruise. The fact that this tendency is most marked in cancer of the head of the pancreas, and in pancreatitis associated with jaundice, naturally suggests that it is dependent upon the cholsemia, but since a haemorrhagic tendency is also encountered in patients suffering from pancreatitis un- attended by jaundice, it cannot be altogether due to that cause. Estimations of the coagulation time of the blood by one of us, with Wright's method, have shown that it may be prolonged in cases of cancer of the head of the pancreas to seven or eight minutes, or even longer, and that in pancreatitis with deep jaundice similar figures may be obtained. In pancreatitis without jaundice, Chemical Pathology 231 although the alteration is not so marked, coagulation times of four or five minutes are not uncommon. We were at one time disposed to think that the explana- tion of the haemorrhagic tendency, and the delayed coagula- tion time, in pancreatic disease might be dependent upon the presence of glycerine in the blood stream, derived from areas of fat necrosis, but we have not been able to ob- tain any experimental evidence in support of this theory, either directly by examining the blood or indirectly from the urine. It is w^ell known that a diminution of the lime salts in the blood leads to a tendency to haemorrhage, and that the haemorrhagic tendency in pancreatic disease is dependent on this cause is highly probable for several reasons. We have found that the administration of cal- cium chloride to patients suffering from diseases of the pancreas not only reduced the coagulation time, often by several minutes, but is also an efficient preventive of the haemorrhage that is liable to occur in such cases during and subsequent to operation. Further, the fact that pancreatic calculi contain 50 per cent, or more of calcium salts, whereas the normal secretion contains under 2 per cent., and the presence also of calcium oxalate crystals in the urinary deposit of many cases of pancreatitis, suggest that inflammation of the gland is associated with a dis- turbance of metabolism which results in an abnormal excretion of the lime salts and a consequent improverish- ment of the blood. There is usually a diminution in the number of erythro- cytes in pancreatitis which, in advanced and untreated cases, may be very marked indeed, even when jaundice is absent. Thus, in one case of chronic pancreatitis, in which operation was refused, we found 3,120,000 red cells per cubic millimetre; nine months later there were 1,889,000, and three months subsequently 1,501,000. In another case, where operation was followed by rapid improvement there were 3,472,000 red cells per cubic 232 The Pancreas: Its Surgery and Pathology millimetre before operation and 4,634,000 three weeks subsequently. In some instances, and especially where the disease is of long standing, we have found that the haemoglobin has not suffered a proportional decrease with the red corpuscles, and there has consequently been a high haemoglobin index, similar to that found in pernicious anaemia. In one case of advanced chronic pancreatitis, which had previously been diagnosed as malaria, and in which a floating biliary calculus was removed from the common duct by one of us, a haemoglobin index of 1.4, with 2,525,000 erythrocytes per cubic millimetre, was found shortly after the opera- tion. Six months later the haemoglobin index was 1.5, and a blood count showed 1,735,000 red cells per cubic millimetre. The patient passed from under our observa- tion, but we have heard that he died shortly afterwards with all the symptoms of pernicious anaemia, Unfortu- nately no post-mortem examination was made. Another and similar case, where a haemoglobin index of 1.5 and a red blood count of 1,293,000 were obtained, was examined after death, and the pancreas found to be small, hard, and cirrhosed. These cases suggest that in some in- stances inflammatory changes in the pancreas are liable to be associated with alterations in the blood, similar to those met with in so-called idiopathic pernicious anaemia, and that in making a prognosis in cases of pan- creatitis the condition of the blood must be considered. In our experience it is rare to meet with a well-marked leucocytosis in diseases of the pancreas, but up to the present our observations have been limited to cases of chronic pancreatitis and cancer of the gland. So far we have not had the opportunity of examining the condition of the blood in acute inflammation, but Woolsey in three cases of acute pancreatitis obtained leucocyte counts of 39,000, 17,600, and 26,000 per cubic millimetre respec- tively. Chemical Pathology 233 The changes met with in the urine in diseases of the pancreas arise in part from the altered conditions existing in the intestinal tract, and in part from perverted metab- olism and excessive tissue waste. Ethereal Sulphates and Indican. — Since a reduced se- cretion of pancreatic juice is followed by an impaired digestion of proteids, and these, as we have seen, are likely to be attacked and broken down by bacteria, it might be expected that the urine would in such cases show signs of excessive intestinal putrefaction, in the shape of an increased excretion of ethereal sul- phates and a pathological excess of indican. According to Herter, this does in fact occur, for he states that when both the bile and pancreatic secretion are completely excluded from the intestine there is an excess of indican and the ethereal sulphates are always very largely in- creased, the proportion to preformed sulphates rising to I : 6, or I : 4, or even i : i, as compared with the normal of about I : 10. Edsall, on the other hand, considers that a diminution in the amount of ethereal sulphates in the urine is an indication of pancreatic disease, for he points out that, although the products of proteolytic digestion are readily decomposed by bacteria, the native albumins are not. If, therefore, there is little or no proteolytic digestion going on in the intestine, as is the case in severe lesions of the pancreas, the products of bacterial activity will be lessened and the quantity of ethereal sulphates and indican in the urine decreased. Pisenti has estimated the amount of indican in the urine of dogs before and after tying the pancreatic duct. In one instance he found 11.70 to 19.90 mg. and in another 1 5.0 to 2 1. o mg. per day, before the operation, as compared with 4.30 to 4.20 mg., and 6.0 to 9.0 mg., per day, respec- tively, after ligature, thus showing a marked diminution. The administration of pancreas-peptone to animals in which the duct had been tied he found increased the 234 The Pancreas: Its Surgery and Pathology quantity of indican excreted. In 1886 Gerhardi reported a case of pancreatic disease which he had successfully diagnosed during life from the absence of indicanuria, when the clinical symptoms suggested obstruction of the upper part of the intestine. Absence of indicanuria has also been observed by Stefanani in a case of purulent pancreatitis, and by Biondi in a case of adenoma of the pancreas. The question has been carefully investigated by Katz in depancreatised dogs. He states that when the animals were fed with easily digested and rapidly absorbed food, the excretion of ethereal sulphates was low — 0.032, 0.022, 0.069 gram daily; but that when a diet of pure meat was substituted the daily excretion was unusually high— 0.076, 0.089 gram, although readings as low as 0.024 were sometimes obtained even under these conditions. He also failed to detect any diminution in the amount of indican after lesions of the pancreas; in fact, in many instances there was marked indicanuria. On a pure meat and milk diet he found that there was an abundance of indican in the urine, the amount being greater, and its increase distinctly marked, on the day following the attack, especially in those cases where the animals took no nourishment after the operation, and even when they died quickly from duodenal necrosis. In those animals which long survived the operation no dim- inution in the excretion of indican was observed. Similar results have also been obtained by de Renzi. Schlagen- haufer records an increase of indican in a case of syphilitic interstitial pancreatitis that he investigated, and Hen- nige has referred the indicanuria found in cholera and lead-colic . to an alteration in the pancreatic secretion caused by nervous influences. Our own investigations of the relation between the preformed and ethereal sulphates in cases of pancreatic disease have given such varied results that we have come to the conclusion that they are due to factors which have Chemical Pathology 235. no direct relation to the activities of the gland. Simi- larly, although we have found an excess of indican in 49 per cent, of our cases of chronic pancreatitis and in 54 per cent, of cancer cases, there has been no relation be- tween the intensity of the lesion and the degree of indi- canuria. The truth appears to be that, although absence or a diminished secretion of pancreatic juice provides conditions under which there may be an abnormal pro- duction of aromatic derivatives in the intestine, these do not make their appearance in the urine unless there is at the same time some affection of the intestinal wall which facilitates absorption, and we have therefore come to look upon an excess of indican and ethereal sulphates in the urine in pancreatic diseases as an indication of an associated enteritis. Bile. — Owing to the anatomical relations of the common bile-duct, the duct of Wirsung, and the head of the pan- creas, circumstances which interfere with the free flow of the pancreatic secretion into the intestine are, in many cases, likely to obstruct the passage of bile at the same time, giving rise to jaundice and the appearance of bile in the urine. Hence the urine in diseases of the pancreas is frequently of a deep yellow or brown colour, and gives a reaction for bile-pigment. Bilious urine and jaundice are, however, by no means constantly found in diseases of the pancreas, even in pancreatitis associated with gall-stones in the lower part of the common bile-duct. Bile-pigment has been detected in the urine in 62 per cent, of our cases of chronic pancreatitis associated with cholelithiasis, and in only 16 per cent, of those in which no biliary calculi could be found in the common duct at the time of operation. Bile-pigment, in large amounts, was present in the urine of all the twenty-four cases of malig- nant disease of the pancreas included in the list on page 214. The relation of urobilin to the bile-pigments has been the 236 The Pancreas: Its Surgery and Pathology subject of much controversy, but it is now generally accepted that if the bile is completely shut off from the intestine no urobilin can be found in the urine. The fact that the urine of only three of the above mentioned cases of malignant disease gave a reaction for urobilin is interest- ing in this connection, for they were the only three in which stercobilin could be found in the faeces. A patho- logical excess of urobilin was present in 61 per cent, of our cases of chronic pancreatitis with an obstruction of the common bile-duct, and in 40 per cent, of those in which no obstruction existed at the time of operation. The urobilinuria coexisted with jaundice, and the presence of bile-pigment in the urine, in 43 per cent, of the former, but in only 6 per cent, of the latter. Azoturia. — Disturbances of intestinal digestion are said frequently to give rise to an increased excretion of nitro- genous compounds in the urine, but as the increase appears to be closely related to excessive putrefactive changes in the intestine, and, as we have seen, these are not by any means a constant accompaniment of diseases of the pan- creas, no constant variation from the normal in this respect can be looked for. Azoturia is well known to occur in diabetes, and it has been observed, without glycosuria, by de Dominicus, Hedon, and Thiroloix after extirpation of the pancreas in animals. The excess is here, however, probably due to abnormal tissue destruc- tion, for it is not met with after partial extirpation of the gland and is associated with a constant dextrose-nitrogen ratio. In most of our cases of pancreatitis the excretion of urea has not been excessive, the uric acid has varied little from the normal, and the total nitrogen, in the few cases in which we have estimated it, has fallen within normal limits. In cancer of the pancreas we have usually found that there was a subnormal proportion of urea. The excretion of phosphates is said to be increased by disease of the pancreas, and de Dominicus states that Chemical Pathology 237 an increase of phosphoric acid is characteristic of pancre- atic lesions, even in those cases where there is no glyco- suria. But since the chief source of the phosphoric acid in the urine is the food, the nature of this will largely control the output. Thus David Young, experimenting with the case of pancreatic infantilism reported by Byrom Bramwell, found that when the patient was taking a milk diet the amount of phosphoric acid was extremely small, but that during the administration of pancreatic extract the quantity underwent a very marked and rapid increase. The explanation he offers is that the caseino- gen of the milk was the source of the phosphorus in the urine. In the stomach it was broken up into paranuclein, containing 4 per cent . of phosphorus , and a proteid . Para- nuclein itself is insoluble, but when it is acted upon by the alkaline pancreatic secretion it is dissolved and split into paranucleic acid and an albumose, from which the phosphoric acid of the urine was derived. We have been unable to detect any marked variation from the normal as regards the excretion of phosphates in any of our cases, and cannot confirm the statement of de Dominicus or the observation of Young, although with regard to the latter our investigations have been limited to two cases of malignant disease of the pancreas. The excretion of chlorides in our cases of pancreatitis has not, as a rule, been noticeably disturbed, but we have found that in cancer of the pancreas the output has been frequently subnormal, possibly owing to the presence of pathological exudates. Acetone Bodies. — Among the effects produced by extir- pation of the pancreas in animals is the appearance of acetone, diacetic acid, and, occasionally, /^-oxybutyric acid in the urine. Thus Baldi found 1.043 grams of acetone on the second day after the operation, 0.652 gram on the third day, and later 0.385 gram, 0.282 gram, and 0.049 gram, as compared with the normal of 0.0 gram to 238 The Pancreas: Its Surgery and Pathology 0.105 gram. Minkowski observed that the excretion of these substances was most marked when the animals ex- perimented upon became emaciated, and that the largest amounts of /?-oxybutyric acid were found when the quan- tity of sugar was diminishing in the later stages of the disease. Acetone and diacetic acid have been found in the urine in four out of the five cases of acute pancreatitis in which we have had the opportunity of investigating the urine, in 29 per cent, of our cases of chronic pancrea- titis, and in 31 per cent, of the cases in which there was a malignant growth of the pancreas. The cases of chronic inflammation in which these substances were found were all of some standing and showed evidence of consider- able tissue wasting. In one case where the pancreatitis had followed typhoid fever, and the patient was in an extremely serious condition when she came under our observation, the urine contained enormous quantities of both acetone and diacetic acid. Three days after opera- tion the urine was again examined, and acetone bodies were found to be still present in very large amounts. As the patient was becoming comatose and her condition was serious, she was injected intravenously with three pints of normal saline solution. This caused a temporary increase in the excretion of the acetone bodies, but in about seventy-two hours they began to diminish, and six days after the injection they could no longer be detected. The patient's appetite and general condition improved at the same time, and she eventually made a complete recovery. The source of the acetone bodies, both in wasting conditions and diabetes, has been much debated. It would appear that they are formed within the system and not in the intestinal tract, for the administration of purgatives does not diminish the quantity of acetone ex- creted in the urine ; indeed, according to von Noorden, Chemical Pathology 239 it may occasionally cause an increase. The carbohy- drates were at one time looked upon as the material from which they originated, but this view was discarded, and it was then supposed that they were formed in the process of disintegration of proteids. More recently attention has been turned to the fats, since /5-oxybutyric acid and its derivatives, acetone and diacetic acid, can be derived from some fats by a simple chemical process, and at the present time this theory has the largest num- ber of supporters. It is also possible, however, that they may arise synthetically within the body from simple carbon derivatives resulting from the breaking down of carbohydrates, proteids, and fats, and that this is a probable explanation is suggested by the conditions under which they may make their appearance in the urine and the circumstances that have been found to control the quantities excreted. Calcium Oxalate. — In examining the urine from cases of pancreatic disease we were early struck by the frequent occurrence of well-marked deposits of calcium oxalate crystals in many of them. Further experience, now extending to some five hundred examinations, has only served to emphasise this early observation, and to suggest that there is probably a connection between chronic inflammatory lesions of the pancreas and oxal- uria. Microscopical examination of the centrifugalised deposit from the urine has shown that oxalate crystals were present in 63 per cent, of our cases of chronic pan- creatitis, or in 73 per cent, if those in which the urine contained bile-pigment be excluded. The crystals are generally numerous and are frequently very small, so that their nature can sometimes be only recognised by examin- ing them with high powers of the microscope and by their chemical reactions. Quantitative examination in five cases has shown that the urine contained an actual excess of oxalic acid, and that the deposit was not merely caused 240 The Pancreas: Its Surgery and Pathology by physical conditions. Thus in one instance the output for twenty-four hours was found to be 0.03 gram, and in another 0.037 gram . We have not observed this condition in acute pancreatitis and rarely in cancer of the pancreas. Its association with chronic pancreatitis is interesting in view of the fact that a similar deposit of oxalate crys- tals is not infrequently met with in diabetes, and that a diminution in the output of sugar in this disease is accom- panied by an increase in the oxalate deposit (vicarious oxaluria) . Diabetes has also been noticed occasionally to follow long-continued oxaluria. Experimenting on dogs, J. Scott found that potassium oxalate was a depressing drug and that large doses increased nitrogenous metabol- ism, but he was unable to induce glycosuria by the subcu- taneous injection of from 0.25 to 0.75 gram. The origin of the oxalic acid in the urine is disputed. Part, no doubt, is directly derived from the food, but part probably arises from the breaking down of purin bodies within the organism. The experiments of Helen Baldwin show that, in dogs, the excretion of oxalates is increased by the administration of cane-sugar and glucose for long periods. Herter attributes this to the excessive fermenta- tion and induced gastritis, but, since Ssobolew has shown that overfeeding animals with carbohydrates gives rise to changes in the islands of Langerhans, it is not improbable that the oxaluria found in these experiments, and in chronic pancreatitis, may be due to a disturbance of metabolism arising from changes in the functions of the cell islets. Carbohydrates. — The alterations in the urine that we have so far considered, although sufficiently striking in many instances, are not peculiar to diseases of the pan- creas . They probably result directly or indirectly from the disturbances of digestion, or from alterations of internal metabolism, to which the pancreatic lesions give rise, but they may also be brought about by other and quite distinct Chemical Pathology 241 causes. So far as experimental research goes, however, the appearance of certain carbohydrates in the urine is directly and peculiarly the result of a failure on the part of the pancreas to perform its functions in the internal economy of the body. The question of pancreatic diabetes will be fully considered in a subsequent chapter, so that it will be sufficient to mention here that extirpation of the pan- creas in animals has been shown to give rise to the ap- pearance of dextrose in the urine, and that more or less profound changes in the structure of the gland have been met with in many cases of human diabetes. In addition to dextrose, however, other sugars have been described as present in the urine in a few cases of pancreatic disease. Le Nobel has described a case of glycosuria with fatty stools, in which there w^as a reducing substance in the urine having the characters of maltose. Von Ackeron found a similar substance in a case of pancreatic carci- noma. Rosenheimhas recorded a case of maltosuria, w4th steatorrhoea and considerable loss of w^eight, in which interstitial pancreatitis w^as found post-mortem. A similar case has also been reported by Lepine. In an examination of two hundred and forty-five specimens of urine from cases of pancreatic disease by the phenylhy- drazin test, we met with two in which an osazone having the characters of maltosazone was obtained in sufficient quantities for a careful examination, and five in which a small deposit of crystals, probably also maltosazone, was given. One of the former was a patient on whom an operation for stone in the common bile-duct had been performed five years previously. Maltosuria is thus a rare condition, and it is doubtful how far it is directly dependent upon disease of the pancreas. Attention was first directed to the occurrence of pen- tosuria by the observations of Salkowski and Jastrowitz, who discovered a pentose in the urine of a morphine-eater, with temporary glycosuria, in 1892. Previous to this 16 242 The Pancreas: Its Surgery and Pathology observation pentoses had been met with only in plants, and it was believed that the animal organism was incapa- ble of building them up. Subsequently two cases of pure pentosuria, without any dextrose, were described by Sal- kowski and Blumenthal. The excretion of pentose in these cases was found to be independent of the diet, and did not in any way affect the general condition of the patients. A diminution of the amount of indican was observed in both. On this account, but more particu- larly because the osazone obtained from the urine ap- peared to be identical in its appearance, melting-point, and solubilities with that obtainable from the pancreas, Salkowski assumed that the pentosuria was dependent upon an abnormally increased formation and destruction of the nucleo-proteid of that organ. This assumption derived support from the observations of Kulz and Vogel, who found that a pentose could be detected in the urine of starving dogs after removal of the pancreas. They also examined the urine of eighty diabetics and found a well-marked pentose reaction in sixty-four, in twelve the test gave doubtful results, and in four no reaction could be obtained. Salkowski and Blumenthal, however, were unable to detect any pentose in ten diabetics whose urine they carefully examined, and in none of the cases of in- flammatory or malignant disease of the pancreas that we have investigated has any evidence of the presence of a pentose been found. Although Salkowski regarded pen- tosuria as an important indication of pancreatic disease, it is so rarely met with that it is of little practical value, and apart from the evidence already quoted, there is no proof that when present it is dependent upon lesions of the gland. It has also to be remembered that minimal traces of pentose may be met with after the ingestion of large quantities of plums, cherries, bilberries, and other substances comparatively rich in that variety of sugar. The pentose excreted under these conditions is, however. Chemical Pathology 243 the dextrorotatory form, whereas that met with in chronic pentosuria, of which a few cases have since been reported, is optically inactive. The ''Pancreatic'' Reaction (Cammtdge). — Although it cannot be considered as proved that the presence of a pentose in the urine is dependent upon disease of the pancreas, the results of the investigations that we have been carrying out since the early part of the year 1901 suggest that, in inflammatory lesions of the gland, there is excreted by the kidneys a substance which, on hydrolysis, yields a body giving the reactions of a pentose. The initial stages of these investigations were referred to and described in our Hunterian, and Arris and Gale lectures, delivered at the Royal College of Surgeons in 1904, and the results of further research were embodied in a paper read by one of us before the Royal Medical and Chirurgical Society in March, 1906. The most striking indication of inflammatory lesions of the pancreas is undoubtedly furnished by the dis- covery of fat necrosis, either during life or post-mortem. This condition is most characteristically met with in acute and gangrenous pancreatitis, but it is not uncommon to find less marked changes in chronic inflammation of the gland, and it appeared to us possible that, even in those cases of chronic pancreatitis where no visual evidence existed of the fat-splitting process, there might still be molecular changes whigh could be recognized by the alterations produced in the chemical composition of the blood. The hemorrhagic tendency we had noticed, and the microscopical changes found in the blood of patients suffering from diseases of the pancreas, pointed to there being some connection between the two conditions, and, bearing in mind the effects produced in animals by the subcutaneous injection of glycerine, it occurred to us that in man the continued action of minute doses, such ( 244 The Pancreas: Its Surgery and Pathology as fat necrosis would give rise to, might produce such a blood state as these patients exhibited. Starting on the theory that there was some such con- nection between fat necrosis and the blood state found in pancreatic disease, we commenced our investigations by examining the blood of several cases of pancreatitis for glycerine and glycerine derivatives. But, as we quickly realised that such direct proof was not practicable, owing to the small amounts of blood available at the bedside, and to the relatively small quantity of glycerine likely to be present in it, we turned our attention to the urine, in the hope of finding indirect evidence in favour of our hypothesis. The well-known selective power of the kid- neys, by which they detect and pick out abnormal constit- uents of the blood, favoured the view that the soluble products of fat necrosis, or their derivatives, might cause changes in the urine which, from the large bulk of material readily available for examination, could be satisfactorily detected. It is well known that glycerine on being boiled with nitric acid gives rise to glycerose, which can be recognised by the osazone that it forms with phenylhy- drazin. When this test was applied to the urine from known cases of pancreatitis, the appearance of a much more marked deposit of crystals than was obtained from normal and most pathological specimens seemed at first sight to lend support to our theory, but as subsequent investigations showed that other mineral acids gave similar results, and it was found that the precipitate had not the characters of glycerosazone, it became necessary to look for some other explanation of the different be- haviour of the urine from pancreatic and non-pancreatic cases. Since comparative tests proved that cleaner and more satisfactory preparations could be obtained with hydro- chloric than with nitric or sulphuric acid, it was adopted for routine work and the original nitric acid method aban- Chemical Pathology 245 doned. The procedure we made use of for our clinical investigations has been described under the term "the A-reaction," and was carried out as follows: A specimen of the urine to be examined was carefully fil- tered, and 10 c.c. of the filtrate poured into a small flask. One cubic centimetre of strong hydrochloric acid (sp. gr. 1. 16) was added, and, a small funnel having been placed in the neck of the flask to act as a condenser, it was placed on a sand-bath and gently boiled for from five to ten minutes, after the first sign of ebullition was detected (Fig. 97). A mixture of 5 c.c. of the filtered urine and 5 c.c. of distilled water was then poured into the flask, which was afterwards well cooled in running water. The excess of acid was now neutralised by slowly adding 4 grams of lead carbonate, and, after standing for a few minutes to allow of the completion of the reac- tion, the flask was cooled in water to the lowest possible temperature, and the precipi- tate removed by careful filtra- tion through a well-moistened, close-grained filter-paper. The clear filtrate was then made up to 15 c.c, and added to 2 grams of powdered sodium acetate, 0.75 gram of phenylhydrazin hydrochlor- ate, and i c.c. of 50 per cent, acetic acid contained in a small flask fitted with a funnel-condenser. The mixture was gently boiled on a sand-bath for five minutes, poured into a test-tube, made up to 15 c.c, and allowed to cool undisturbed. After a period, varying with the severity Fig. 97. — "Pancreatic" re- action flasks fitted with fun- nel condensers on a sand-bath. 246 The Pancreas: Its Surgery and Pathology of the case, of from one to twenty-: Fig. 98. — "Pancreatic" reaction-crys- tals, prepared by the A-method; from a case of acute pancreatitis (X 192). Fig. 99. — "Pancreatic" reaction-crys- tals, prepared by the A-method; from a case of chronic pancreatitis due to the presence of a gall-stone in the pancreatic portion of the common bile-duct (X 192). four hours, a more or less abundant floc- culent yellow pre- cipitate, occupying a quarter or more of the test-tube, was seen to have formed, and this, when examined un- der the microscope with a -g-inch objec- tive, was found to consist of sheaves and rosettes of golden yellow crys- tals. The deposit met with in non- pancreatic urines was usually much less abundant, and had not, as a rule, the light fiocculent appearance of that obtained in speci- mens from well- marked cases of pancreatitis. Comparison of the crystals from differ- ent cases showed that they were not always of the same type, and experi- ments indicated that there was also a difference in their Chemical Pathology 247 Fig. 100. — "Pancreatic" reaction- crystals, prepared by the A-reaction; from a case of malignant disease of the pancreas (X 192). solubilities. The crystals fjom cases of acute and sub- acute pancreatitis were found to be fine, slender, and hair -like in form, and on being irri- gated with 7,7, per cent, sulphuric acid under the microscope were observed to dis- appear in under half a minute after the acid first touched them, taking the average of three or more determinations. In malignant disease, on the other hand, the more typical crystals were broad, coarse, and sword- like, and took from three to five minutes to dissolve in 33 per cent, sulphuric acid. Those from cases of chronic pancreatitis, and most non-pan- creatic diseases, were intermediate in form and rate of solution, dissolving in dilute sulphuric acid in from one to two min- utes after the acid first reached them. Exceptions were frequently met with, but careful observation of these Fig. I o I . — ' ' Pancreatic " reaction-de- posit from a normal urine treated by the A-method (X 192). 248 The Pancreas: Its Surgery and Pathology points was often found to be of considerable assistance in diagnosing the condition. The most striking excep- tions were encountered in cases of pancreatic cancer, for while there was, in some cases, a much more abundant deposit than usual, and this was found to consist of slender readily soluble crystals, in others a deposit of the more characteristic, coarse crystals was only secured after repeated trials with different periods of boiling. The results obtained by this method, although sug- gestive of some difference in the composition of the urine from pancreatic and non-pancreatic cases, and possibly also in different types of pancreatic disease, were not sufficiently distinctive to be entirely relied upon for pur- poses of diagnosis, and further investigation was obviously necessary. After considerable experiment we found that the formation of the crystals in the A-reaction was pre- vented, or interfered with, in inflammation of the pancreas by preliminary treatment of the urine with perchloride of mercury, while such treatment did not affect the ap- pearance of the crystals in cancer of the pancreas and other conditions that gave rise to a positive reaction. This differential test, which we described as the " B- reaction," was carried out as follows: Ten cubic centimetres of the filtered urine were tho- roughly mixed with 10 c.c. of a saturated solution of perchloride of mercury in distilled water. After standing for a few minutes the mixture was filtered through a well moistened filter-paper, and to 10 c.c. of the filtrate i c.c. of strong hydrochloric acid was added. It was then boiled on the sand-bath for five to ten minutes, and di- luted with 5 c.c. of the mixed urine and mercuric chloride solution with 10 c.c. of distilled water. After being cooled in running water the excess of acid was neutralised with lead carbonate and the succeeding stages of the operation carried out as in the "A-reaction." The indications given by a comparison of the results Chemical Pathology 249 yielded by these two reactions proved, in our hands, of very considerable assistance in diagnosis, and enable4 us to arrive at a correct opinion in several cases where the clinical signs and symptoms were doubtful or misleading. The results obtained in the examination of five hundred specimens of urine, two hundred and ninety-seven of which were from patients in whom there was evidence of pan- creatic disease at operation or post-mortem, and two hun- dred and eighty-three from normal individuals or patients suffering from diseases in which there was no reason to think that the pancreas was involved, are shown in the subjoined table: RESULTS OF THE "A AND B-PANCREATIC" REACTION IN FIVE HUNDRED CONSECUTIVE EXAMINATIONS. Group. Diagnosis. No. A Deposit Greater ThanB. A AND B Deposit Equal IN Amount. A and B Both Negative. I Acute pancreatitis 4 4 II Chronic pancreatitis: (a) With obstruction of common duct .... (6) No obstruction of common duct 68 116 49 98 19 15 3 Ill Cancer of the pancreas . 29 8 21 IV No pancreatitis: (a) Gall-stones in gall- bladder or common duct 56 117 7 9 31 132 18 (6) Miscellaneous 36 V Normal 50 II 39 From this it will be seen that, as a rule, the deposit yielded by the A-reaction was greater than that obtained by the B-method in those cases where there was evidence of pancreatitis, but that in those instances where the pan- creas was not inflamed there was no reaction, or the amount of precipitate was approximately the same in the 250 The Pancreas: Its Surgery and Pathology two preparations. In the majority of the cancer cases examined the result was similar to that obtained in non- pancreatic disease, but in about 25 per cent, there was an appreciable difference in the amount of deposit yielded by the A- and B-reactions. Like most comparative tests, to which accurate meas- urements cannot be applied, this method suffered from the great disadvantage of being dependent, to a certain extent, on the experience of the observer for its inter- pretation. Further, unless considerable care was exer- cised in the details of the experiment errors in technique were liable to occur and confuse the issue, as the published accounts of some who have attempted to use these reac- tions for diagnostic purposes have shown. To overcome these difficulties and, as far as possible, eliminate the per- sonal element, we have introduced an "improved " or "C- reaction," in which the presence or absence of pancreati- tis is indicated by the examination of a single preparation. The manipulation is slightly more complicated and still requires a reasonable amount of skill and care, particu- larly in the details of the experiment, but the result is an absolute one, and is therefore independent of the personal bias of the investigator. Examination of the phenylhydrazin precipitate de- rived from the urine in cases of pancreatic inflam- mation, after treatment with hydrochloric acid, showed that it consisted of two parts, one a phenylhydrazin compound of glycuronic acid and the other the osa- zone of a sugar. Although there is reason to believe that the excretion of glycuronic acid is increased in pancreatitis, an augmentation of the output occurs in so many pathological conditions that no helpful diagnos- tic method could be based upon this, in the present state of our knowledge. On turning to the precipitate ob- tained from the urine after treatment with mercuric chloride we found that it consisted entirely, or almost Chemical Pathology 251 entirely, of a glycuronic acid compound of phenylhydra- zin, so that the difference noticed between the A- and B- reactions in characteristic cases of pancreatic inflamma- tion appeared to be dependent upon the presence of the sugar. By collecting large quantities of urine from well-marked cases of pancreatitis, we are able to investi- gate the characters of this, and found that it gave the reactions of a pentose. As we have already said, we have been unable to discover any evidence of the presence of a pentose in the untreated urine from any of our cases, so that it was probable that the pentose giving rise to the characteristic "pancreatic" reaction was formed by hydrolysis from some antecedent substance in the urine during the process of heating it with the dilute acid. We are not as yet in a position to make any definite state- ments with regard to the nature of the mother-substance from which the sugar is derived, but our earlier experi- ments proved that it was not the so-called animal gum of the urine, and the fact that a positive reaction has not, so far, been obtained by the "improved method" with the urine from any but pancreatic cases suggests that it is probably a body resulting from changes in the pancreas, and possibly derived directly from that organ. The relatively large proportion of pentose-yielding material in the pancreas, as shown by Neuberg, who gives the following as the results of his analyses of various organs : 1 Pancreas Liver 0.56% Thymus 0.56% Submaxillary gland. • . 0.53% Thyroid 0.50% Kidney 0.49% Spleen 0.46% Brain 0.22% Muscle 0.11% As pentose in the dry ' substance points to the pancreas as the most likely source. It can- not be denied, however, that the disintegration of other tissues may also, at times, influence the urine in this 252 The Pancreas: Its Surgery and Pathology respect, and it has also to be remembered that the inges- tion of large amounts of penton-containing food materials may also cause small quantities of pentose to be excreted in the urine; therefore, while we maintain that a positive reaction by the ' ' improved method' ' of performing the so-called "pancreatic reaction" is strongly suggestive of inflammatory disease of the pancreas, we are not pre- pared to contend that it is pathognomonic of pancreatitis. The ' ' improved method, ' ' or " C-reaction, ' ' is based upon the different behaviour of glycuronic acid and the sugars, in acid solutions, to tribasic lead acetate, the former being precipitated and the latter remaining in solution. If therefore the acid filtrate, left after the urine has been boiled with hydrochloric acid and the excess of acid neutralised with lead carbonate, is treated with tribasic lead acetate, the glycuronic acid set free in the process will be thrown out, while any sugar remaining in the solution can be detected by the phenylhydrazin test, after the precipitate has been filtered off and the excess of lead removed by appropriate methods. In performing the reaction a specimen of the twenty-four hours urine, or of the mixed evening and morning secre- tions, is filtered several times through the same filter-paper. If it is found to be free from sugar and albumin, and is acid in reaction, 2 c.c. of strong hydrochloric acid (sp. gr. 1.16) are mixed with 40 c.c. of the clear filtrate, and the mix- ture gently boiled on a sand-bath in a small flask, fitted with a funnel condenser (Fig. 97). After ten minutes' boiling the flask is well cooled in a stream of water, and the contents made up to 40 c.c. with cold distilled water. The excess of acid is then neutralised by slowly adding 8 grams of lead carbonate. After standing for a few min- utes to allow of the completion of the reaction, the flask is again cooled in running water, and the contents filtered through a well-moistened, close-grained filter-paper until a perfectly clear filtrate is obtained. The acid filtrate is Chemical Pathology 253 then well shaken with 8 grams of powdered tribasic lead acetate, and the resulting precipitate removed by filtra- tion, as clear a filtrate as possible being secured by repeat- ing the filtrate several times if necessary. Since the large amount of lead now in solution would interfere with the subsequent steps of the experiment, it is removed, either by a stream of sulphuretted hydrogen, or, what we have found to be equally satisfactory and less disagreeable, by precipitating the lead as a sulphate. For this purpose the filtrate is well shaken with 4 grams of powdered sodium sulphate, the mixture heated to the boiling-point, then cooled to as low a temperature as possible in a stream of cold water, and the white precipitate removed by careful filtration. Ten cubic centimetres of the perfectly clear, transparent filtrate are taken and made up to 17 c.c. with distilled water ; it is then added to 0.8 gram of phenylhy- drazin hydrochl orate, 2 grams of sodium acetate, and i c.c. of 50 per cent, acetic acid, contained in a small flask fitted with a funnel condenser. The mixture is boiled on a sand-bath for ten minutes and filtered hot through a small filter-paper, moistened with hot water, into a test- tube provided with a 15 c.c. -mark. Should the filtrate fall short of 15 c.c, it is made up to that amount with hot distilled water, the added water being well mixed with the fluid by stirring with a glass rod, but in our own work we find that any addition is rarely necessary, as, with a little practice, it is possible to so regulate the boil- ing that the final result almost always comes out at between 15 and 16 c.c. In well-marked cases of pancreatic inflammation a light yellow, flocculent precipitate should appear in a few hours, but in less characteristic cases it may be necessary to leave the preparation over-night before a deposit occurs. Under the microscope the precip- itate is seen to consist of long, light -yellow, flexible, hair-like crystals arranged in delicate sheaves, which 254 The Pancreas: Its Surgery and Pathology when irrigated with 33 per cent, sulphuric acid melt away and disappear in ten to fifteen seconds after the acid first touches them. The preparation must always be examined microscopically, as a small deposit may be easily over- looked with the naked eye, and it is also difficult to deter- mine the exact nature of a slight precipitate by macro- scopical investigation alone. To exclude traces of sugar, undetected by the prelimi- Fig. 102. — Improved, or C-, "pancreatic" reaction-crystals from a case of chronic pancreatitis with gall-stones in the common bile-duct (X 200). nary reduction tests, a control experiment is carried out by treating 40 c.c. of the filtered urine in the same way as that in the test just described, except that it is not boiled with hydrochloric acid. Any albumin that may be present in the urine is removed, previous to commencing the test, by faintly acidulating, boiling, filtering off the albuminous precipitate, cooling, and making the specimen up to its orig- inal bulk with distilled water. The urine employed for Chemical Pathology •:)D the experiment should be fresh, and not have undergone fermentative changes. If alkaline in reaction, it should be made distinctly acid with hydrochloric acid before the test is commenced. Any dextrose that may be present can be removed by fermentation after the urine has been boiled with the acid and the excess neutralised. The administration of calcium chloride, as advised by one of us in all cases of pancreatic disease previous to operation, has been found to interfere with the success of the reaction. In the following table the results obtained by this method in two hundred consecutive examinations, in which it has been possible to confirm the diagnosis post- mortem or at operation, are given, and, for the sake of comparison, the findings in fifty specimens from presuma- bly healthy persons are also included : RESULT OF THE "IMPROVED" OR " C-PANCREATIC REAC- TION" IN TWO HUNDRED AND FIFTY CONSECUTIVE EXAMINATIONS. Group. Diagnosis. No. Positive. Negative. I Acute pancreatitis 2 2 II Chronic pancreatitis : (a) With obstruction of the common duct: (i) By gall-stones (2) By growth (6) No obstruction of common 19 2 19 2 duct : (i) No gall-stones found .. . 32 32 (2) Gall-stones in gall- bladder 12 12 Ill Cancer of the pancreas 16 4 1 2 IV No pancreatitis: (a) Gall-stones in common duct 10 10 (b) Gall-stones in gall-blad- der II 96 4 1 1 {c) Miscellaneous 92 V Normal 5° 50 256 The Pancreas: Its Surgery and Pathology It will be seen that a positive reaction was obtained in seventy-five, and that in one hundred and twenty-five no crystalline deposit was observed. Two of the former were cases of acute pancreatitis. In thirty-three there was chronic pancreatitis, associated with gall-stones in the common duct in nineteen, with growth of the common duct invading the pancreas in two, and with stones in the gall-bladder in twelve. In twelve, although the pancreas was stated to be distinctly larger and harder than normal when examined at operation, no biliary calculi were found, but in one of these cases gall-stones were found in a specimen of feeces examined at the same time as the urine, and in another several had been found in the stools shortly before the examination was made. There was an ulcer of the duodenum in six, a gastric ulcer adherent to the pancreas in one, in three there were nu- merous adhesions about the head of the gland. Of the sixteen cases of cancer of the pancreas, twelve gave no reaction, but in four a more or less marked deposit of crystals was obtained. In addition to the twelve cases of cancer just mentioned, no reaction was obtained in ten specimens from cases where gall-stones were found in the common duct at the time of operation and the pancreas was said to be normal ; in eleven where biliary calculi were present in the gall-bladder, but no evidence of pancrea- titis was found either pathologically or clinically ; and in ninety-two samples from cases of miscellaneous diseases, including cancer of the stomach, colon, rectum, or liver, gastric ulcer, duodenal ulcer, gastritis, colitis, appen- dicitis, tuberculosis of the intestine, intestinal obstruc- tion, cirrhosis of the liver, hepatic abscess, nephritis, floating kidney, tuberculosis of the kidney, cystitis, mumps, and Addison's disease. In four cases of cancer of the stomach or duodenum a positive reaction was ob- tained, but in these the growth was adherent to the pan- creas. One, in which the growth was situated in the Chemical Pathology 257 first part of the duodenum, gave no reaction when first examined, and on abdominal section it was then found that the pancreas was free; but a month later, when a positive reaction was obtained, a second exploratory operation showed that the pancreas had become involved in the growth. No reaction was obtained with any of the fifty specimens from apparently healthy individuals. On looking through the table it is interesting to note that nineteen out of the twenty-nine cases (or 65 per cent.) in which gall-stones were found in the common duct at the time of operation gave a positive reaction, whereas ten (or 35 per cent.) gave no reaction, which corresponds fairly closely with the 62 per cent, and 38 per cent, given by Helly as the proportion of cases in which the common bile-duct is embraced by, and free from, the pancreas respectively. The urines from twenty- two cases, which previous to operation had given a well-marked reaction, were re- examined one to two weeks after cholecystenterostomy had been performed for the relief of pancreatitis, but no reaction could be obtained, and in four cases where an opportunity presented itself of making a further investi- gation at a subsequent date, one three months, another six months, a third seven months, and the fourth nine months after operation, there was still no reaction. It has been possible to test the findings of the "pan- creatic" reaction in twenty-four cases by histological examination of the pancreas. In one a small piece of the gland was removed at operation and showed evidence of interstitial pancreatitis, which confirmed the diagnosis based upon the urine examined. The remaining twenty- three were examined post-mortem. Three were cases of cancer of the pancreas. In one, where the whole organ was invaded by the growth, the results of the examina- tion of the urine had suggested during life that there was a considerable chronic inflammation; in the other two 17 258 The Pancreas: Its Surgery and Pathology pig_ lo^, — Microphotographs of the pancreas from six cases the urine of which had given during life a well-marked "pancreatic" reac- tion (X ca 40). Chemical Pathology 259 the disease chiefly affected the head of the gland and a correct diagnosis had been arrived at. Eleven had been diagnosed from the urinary reaction as chronic pancrea- titis. In ten of these a more or less marked overgrowth of the interstitial connective tissue was found microscop- ically (Fig. 103). It is noteworthy that in four of them, although the organ had been said at operation to be larger and harder than usual, no pathological change could be detected by the naked eye after death. In one case no interstitial overgrowth could be discovered either macro- scopically or microscopically, but the blood-vessels were much dilated and there were small patches of round- celled infiltration in the neighbourhood of the ducts, pointing to an early inflammatory change. No reaction suggestive of a pancreatic lesion had been obtained in the remaining nine cases during life, and post-mortem the pancreas appeared to be normal, both to the naked eye and on microscopical examination. It has been contended by Ham and Cleland that the crystals obtained from the urine by the A- and B -reac- tions are crystals of a lead salt, formed by the action of phenylhydrazin and sodium acetate upon the lead car- bonate used in neutralising the hydrochloric acid em- ployed in the test, and that therefore the results obtained by these methods are absolutely unreliable. The un- soundness of this argument is at once apparent when it is remembered that sulphuric acid, neutralised with barium carbonate, or nitric acid neutralised with urea, give similar results, and that hydrochloric acid was only selected for routine work because it gave cleaner prepara- tions and its action was more easily controlled. Lead carbonate was selected as the neutralising agent, since the caustic alkalies were found to interfere with the reaction, and lead was the most insoluble chloride that could be formed. There are possibly some, however, who, in consequence 26o The Pancreas: Its Surgery and Pathology of this and similar criticism, are unconsciously biased against the reaction, and on this account it may not be out of place if we here summarise the arguments brought forward by these and other writers, and briefly consider the experimental data on which they were based. The conclusion reached by Ham and Cleland was arrived at on the following grounds: (i) They stated that they were able to obtain crystals from all urines, provided that the solution was sufficiently concentrated by boiling ; (2) rosettes of pale crystals were also obtained when the reaction was performed with distilled water; (3) if the urine or distilled water, after being boiled with hydro- chloric acid, and neutralised with lead carbonate, was treated with ammonium sulphide, filtered, and boiled, the reaction with phenylhydrazin was prevented; (4) lead acetate solution treated with sodium acetate and phenylhydrazin hydrochlorate gave fine needle-like crys- tals in rosettes and sheaves. There is no doubt that a soluble lead salt does form needle-like crystals when boiled with sodium acetate and phenylhydrazin hydrochlorate, but these crystals never appear in a properly performed "pancreatic" reaction, and, should they do so, can be easily distin- guished from the true osazone crystals by the naked- eye characters and their appearance under the micro- scope. They are very much larger (Fig. 104), colourless instead of yellow, and form solid masses at the bottom of the test-tube or appear as tufts adherent to its walls. In a carefully performed reaction the quantity of lead that passes into the filtrate is small, and does not in any way affect the results of the test, but if, from faults of manipulation or errors in technique, a large amount is present, it is not unlikely to give rise to difficul- ties in inexperienced hands, especially when the preparation has been unduly concentrated by excessive or furious boiling. It is to this cause that the positive reactions Chemical Pathology 261 obtained by Ham and Cleland with all urines and with distilled water were no doubt due. The removal of any lead in solution, moreover, does not interfere with the success of the reaction, as these writers state; indeed, this is intentionally done in the "improved reaction," as there the large amount of lead acetate employed would Fig. 104. — Microphotograph of the "lead-salt crystals" formed as the result of faulty technique in carrying out the "pancreatic" reaction by the original A-method (X 200). introduce a serious difficulty, unless it were removed before performing the phenylhydrazin test. Ammonium sulphide cannot, however, be used for the purpose, as the ammonia set free destroys any sugar that may be present when the fluid is subsequently heated, but Ham and Cleland do not appear to have considered that, when using ammonium sulphide in the manner they describe, they were possibly performing a modification of Moore's test for sugar. Finally, the fact that the purified crystals 262 The Pancreas: Its Surgery and Pathology from the urine in cases of pancreatic disease have a defi- nite melting-point, are free from any trace of lead, and correspond in their other characters with osazone crys- tals, conclusively proves that, when the reaction is care- fully and properly performed, they are not "lead salt crystals." In a paper on "the use of phenylhydrazin in the clinical examination of urine" W. H. Willcox gives an account of some experiments which were, more or less, based upon the reactions described in our original communications, and he there states that "the production of characteristic yellow crystals in the urine after hydrolysis with hydro- chloric acid can in no sense be used as a specific test for any pathological conditions, since such crystals are constantly obtained from normal urines." The experimental por- tions of this paper cover much of the ground traversed by us previous to the elaboration of the "pancreatic" reac- tion, but we cannot confirm all the results or agree with the deductions drawn from them. We are quite willing to admit that "the production of characteristic crystals in the urine after hydrolysis cannot be used as a specific test for any pathological condition," including pancrea- titis, if the method employed is that of the author of that paper, but we do maintain that, if the methods we have described are carefully and conscientiously carried out, and the results considered in conjunction with the clinical symptoms, and the indications given by other methods of pathological research, the presence of pan- creatitis can be determined with very much greater cer- tainty than by the clinical evidence alone. Our results with normal urines do not correspond with those obtained by Willcox, as the tables on pages 249 and 255 show. We do not think that any difficulty of diagnosis was likely to arise even with a normal urine when the original A- and B -reactions were employed, but the "improved method" has now got rid of any slight difficulty that there Chemical Pathology 263 might possibly be and has simplified the diagnosis of pancreatitis from non-pancreatic diseases. J. H. Schroeder and P. S. Haldane have published some criticisms of our original methods under the mistaken idea that we claimed to have proved the presence of glycerine in the urine of patients suffering from diseases of the pancreas. The glycerine theory was but a working hypothesis on which we commenced our investigations, and, as we have explained, had to be abandoned as the research proceeded. They also assume that the crystals obtained in acute and chronic pancreatitis and in cancer of the pancreas by the A-reaction are identical, and point out that they are therefore unlikely to have different shapes and solubilities. The assumption is without foundation, for the difference in appearance and in rate of solution in sulphuric acid is due to the differences of chemical composition. Fat-splitting Ferment {Opie). — Since the fat necrosis associated with acute pancreatitis is due to the fat- splitting ferment of the pancreatic juice, it occurred to Opie that this ferment, which is free in the tissues, might be excreted by the kidneys. In one case, in which he tested the truth of this assumption, he found evidence that tended to show that such an excretion did occur. The specimen of urine examined was taken, after death, from the body of a man who died of heemor- rhagic pancreatitis. It was neutralised with potassium hydrate and divided into two parts. To one a few drops of ethyl butyrate and a little neutral litmus were added. The other was treated in the same way, after any fer- ment present had been destroyed by boihng. Both specimens were incubated at 37° C. for twenty-four hours. At the end of that time the unboiled specimen had ac- quired a well-marked acid reaction, while the control specimen showed little, if any, change. We have not had the opportunity of carrying out a similar test in a case 264 The Pancreas: Its Surgery and Pathology of acute inflammation of the pancreas, but in two cases of subacute pancreatitis, and several of chronic pancrea- titis, in which we have employed it no difference could be observed between the two preparations. Lipuria has been described by some writers as occurring in disease of the pancreas. Clark and Bowditch have reported cases of cancer of the gland in which fat globules were present in the urine. Tulpius and Elliotson also record similar cases, but without any other confirmatory evidence of pancreatic disease than the presence of fat in the stools. In a case of acute pancreatitis, operated on seventy- two hours after the onset, Cooke found fat in the urine on one occasion. We have met with lipuria only once in our series of cases, and that was in a case of chronic pancreatitis in a woman, aged forty-four, the cause being apparently an extension from duodenal catarrh. It was associated with liporrhoea, azotorrhoea, and bulky stools, and with a well-marked pancreatic reaction in the urine. The abdomen was opened and a swelling of the pancreas was discovered with a number of adhesions surrounding it, but no gall-stones were found. Drainage of the bile-ducts by a simple cholecys- totomy completely cured the pancreatic condition, and when the urine was examined a year later there was an entire absence of the pancreatic reaction. Fat is met with in the urine in so many different con- ditions, and is so rarely found in pancreatic disease, that its association with lesions of the pancreas is possibly acci- dental. It is to be remembered, however, that lipuria occurs in diabetes mellitus, and that in this disease a large amount of fat is also occasionally encountered in the blood. Detection of Pancreatic Enzymes. — It is sometimes neces- sary to determine whether a pathological fluid, obtained from a cyst or fistula, has originated in connection with the pancreas. The general physical and chemical characters of the fluid may afford some indication of its probable source, Chemical Pathology 265 but the most reliable proof is obtained by an investiga- tion of its behaviour to proteids, fats, and carbohydrates. When the fluid is found to contain ferments capable of readily digesting all three forms of food material there can be little doubt as to its origin. The presence of a diastatic ferment alone is of little value in diagnosis, since diastase may be met with in other fluids of the body. The detec- tion of a proteolytic ferment, capable of digesting albu- min in an alkaline medium, is much more important, for no other ferment than trypsin can dissolve albumin in the presence of an alkali. It is frequently found, however, that the contents of cysts, undoubtedly of pan- creatic origin, have little or no proteolytic power. This is stated to be particularly the case with old encapsuled cysts of long standing. It has to be remembered, how- ever, that the normal pancreatic juice possesses but feeble powers of digesting proteid until it has been activated by the enterokinase of the intestine, so that the absence of this property in the contents of pancreatic cysts is not surprising. The observation of Delezenne that the addi- tion of a small quantity of a soluble calcium salt activates the pancreatic secretion as powerfully, but more slowly, than enterokinase suggests a way of overcoming the diffi- culty when examining the contents of a cyst for diagnos- tic purposes. The power of splitting neutral fats into glycerine and fatty acids is the most characteristic prop- erty of pancreatic juice, and unless a pathological fluid possesses this property it cannot be stated, with certainty, that it has originated in connection with the pancreas. The method of testing for the proteolytic ferment devel- oped by Boas is that which has been usually employed. The fluid to be examined is added to milk, placed for some time in an incubator, and, after the casein has been precipitated, is examined by the biuret reaction. A positive result of the test indicates that the fluid can digest albumin in the presence of an alkaline reaction. 266 The Pancreas: Its Surgery and Pathology Pawlow, in his investigations of the digestive juices, employed Mett's tubes, and these have the advantage that they afford a means by which the digestive power of the fluid can be expressed numerically for purposes of comparison. Egg-albumen is employed in this method and the tubes containing it are prepared as follows: The egg-albumen is filtered through gauze into a small beaker, or wide test-tube, and short glass tubes, having a lumen of about 2 mm., are slowly dropped into it. Air- bubbles are allowed to escape, aided by gentle tapping, and the vessel containing the tubes is then placed in a bath of boiling water for five or ten minutes. The flame is removed and the glass allowed to cool for several hours. The test-tube or beaker is then broken, and the small tubes filled with, and embedded in, the coagulated albu- men are cut out and preserved in glycerine. One of the tubes is used for each test. It is first washed with water, then placed in a test-tube containing the fluid to be tested, which, if necessary, has previously been made faintly alkaline. After being incubated for from three to ten hours, the small tube is examined and the presence of a proteolytic ferment is shown by a portion of the column of coagulated albumen having been dissolved. To de- termine the digestive power of the fluid, the length of the tube, and of the undigested remains of the proteid col- umn, are measured off on a millimetre scale with a low power of the microscope ; the difference gives the length of the digested cylinder in millimetres and fractions of a millimetre. The quantity of proteolytic ferment in the fluid is proportional to the square of the column of albu- men digested in a definite time. The fat-splitttng ferment may be tested for by the method of Castle and Loevenhart, in which purified ethyl buty- rate and neutral litmus are added to the fluid to be ex- amined (see "Opie's test" for fat-splitting ferments in the urine, page 263). A neutral fat, obtained by tho- Chemical Pathology 267 roughly shaking olive oil with sodium carbonate solution and ether, pipetting off the ethereal layer, filtering it if necessary, and then recovering the fat from the ether by allowing the latter to evaporate, may be used instead of ethyl butyrate. A numerical expression of the fat-splitting power of the fluid may be obtained by titrating the acidity of the emulsion, after it has been incubated for a definite period with periodical shaking, with baryta solution. The diastatic ferment can be recognized by incubating the fluid with boiled starch paste, and then testing for sugar with Fehling's solution. The activity of the fer- ment may be determined by titrating the sugar, formed in a given time from a definite amount of starch paste, with Fehling's solution, but the more rapid method advo- cated by Pawlow gives results that are sufficiently reliable. This is a modification of Mett's method for proteolytic ferments. Thin glass tubes filled w4th coloured starch paste are incubated with the fluid to be tested for a defi- nite time, — half an hour is usually sufficient, — and then examined with a lower power of the microscope. The activity of the amylolytic ferment is found to follow the same law as the proteolytic and vary directly as the square of the column digested in a given time. In all investigations on the digestive power of fluids it is advisable to conduct a control experiment with a portion of the material that has been boiled to destroy any ferments that may be present, so that in making the final deductions it may be used as a standard of compari- son. The action of bacteria should also be excluded by adding to both the test fluid and to the control a mild antiseptic, such as thymol. Literature Abelmann; Dissertation, 1890. V. Ackeron: Berliner klin. Wochenschr, 1889, S. 293. Baldi: Arch, di farmacol. e. terap., 1894, Nr. 10. Baldwin: Cited by Herter, "Lectures on Chem. Path.,'" 1898. Boas: Deutsche med. Wochenschr., 1890, xvi, 1095. Blumenthal: Berliner klin. Wochenschr., 1895, 1897. 268 The Pancreas: Its Surgery and Pathology Bowditch: Boston Med. and Surg. Journ., 1852, xxv, 8. Cammidge: "Arris and Gale Lect. "; Lancet, March 19, 1904; Tr. Royal Med. and Chir. Soc, 1905 ; Ibid., 1906; Brit. Med. Journ., Oct. 28, 1905, p. 1102; Ibid., May 19, 1906. Castle and Loevenhart: American Chem. Journal, i960, xxiv, No. 6. Cavazzanni: Arch, di clin. med., 1893. Clark: Lancet, 185 1, ii, 152. Cooke: Brit. Med. Journ., May 19, 1906, p. 1156. Demme; Wiener med. Blatter, 1884, Nr. 51. Deucher: Cited in Schmidt and Strasburger "Faeces des Menchen, " 1903, p. 161. Edsall: American Journ. of Med. Sci., April, 1901. EUiotson: Med. Chir. Trans., 1838, xviii, 67. Gerhardi: Virchow's Arch., 1886, S. 303, Dissert. Zurich, 1886. Gordon: Lancet, Dec. 9, 1905, p. 1687. Haldane, P. S.: Edinburgh Medical Journal, Nov., 1905. Ham and Cleland: Lancet, May 14, 21, June 11, 18, 1904; Brit. Med. Journ., June 14, 18, 25, July 7, 16, 1904. Hennige: Deutsch. Arch. f. klin. Med., xxiii, 285. Herter: "Lectures on Chemical Path.," 1898. Katz: Cited by Oser, in Nothnagel's "Encyclop. of Pract. Medicine," 1903. _ Krehl: "Clinical Pathology," tr. Hewlett, 1905. Kiilz and Vogel: Centralbl. f. med. Wissensch., 1898, S. 817; Zeitsch. f. Biologic, xxxii, 1895. L6pine: Berliner klin. Wochenschr., 1898, S. 317. Muller: Zeitschr. f. klin. Med., 1887, xii, 45. Neuberg: Rev. d. deutsch. chem. Gessellsch., 1902, Ixxxv, 147. Le Nobel: Arch. f. klin. Med., 1888, xliii, 285. Opie: Johns Hopkins Hosp. Bull., May, 1902, p. 117. Oser: Nothnagel's "Encyclop. of Pract. Med.," 1903. Pawlow: "The Work of the Digestive Glands," 1902, p. 25. Pisenti: Arch, per le sc. med., 1888, p. 87. de Renzi: Berliner klin. Wochenschr., 1892, Nr. 23. Robson, Mayo: "Hunterian Lectures," Lancet, March 19, 26, April 2, 1904. Rosenheim: Berliner klin. Wochenschr., 1889, S. 293. Rosenberg: Sitzungsberichte de physiol. Gesellsch. du Berlin, 1896, xxxiii, 6. Du Bois Arch., 1896, S. 535. Salkowski and Jastrowitz: Centralbl. f. d. med. Wiss., 1892, S. 593; Berliner klin. Wochenschr., 1895. Salmon: Verhandl. d. Cong. f. innere Med., 1902, xx, 244. Sandmeyer: Zeitschr. f. Biologic, 1891, xxix; 1892, xxxi, 86. Schlangenhaufer : Arch. f. Dermatol, u. Syph., 1895, xxxi, 43. Schmidt: " The Test Diet in Intestinal Diseases," tr. Aaron, 1906, p. 32. Schroeder: American Medicine, Sept. 3, 1904. Scott, J.: Brit. Med. Journ., Supplem., July 4, 1903. Sterling: Trans. Path. Soc. of London, liv, p. 255. Stefanani: Gazz. degli. osped., 1896, p. 848. Stolnikow: Virchow's Arch., 1882, xc, 389. Tulpius: Observat. Med., 1672, p. 328. Walker: Trans. Roy. Med. and Chir. Soc, 1889. Weintraud: Cited in Schmidt-Strasburger "Faeces des Menchen," 1903, S. 161. Willcox: Lancet, July 23, 1904, p. 211. Woolsey: Annals of Surgery, Nov., 1903. Young: Cited by Byrom Bramwell, Scot. Med. and Surg. Journ., April, 1904. Ziehl: Deutsch. med. Wochenschr., 1883, Nr. 37. CHAPTER XI DIABETES The dependence of diabetes upon disease of the pan- creas was first suggested by Bouchardat in 1875, and, although two years later Lancereaux emphasised the importance of the connection, and sought to distinguish a special type of the disease which he believed was always associated with profound alterations in the structure of the panceas, it was not until the classical experiments of von Mering and Minkowski were published in 1889 that the pancreatic theory of diabetes was placed upon a firmly established footing. The failure of experimental attempts to produce dia- betes through damming back the pancreatic juice by ligature of the ducts, or by conveying the secretion out- side the body by a fistula, had led such an authority as Cohnheim to regard all pancreatic changes met with in diabetics as secondary, or accidental, complications. It was recognised, however, that the pancreas might exert an influence on carbohydrate metabolism through some other channel than its external secretion, but complete extirpa- tion of the organ in dogs was so speedily followed by the death of the animals from shock, hemorrhage, injury to the duodenum, or peritonitis, that no opportunity was afforded of observing the remote effects of the operation. Martinotti, in 1888, was the first to successfully over- come these difficulties and to point out that the failure of previous experimenters might be avoided by rigid antisepsis, careful ligature of all bleeding vessels, and preservation of the vascular supply of the duodenum. It was not until the following year, however, that the 269 270 The Pancreas: Its Surgery and Pathology physiological relation of the pancreas to carbohydrate metabolism was conclusively proved by the publication of von Mering and Minkowski's work. These and other observers have shown that total extir- pation of the pancreas in dogs gives rise to a condition com- parable in every respect to diabetes as seen in man. Usually within twenty-four hours of the operation the urine con- tains sugar, which gradually increases in amount until it reaches a maximum of 8 to 10 per cent, about the third day. On a diet of bread and meat a dog of 8 kilos is then found to pass from 70 to 80 grams of sugar in the twenty-four hours, and even after seven days' starvation glycosuria is still present. The amount of sugar in the urine in the latter case bears a constant ratio to the nitrogen of about 2.8 : I, suggesting that the total quantity of sugar formed from albumin within the body is being excreted unutil- ised by the tissues. Although dextrose given with the food is passed in its entirety into the urine, Isevulose is made use of to a fairly large extent, in contrast to what is found to be the case in diseases of the liver (Strauss). There is a marked increase in the quantity of urine passed, a dog of 7 kilos voiding from 1000 to 1200 c.c. in the twenty-four hours. Although an animal from which the pancreas has been removed eats and drinks voraciously, it rapidly wastes and loses strength, so that death takes place from inanition in about four weeks, even when lung disease, or trouble arising from the invariable disin- clination of the operation wound to heal, does not bring about a fatal issue at an earlier date. When the animal is too weak to move about, the excretion of sugar begins to diminish, although food is being taken, and a few days before death it may altogether disappear, especially when there is suppurative peritonitis. Coincidentally with the fall in the excretion of sugar, acetone, diacetic acid, and /3-oxybutyric acid make their appearance in the urine. When the animal is killed a few days after Diabetes 271 the operation, the glycogen normally present in the liver and other organs is found to be absent, or only present in small amounts, unless it has been fed with laevulose in the interval, when a high percentage may be met with. Examination of the blood shows that it contains a jjro- portion of sugar much in excess of the normal o. i per cent., sometimes as much as 0.4 per cent, being found, and that when the ureters are tied or the kidneys are removed the proportion is still further increased, thus pointing to the accumulation of sugar in the blood as the immediate cause of the glycosuria. The effects of removing the pancreas have been most thoroughly and completely investigated in dogs, but analogous results have also been obtained with many other members of the vetebrate series, including cats and pigs (Minkowski and Harley), carnivorous birds (Wein- traud, Kausel, and Langendorff ) , frogs and turtles (Alde- hoff and Markuse), eels (Capparelli). The proportion of sugar in the blood has been shown, by Kausel, to be in- creased in herbivorous birds by removing the pancreas, but glycosuria was found to only occasionally occur, probably because the kidneys of these animals are not readily pervious to sugar. Experiments performed upon rabbits have usually been unsuccessful, because of the great technical difficulties encountered in totally extirpat- ing the gland, but Hedon, and later Sauerbeck, have succeeded in producing atrophy and transient glyco- suria by injecting oil into the duct of Wirsung, the gly- cosuria appearing at the earliest on the twentieth day, and being at its height from the thirtieth to the thirty- eighth day after the injection. Partial extirpation of the pancreas may or may not give rise to diabetes, according to the amount left behind and its condition. If about a fourth or fifth of the gland is left, glycosuria only occurs if carbohydrates are present in the food (" alimentary glycosuria"). A larger por- 272 The Pancreas: Its Surgery and Pathology tion usually prevents the condition. Less generally gives rise to frank diabetes. Even when sugar does not appear in the urine after partial extirpation it will do so if the remnant is subsequently removed, and may gradually develop as the fragment atrophies. Sandmeyer found that the first trace of sugar appeared in the urine of a dog, part of whose pancreas he had removed, seven weeks after the operation, and it was not until after the lapse of thirteen and a half months that permanent diabetes developed. Death occurred eight months later. At the post-mortem a remnant weighing 0.36 gram, and showing no trace of gland structure, was found adherent to the posterior wall of the stomach, while attached to the lowest part of the duodenum was a piece of slightly changed gland-tissue the size of a pea. The first explanation of the results of these experiments that suggests itself is that the removal of the pancreas leads to impaired digestion from absence of the pancreatic juice, and that this is in some way responsible for the onset of glycosuria. But the fact that diabetic symptoms do not supervene unless almost the entire gland has been removed is against such a theory; moreover, if the se- cretion of the gland is diverted, and intestinal digestion thus prevented, diabetes does not follow, although marked wasting may occur. Ligature of the pancreatic duct likewise fails to give rise to glycosuria, as a rule. Disease of the solar plexus has been regarded by some as a cause of the diabetes, and, as the plexus is almost unavoid- ably injured in the removal of the pancreas, this might possibly be the explanation of the symptoms caused by the depancreatisation of animals. It was shown by Minkowski, however, that if the descending portion of the gland is transplanted into the subcutaneous tissue of the abdom- inal wall, and allowed to become engrafted there, the intra-abdominal portion can be removed, after the graft has been severed from all its nervous connections, without Diabetes 273 producing diabetes, but that if the graft is subsequently removed — or atrophies — diabetes develops. The cause of the glycosuria, and of the accumulation of sugar in the blood, appears therefore to be dependent upon some influence which the pancreas exerts by way of the blood or lymph stream. There are two possible means by which this can be effected: first, it may be that the cells of the pancreas normally destroy, or modify, some toxic substance, produced in other parts of the body, which interferes with the utilisation of sugar by the tissues; or, secondly, that the pancreas produces an internal secretion which is necessary for the splitting up and use of sugar by the other cells of the organism. The first, or auto-intoxication, theory is that which was originally favoured by Minkowski, but it was later aban- doned by him in favour of the second hypothesis. Bosan- quet,in his Goulstonian lectures, favours the view that dia- betes, is due to an increased internal dissociation of tissue (possibly fat) into sugar, caused by a toxic substance that is produced in the course of normal metabolism, and which is normally neutralised by the pancreas. He points out that poisonous doses of phloridzin, diuretin, and uranium nitrate give rise to glycosuria, and that suprarenal extract and other reducing substances, when applied directly to the pancreas, produce a similar effect. That it is not necessary for the bulk of the blood to come into actual contact with the pancreatic cells to prevent hypergly- csemia and glycosuria is suggested by the restraining effect of even a small proportion of the pancreatic tissue, and by the results of grafting a portion of the gland into the abdominal wall, for under such conditions only a small fraction can be directly influenced by the gland cells. It is possible, however, that when only a small part of the pancreas is left in an experimental partial extirpation of the gland, a sufficient amount of an internal secretion may pass into the bloodstream to neutralise any toxic substance 274 The Pancreas: Its Surgery and Pathology that may be present there. Tuckett has suggested that the pancreas normally forms such an internal secretion, which enters the circulation by way of the thoracic duct, and there neutralises a toxine absorbed by the lymphatics from the intestine during digestion. In support of his hypothesis he states that if the thoracic lymph from a fasting dog is injected into the portal circulation of a cat, no hyperglycasmia or glycosuria results; but that if the lymph from a dog during digestion is similarly injected, a hyperglycsemia, varying from 0.3 to 0.9 per cent., and a glycosuria, varying from i.o to 9.0 per cent., are pro- duced. Confirmation of his results is, however, as yet lacking. Minkowski has shown that if the efferent vessels from a pancreatic graft are ligatured, so as to ensure that all the returning blood is passed into the general circula- tion, diabetes does not develop, thus demonstrating that the transmission of the efferent blood into the portal circulation is not necessary to prevent the onset of glyco- suria. As the result of experiments upon dogs Lorand has recently stated that there is a relation between the islands of Langerhans and the thyroid, the former secret- ing a substance which neutralises a poison produced by the latter, and that diabetes may arise either from increased functional activity of the thyroid or from failure of the cell islets to perform their function. The theory that the blood normally contains a sugar- splitting ferment, which is absent in diabetes, was warmly advocated by Lepine, who stated that the blood of dia- betics has a diminished capacity for transforming sugar. ■Crofton, who supported Lepine's observations, claimed to have isolated the ferment by which glycolysis is brought about, and to have identified it with trypsin. Other observers have found, however, that when precautions are taken to prevent contamination, normal blood possesses no glycolytic power, and have regarded the positive results as being due to the action of micro-organisms. Diabetes 275 Indeed, Lepine himself subsequently gave up the idea that glycolysis occurs in the blood, and referred it to the tissue cells. Blumenthal and others have asserted that the cells from the pancreas, liver, spleen, muscle, etc., possess a strong glycolytic power; which is much increased if pancreatic extract be mixed with the cell juice from other organs. Giacco found that this power varied in different organs, being great in the heart and little in the pancreas itself. He also stated that it was not a vital phenomenon, for it persisted after the tissues had been boiled. According to Umber, however, when careful precautions are taken against contamination with micro- organisms, the tissues outside the body exhibit only very slight glycolytic powers. The same observer also found that the sugar-splitting power of the blood was not greater in the pancreatic vein than in the general arterial or venous systems, as it would have been if the pancreas secreted a sugar-destroying substance. An explanation of the phenomena of glycolysis on the lines of Ehrlich's "side-chain" theory has been advanced by Cohnheim, for he found that expressed muscle juice is inactive to sugar until it has been mixed with expressed tissue juice from the pancreas, or with an ether precipitate from it. From this he argues that the muscle produces a ferment which is itself incapable of decomposing sugar, but which, when acted on by an " activator substance" derived from the pancreas, gains that power, in much the same way as a complement and amboceptor are necessary for hsemo- lysis and analogous processes. The presence of blood in the muscle was found to cause glycolysis without the addi- tion of pancreatic extract, indicating that the activator substance derived from the pancreas was present in the blood. This substance Cohnheim states is soluble in water and alcohol, but is insoluble in ether, and, since it is not destroyed by boiling, he concludes that it is not a 276 The Pancreas: Its Surgery and Pathology ferment, but analogous to adrenalin, iodothyrin, secretin, and other products of internal secretion. Cohnheim's methods and conclusions have been criticised by Claus and Embden, who were unable to confirm his results, but the main fact that the pancreas gives rise to an activator substance for a glycolytic enzyme produced by other tissues of the body appears to have since been firmly es- tablished. Experiments have been conducted by Pavy which point to a "co-ferment-like, or activator sub- stance," being yielded by the pancreas, which aids in the synthetic process concerned in the linking-on of sugar in the construction of proteid, and the absence of which would lead to failure of carbohydrate assimilation and the condition met with in diabetes. There is little or no doubt that the nervous system plays a part in the production of some forms of diabetes, and, although the experiments already detailed show that injury of the nerves in the neighbourhood of the pancreas is not responsible for the symptoms caused by extirpa- tion of the gland, it is possible that indirectly the nutri- tion of the nerve centres may be influenced in such a way as to produce metabolic changes in the pancreas and other tissues of the body. There is, however, no experi- mental work to support such a view, and it is now gener- ally admitted that the most satisfactory explanation of pancreatic diabetes is that which supposes that the disease is due to the absence of some ferment, or co-ferment-like body, which normally reaches the blood from the gland. Injury or rough handling of the pancreas, painting it with piperidine, nicotine, coniine, pyridine, adrenalin, etc., are said to give rise to transitory glycosuria. This subject has been recently re-investigated by Underbill, who found that " insults" of the gland, by freezing with ethyl chloride or rough handling, caused neither hyperglycasmia nor glycosuria, but that the application of the drugs referred to gave rise to both. Further investigation showed, Diabetes 277 however, that suppression of the respiratory process, to the point of dyspnoea, caused an increase of sugar in the blood which could be prevented by the administration of oxygen, and that a similar use of oxygen also prevented the hyperglycasmia generally following the application of piperidine and all the other substances mentioned, except adrenalin, so that they appeared to act through the blood upon the respiratory centre. The action of adrenalin appeared, however, to be directly upon the pancreas, and was unique in this respect. The view has been held by some authors that diabetes is an infectious disorder, and since, as we have already seen, changes may be brought about in the structure of the pancreas by the action of micro-organisms and their toxines, it is not impossible that, at least in some instances, a microbic infection may give rise to the disease. Ham- marschlag and KaufEmann have, indeed, succeeded in producing glycosuria by feeding animals upon bacteria obtained from the intestines of diabetics, and by injecting them intravenously, but evidence that it was a true dia- betes is lacking, and there is no proof that the infection was specific. The first recorded case in which disease of the pan- creas was noticed to be associated with diabetes was described by Cowley in 1 788. In this the gland was found to be atrophied and to contain calculi in its ducts. Cho- part described a similar case in 1821, and Bright, in 1833, gave an account of a diabetic, nineteen years of age, with jaundice and fatty stools, the head of whose pancreas was found at the post-mortem to be converted into a hard, nodtdar tumour, firmly adherent to the duodenum. Sub- sequently other cases were published by Elliotson, Fre- richs, Hartsen, Fles, von Recklinghausen, Munk, Silver, and Bouchardat, but the first to definitely propound the theory of pancreatic diabetes was Lancereaux, in 1877. He claimed that diabetes, accompanied by wasting 278 The Pancreas: Its Surgery and Pathology ("diabete maigre"), was due to disease of the pancreas, while that in which there was no marked loss of flesh ("diabete gras") arose from some other cause. The pancreatic type of the disease, he believed, was also characterised by the bnisqueness of its onset, the gravity of the symptoms, and the rapid progress of the disease. Subsequent observation has not confirmed the clinical distinctions thus drawn by Lancereaux; in fact, many cases of glycosuria, of undoubted pancreatic origin, are of insidious onset, progress but slowly, and show no marked loss of flesh. We have had the opportunity of observing several cases in which glycosuria has developed after operations for gall-stones, or disease of the pancreas, and have been much struck by the slow progress of the disease and generally good condition of the patient, even after the lapse of several years. One patient who was operated on twelve years ago for gall-stones was found to have glycosuria eight and a half years after the operation, and still passes a considerable amount of sugar in the urine, but enjoys good general health. In another and similar case the glycosuria has persisted for six years, and in a third case of cholelithiasis associated with glycosuria, in which we have repeatedly had the opportunity of exam- ining the urine, the patient was said to be alive and well, except for some local irritation due to the sugar, five years after the operation and the discovery of the glyco- suria. It it now generally acknowledged that the character of the symptoms affords no clue as to the pancreatic or non-pancreatic origin of the disease, and that attempts to make a clinical distinction are usually unsuccessful, except in those instances where a history of past disease of the pancreas affords an indication. It is impossible, therefore, to judge from bedside experience how far pancreatic disease is responsible for diabetes in man. On turning to post-mortem records for information on this Diabetes 279 point, we are at once confronted by the divergent ex- perience of different observers. Windle reported that in one hundred and thirty-nine cases of diabetes the pan- creas was diseased in seventy-four (53 per cent.). Seegen analysed the records of ninety-two cases and found a pan- creatic lesion in seventeen (19 per cent.). P>erichs in forty-four cases found disease of the pancreas in sixteen (36 per cent.). Out of fifty-four cases of diabetes, ex- amined in the Berlin Pathological Institute, Hansemann reports that there was a lesion of the pancreas in forty (17 per cent.). Bloch, quoted by Oser, collected twenty- two cases, from the records of the Vienna General Hos- pital, in twelve (55 per cent.) of which the pancreas had been recognised as abnormal. Williamson examined twenty-three cases and in fifteen (65 per cent.) found evidence of pancreatic disease. Opie investigated the pancreas in nineteen cases of diabetes and detected some abnormality of the gland in fifteen (79 per cent). Bosan- quet records nineteen cases, in seventeen (90 per cent.) of which there was disease of the pancreas. These different results are no doubt to a great extent de- pendent upon a divergence of opinion as to what may be regarded as normal and what as pathological when examin- ing the pancreas in the post-mortem room, and also to the use of the microscope by modern observers as an aid to diagnosis in some instances. Opie, who made a histological examination of the gland in the nineteen cases he investi- gated, found evidence of disease in all but four (21 per cent.), whereas Seegen, out of his ninety- two cases, states that the organ was normal in seventy-five (81 per cent.) ; in four of Opie's cases, however, no gross abnormality could be detected, and it was not until they were sub- mitted to microscopical investigation that a lesion was discovered. The statistics of the older observers, not based upon careful macroscopical and microscopical examination, are therefore probably of little value, and, 28o The Pancreas: Its Surgery and Pathology as the number of instances in which such investigations have been performed is as yet too small to allow of any reliable inference being drawn, no definite answer can be given to the question as to what proportion of cases of diabetes are due to disease of the pancreas? Opie con- siders that in considerably more than half the disease results from a destructive lesion of the gland, while Bosanquet thinks that "it is becoming increasingly probable that the pancreas is diseased in all cases of diabetes mellitus." On attempting to determine from published reports the relative frequency in diabetes of the various diseases to which the pancreas is liable, the difficulty of reconciling the results obtained by different investigators is again encountered. According to the older observers, the most common lesion is atrophy of the gland. Windle found it in over 59 per cent, of the cases he investigated and Fre- richs in 75 per cent. The statistics, quoted by Hanse- mann, from the Berlin hospitals in the space of ten years, show forty cases of diabetes with disease of the pancreas, in thirty-six (90 per cent.) of which there was simple atrophy, and in three (8 per cent.) atrophy and sclerosis. The more recent observations of Williamson and Opie, however, give much lower figures, the former finding simple atrophy in four out of eleven cases (27 percent.), and the latter in four out of fifteen (26 per cent.). Some explanation of this difference is afforded by the more careful and exact methods of investigation employed in recent years, and there is no doubt that in the past too great a reliance upon the naked-eye characters caused many cases in which the size of the pancreas appeared diminished to be classified as simple atrophy, which were, in reality, examples of the atrophic changes resulting from chronic inflammation of the gland. The form of atrophy which Hansemann considered was always asso- ciated with diabetes appears to belong to this class, and Diabetes 281 although there is no means of determining whether the atrophy of the pancreatic cells leads to the increase of fibrous tissue, or the fibrosis results from inflammatory changes in the gland, and then, by its contraction, pro- duces alterations in the glandular acini, it is now gener- ally considered that the fibrosis is the principal lesion, and the cell atrophy a minor phenomenon. Simple atrophy, therefore, although it may be the only lesion found in some cases of diabetes, would not appear to play such an important part in the production of the disease as was at one time supposed, and there is no evidence to show that the pancreas is liable to a particular form of atrophy which invariably gives rise to diabetes. In a few cases of diabetes fatty degeneration of the pancreas has been found, after death, as the only discover- able lesion. Bosanquet met with a recognisable degree of fatty change in ten out of one hundred cases, which, in three, was combined with some fibrosis. Williamson in his series met with one case of lipomatosis, in which there was atrophy and fatty degeneration, and one where, besides atrophy and fatty degeneration, there were evi- dences of inflammatory changes. The earliest recorded case in which disease of the pan- creas was recognised as being associated with diabetes was, as we have seen, one of pancreatic calculi. But as Hansemann was only able to find fourteen instances in seventy-two cases (19 per cent.) collected from medical literature, and Oser quotes but twenty-four examples in one hundred and eighty-eight cases of diabetes (14 per cent.), the association is not a very common one, particu- larly as the lesion is so obvious that it would not be readily overlooked. The mere presence of calculi cannot be regarded as directly responsible for the diabetes, for blocking of the ducts, by ligature or otherwise, has been proved not to cause glycosuria ; it is to the fibrotic changes accompanying them that we must look for the explanation. 282 The Pancreas: Its Surgery and Pathology That this is the true cause is shown by the fact that dia- betes is only found in those cases where there is very marked overgrowth of fibrous tissue, whereas in those instances where the concretions are not associated with advanced interstitial changes glycosuria does not occur. In a similar way, although cysts of the pancreas have been found in from 5 per cent. (Oser) to 7 per cent. (Dieck- hoff) of diabetics showing a pancreatic lesion, there are many cases of cysts in which glycosuria does not occur. In some instances sugar may appear in the urine some time after a cyst has been recog- nised and surgically treated, owing probably to the advance of the chronic inflamma- tory changes to which the forma- tion of the cyst was originally due. We have had the opportunity of in- vestigating a case of this description through the kindness of Dr. Chur- ton, under whose care it came at the General In- firmary, Leeds. The patient was operated on by one of us in June, 1896, for a cyst of the pancreas. The urine was then, free from sugar and showed no other abnormality, save that it gave a well-marked "pancre- atic" reaction. In February, 1905, we heard that the patient had been admitted to the Infirmary suffering from diabetes, and, by the courtesy of the house physician, we were able to obtain a twenty-four hour sample of the Fig. 105. — Fibrosis of the pancreas, from a case of diabetes associated with the pres- ence of pancreatic calculi (X 32). Diabetes 283 urine and a specimen of the faeces. The former measured 62 ounces, was strongly acid in reaction, specific gravity 1 .030. There was no albumin, but a well-marked reaction for nucleo-proteid was obtained. Acetone was absent, but there was a trace of diacetic acid. No reaction for bile-pigment, urobilin, or indican was obtained. The urine reduced Fehling's solution and gave a characteristic reaction with phenylhydrazin. Titration with Fehling's solution showed 4.5 per cent, of sugar (80 grams in the twenty-four hours). No indication of the pres- ence of a pentose could be found. The total ni- trogen, urea, uric acid, chlorides, phosphates, sulphates, and oxalates were estimated, and found to be normal, ex- cept that the oxalates showed an excess (0.32 gram in the twenty-four hours). The "pancre- atic " reaction gave many fine crystals, sol- uble in ^^ per cent, sulphuric acid in ten to fifteen seconds. The fasces were of a light yellow coloiu" and faintly alkaline in reaction. They gave a well-marked reaction for stercobilin. There was no marked excess of unabsorbed fat, but the normal relation between the "neutral fats" and "fatty acids" was disturbed, the for- mer constituting 15 per cent, and the latter only 5 per cent, of the dry weight of the faeces, thus indicating some interference with the digestive functions of the pancreas. The association of cancer of the pancreas with diabetes is relatively uncommon. Windle found it in 4 per cent. Fig. 106. — Columnar-celled car- cinoma of the pancreas undergoing colloid change, from a case of diabetes (X 5°)- 284 The Pancreas: Its Surgery and Pathology of his cases; Frerichs in 6 per cent., Dieckhoff in 7 per cent., and Williamson once in his series of twenty-three consecutive cases. Glycosuria has been met with in only two of the forty cases of primary malignant disease of the pancreas in which we have had the opportunity of examining the urine, and once where the gland was in- volved in a secondary growth. The last is of particular interest from several points of view, for it demonstrates in a very striking manner the importance of the pancreas in carbohydrate metabolism in the human subject, and also the value of the "pancreatic" reaction in diagnosis. When the patient was first seen, early in December, 1905, there was an abdominal tumour which it was thought might be pancreatic, but a specimen of urine on being submitted for examination gave no " pancreatic" reaction and was free from sugar. On opening the abdomen the tumour was found to be due to a growth in the first part of the duodenum, and a gastro-enterostomy was therefore performed. On the i8th of January a second specimen of urine was examined, and, although it was still free from sugar, it was found to give a well-marked and charac- teristic "pancreatic" reaction, suggesting that the pan- creas had now become involved in the growth. At the request of the friends, the abdomen was again opened, and it was seen that the growth had invaded the pancreas and was beginning to involve the common bile-duct. As it was impossible to attempt the removal of the tumour, a cholecystenterostomy was performed. In May, 1906, the urine was again examined, and was found to contain 5.25 per cent, of sugar; a month later this had increased to 7.0 per cent.; in July it had reached 7.25 per cent. ; in August, 7.5 per cent.; and in October, 9.5 per cent. was present. In spite of the high percentage of sugar in the urine the general condition of the patient remained fairly good and she complained of no other symptoms than thirst and a voracious appetite. Considerable Diabetes 285 quantities of acetone and diacetic acid were present in the urine in May, but with careful treatment they gradually diminished in amount, until in the early part of October only traces could be detected. Toward the end of Octo- ber the gall-bladder was found to be distended and a few days later jaundice developed. The patient died deeply jaundiced on November 5, 1906. In some cases of malignant disease of the pancreas glycosuria has appeared as an early symptom, which has later disappeared, while in others it has only been met with towards the termination of the disease. The tem- porary appearance of sugar in the urine in these cases is possibly dependent upon the disturbance in the functions of the gland, caused by an inflammatory reaction attendant upon the spread of the growth, which subsequently quiets down, leaving sufficient unaltered tissue to carry on the work of carbohydrate metabolism. In a case of this description under Macaigni, quoted by Oser, there w^as transient glycosuria for seven months, then eleven months' cachexia without glycosuria. Death took place twenty- three months after the onset of the disease. Post-mortem a large, very hard cancer of the head of the pancreas, replacing one -half of the gland, was found. The rest of the organ appeared to be normal. It has also to be borne in mind that where a portion of the pancreas has been destroyed by growth the condition resembles that pro- duced in animals by partial extirpation of the gland, so that if carbohydrates are excluded from the diet the ali- mentary glycosuria, which previously existed, may dissap- pear. In most recorded cases where sugar has appeared in the urine as a terminal symptom either the whole organ has been replaced by a mass of growth, or the portions that have remained have undergone sclerotic changes, so that no normal pancreatic tissue was left to carry on the functions of the gland. The absence of permanent diabetes in most cases of 286 The Pancreas: Its Surgery and Pathology cancer of the pancreas is due to the growth being Hmited in many instances to one portion of the gland, usually the head. In about 29 per cent, of cases, however, this explanation will not hold good, for in that proportion there is a diffuse growth affecting the whole organ. It is supposed that in these cases either the tumour cells possess a glycolytic power, or the new-growth insinuates itself between the pancreatic cells in such a way as to obliterate the normal structure of the gland without de- stroying it entirely. That such a process of growth is possible is shown by the presence in some instances of unaltered island of Langerhans in the midst of the can- cerous material, while in support of the former hypothesis Hansemann points out that in primary carcinoma of the suprarenals Addison's disease is rare. When considering the general pathology of the pancreas, we pointed out that the commonest of all lesions to which the organ is liable are those of an inflammatory nature, although until recently they have failed to receive, both from clinicians and pathologists, that recognition which their importance deserves. The association of diabetes with inflammatory changes, and their sequelcB, have in a similar way been largely overlooked, or the disease has been referred to some other cause. As we have seen, the special form of atrophy described by Hansemann is in reality a fibrosis due to inflammatory changes in the gland ; calculi and cysts are also probably not responsible for the glycosuria with which they are associated, but occur in the course of a chronic inflammation which ultimately destroys the structure of the gland, and some cases, at least, of diabetes associated with malignant disease of the pancreas are caused by the changes brought about by secondary inflammation. Dieckhoff in his analysis of fifty-three cases found acute pancreatitis in 10 per cent, and chronic pancreatitis in 36 per cent. Williamson met with four instances of cirrhosis of the pancreas in twenty-three Diabetes 287 cases, and Opie with four of chronic inflammation in nineteen cases, so that it is probable that inflammatory changes play a not unimportant part in the production of the disease, especially if the different manifestations to which reference has been made are taken into account. Acute pancreatitis is not itself a common disease, and is for this reason alone not frequently met with as a cause of diabetes. In the one hundred and eighty-eight cases collected by Oser there were three in which diabetes was associated with hemorrhage into the pancreas, three of necrosis of the gland, and six of abscess; yet in about one hundred cases of acute inflammation, collected by Fitz and by Seitz, diabetes was only present in two. The reason for the comparative rarity with which glyco- suria occurs in acute pancreatitis appears to be that when the whole organ is destroyed death usually follows very rapidly, and when the progress of the disease is less acute portions of the gland are left unaffected. The experi- ments of Guleke on dogs have shown that, when complete necrosis of the pancreas has been induced, by injecting oil into the ligatured pancreatic duct, glycosuria always occurs, but that when a portion of the pancreas has been left intact no sugar can be found in the urine. This observer also found that in animals where chronic pan- creatitis had been produced by the same means glycosuria was present in some and not in others. A case of haemorrhage into the pancreas, causing destruc- tion of the whole gland, and associated with the appear- ance of sugar in the urine, is described by Bosanquet in his Goulstonian lectures. The patient, a laundress aged fifty- three, was admitted into Charing Cross Hospital, under the care of Dr. J. M. Bruce, on January 24, 1893. A week be- fore she had been seized with pain in the abdomen, which rapidly swelled and became hard to the touch. She had previously had no symptoms of diabetes, but then com- plained of thirst, and on examining the urine it was found 288 The Pancreas: Its Surgery and Pathology to contain from 10.12 to 11.25 grains of sugar in the twenty-four hours. Her temperature, which on the 24th was 100° P., gradually rose, and on the 29th and 30th she had rigors. On the latter day acetone was present in the urine. Finally she died in collapse without any appear- ance of coma. At the necropsy the layers of the mesen- tery were everywhere separated by a large mass of disin- tegrated blood-clot and blood-stained fluid. The stomach was adherent to the left lobe of the liver, and, on separat- ing the two, several pints of blood-stained grumous fluid escaped. In the situation of the pancreas was a break- ing-down mass of tissue, along with much bloody fluid. There was also diffuse fat necrosis and evidence of recent peritonitis. Such a case constitutes a natural experi- ment on the removal of the pancreas in a human being, and, as Bosanquet points out, the results exactly corre- spond to those obtained in animals. An example of the association of glycosuria with acute pancreatitis in which recovery took place has been re- corded by Gifford Nash, in the "Lancet" of November II, 1902. The patient was a man of sixty, who for seven years had suffered from "bilious attacks" and dis- comfort at the pit of the stomach. On October 27, 1901, he was seized with sudden pain in the abdomen. There was no jaundice, and the symptoms suggested intestinal obstruction. The urine was increased in amount, and contained, on November 5th, 8.75 grains of sugar to the ounce. Operation was undertaken on November 17th. The pancreas was found to be enlarged, there was fat necrosis in the neighbourhood of the gland, and a large calculus was found in the gall-bladder. Cholecystotomy was performed and the patient slowly recovered. On December 28th the urine contained 4.5 grains of sugar per ounce; on March ist, 1902, 4.5 grains also, but on May 17th the glycosuria had disappeared. Through the kindness of Dr. Gifford Nash and Dr. J. Tait, we were Diabetes 289 able to examine a specimen of the patient's urine in November, 1902, and found that the glycosuria had re- turned. A well-marked "pancreatic" reaction was also obtained. A second specimen, examined on February 2, 1904, gave similar results. It was then stated that the patient was in very good health, and had had no illness since the operation. In January, 1906, a further exami- nation was made, and the urine was found to contain 0.95 per cent, of sugar. No acetone or diacetic acid was present, but it still gave a positive "pancreatic" reaction. Chronic interstitial pancreatitis has been frequently observed in association with dia- betes, but in the ma- jority of cases where interstitial changes in the pancreas arise from obstruction of the ducts by gall- stones, or from other causes, glycosuria is not met with as a symptom. In sixty- five consecutive cases, where biliary calculi were found in the common duct at operation and the pancreas was enlarged and hard, we have detected sugar in the urine of four (16 per cent.). In three of these the amount was under 0.2 per cent., and in the fourth 0.4 per cent, was present. After operation the sugar disappeared from the urine in all but the last, in which it slowly in- creased in amount, the patient dying from diabetic coma ten months subsequently. Since the interstitial changes arising from the presence of gall-stones in the common duct principally affect the head of the gland in the first 19 ^s»^^r-- Fig. 107. — Spheroidal-celled carcin- oma of the pancreas, islands of Langer- hans not affected, no glycosuria (X 40). 290 The Pancreas: Its Surgery and Pathology instance, and the results of experiments upon animals have shown that even a small portion of healthy pancrea- tic tissue will prevent the onset of diabetes, or at least delay the appearance of the symptoms, so long as it remains undestroyed by fibrotic changes, the comparative Fig. 108. — Chronic interstitial pancreatitis following duct obstruc- tion, from carcinoma of the duodenum, showing islands of Langerhans unchanged though embedded in sclerotic tissue; no glycosuria (Opie). rarity of glycosuria in these cases is not difficult to ex- plain. When, however, a great part of the parenchyma has been destroyed, or is functionally impaired, by the progressive changes consequent on repeated or long- continued irritation, either from gall-stones or an unre- Diabetes 291 Fig. 109. — Section of the pancreas from a case of diabetes following gall- stone obstruction (X 42). lieved duodenal catarrh, sugar will make its appearance in the urine, first as an alimentary glycosuria, and later as a permanent dia- betes. Some observers have attempted to define a particular type of pancreatitis associated with dia- betes. Hoppe-Seyler and Fleiner have de- scribed cases of the disease accompany- ing general arterial sclerosis, and Bosan- quet in two out of seven cases he inves- tigated found arterio- sclerosis, accompanied, in one instance in which the glyco- suria had been intermittent, by gangrene of the leg. Bosanquet suggests that- in the last mentioned case the appearance and disappearance of the sugar might be due to intermittent pancreatic failure, analogous to the "intermittent claudica- tion ' ' sometimes met with in arteriosclerosis. Lemoine and Lannois thought that the new- growth of fibrous tissue originated in the peri- vascular tissue, whence it spreads into the parenchyma. In four cases they ■^ 1>J *-i^ Fig. no. — Chronic interstitial pancreatitis with arteriosclerosis, from a case of diabetes (X 40). 292 The Pancreas: Its Surgery and Pathology examined, they described the penetration of fibrous bands into the acini, separating the cells and giving rise to a unicellular sclerosis. Opie considers that diabetes is peculiarly related to interacinar pancreatitis and that in the interlobular form it is rare. The difference, he considers, depends upon the relation of the fibrous tissue overgrowth to the islands of Langerhans, which he, in common with many other observers, believes are responsible for the elaboration of the internal secretion by means of which the pancreas exerts its influence upon carbohydrate metabolism. The newly formed fibrous tissue in interacinar pancreatitis is diffusely distributed within the lobules and between the individual acini, so that the islands are affected at the same time as the other elements of the gland, but in the interlobular form, which is the type following duct- obstruction, the proliferation of fibrous tissue takes place between the lobules and invades them from the periphery, so that the cell islets suffer only when the process is far advanced, and the secreting parenchyma has been re- placed by masses of scar-like tissue. Opie found that diabetes was present in seven out of nine cases of inter- acinous pancreatitis, but that in only one out of twenty- one cases of chronic interlobular inflammation of the gland was there sugar in the urine. In this one case of interlobular inflammation the induration of the gland was far advanced and the islands of Langerhans were fibroid. The two cases of interacinous pancreatitis in which diabetes was absent were both, he found, in an early stage of the disease. The suggestion that the islands of Langerhans are concerned in the production of the "internal secretion" of the pancreas was first made by Laguesse. This view was subsequently adopted by Schafer, Diamare, and others. The theory that such a relationship exists is based partly on histological grounds, and partly upon the Diabetes 293 results of experimental work, but the most important evidence in support of it has been furnished by patho- logical observations, which suggest that pancreatic diabetes is due to a disturbance of the functions of the cell islets. The peculiarity of their structure, their independence of the duct system of the gland, and their comparative resistance to certain morbid changes by which the secret- ing acini are destroyed, point to their being independent organs with an independent function; while their rich blood supply may be taken to indicate that they are possibly vascular glands, engaged in the elaboration of some internal secretion which is poured into the blood stream. Ssobolew, who has sought by experimental means to prove the relationship of the islands of Langer- hans to carbohydrate metabolism, states that when animals are overfed with carbohydrates the granules, which have appeared in the cells during hunger, diminish in number, in the same way as the zymogen granules of the secreting cells diminish as the result of functional activity. Intravenous injections of sugar, he believes, bring about a similar result. But this has been denied by Schmidt, who also failed to observe any change on introducing sugar into the peritoneal cavity of mice and guinea-pigs. Schulze, however, experimenting with guinea-pigs, has confirmed the observations made by Ssobolew. Basing her experiments upon the observation of Schulze, Ssobolew and others, that complete atrophy of the glandular acini of the pancreas is caused by ligaturing the duct, while the islands of Langerhans remain un- changed, Lydia De Witt has attempted to isolate the cell islets in cats, and study the physiological action of an extract made from them. She found that the changes in the glandular parenchyma were much the same as those described by Ssobolew, but that when no special effort was made to avoid including the blood-vessels in 294 The Pancreas: Its Surgery and Pathology the Hgature, the islands, as well as the gland tissue, some- times atrophied, whereas when precautions were taken to avoid interfering with the blood supply of the gland the islands were well preserved, regardless of the extent of atrophy of the gland tissue. No sugar was found in the urine, either after the operation or just before death, but in three out of four cases a positive "pancreatic" reaction was obtained when the urine was examined shortly before the animal died. To test the physio- logical powers of the atrophied gland, it was removed immediately after death, extracted with glycerine or water, and the digestive and glycolytic actions of the extracts investigated. In seven out of twenty cases there was no digestion of starch, fibrin, or fat, while in several others the digestive action was very much weak- ened, the diminution and absence of digestive action being apparently proportionate to the degree of atrophy of the glandular tissue. No appreciable weakening of the glycolytic, or activator, power of the extract, as tested by Cohnheim's method, was noticed in any of the cases, even when the glandular tissue had undergone atrophic changes. The results of these experiments, although suggestive, and tending to support the theory that the islands of Langerhans manufacture a substance analogous to the " activator principle" of Cohnheim which favours the glycolytic action of muscle ferment, were not as decisive as had been hoped, for the isolation of the islands was not always complete, and it was found diffi- cult to obtain from the cat's pancreas sufficient extract to make many satisfactory tests. Rennie has carried out some investigations with ex- tracts prepared from the large cell islets, dissected out free from pancreatic tissue, met with in Lophius piscato- rius and Scorpoena scropha, and found that they had no inverting pOAver. Diamare and Kuliabko state, however, that some inversion takes place on standing for forty- Diabetes 295 eight hours. The latter have also shown that the extract has no digestive power for starch, whereas an extract made from the pancreas of the same fish rapidly converts starch into sugar. The pathological changes that have been met with in the interacinar islets in cases of diabetes are hyaline de- generation, necrosis, atrophy with vacuolisation and liquefaction of the cell-protoplasm, acute and chronic inflammation with haemorrhage, sclerosis, or calcification, and diminution of the number of islets. Opie's description of hyaline degeneration in the islands of Langerhans alone, in diabetes, furnished the most con- vincing evidence of the association of these structures with carbohydrate metabolism up to that time available, and from the publication of his paper in 1901 may be dated a revival of interest in the pancreatic theory of diabetes, and a more searching and minute inquiry on the part of other observers into the condition of the cell islets in fatal cases coming to post-mortem. In nineteen cases of diabetes, in which Opie investigated the condition of the islands of Langerhans, he found hyaline degenera- tion in seven (35 per cent.). The first example of the lesion that came under his observation w^as in a severe case of diabetes. The islands were^so completely altered as to be unrecognisable, and the secreting parenchyma was also in great part destroyed. In subsequent cases of his series the relationship of the diabetes to the lesion in the islands was more conclusively demonstrated, for, although these structures had undergone very grave alter- ations, and were often converted into almost homogene- ous masses of hyaline material, the secreting parenchyma showed in some instances only insignificant changes, and in parts of the gland was unchanged. Other observers, including Wright and Joslin (two cases with only very slight alterations of the glandular acini) , Herzog (one with slight chronic interstitial infiammation) , 296 The Pancreas: Its Surgery and Pathology Fig. III. — Sections of the pancreas from a case of diabetes showing sclerosis of the capsule of the islands of Langerhans: a, Low power; b, the same islet under higher magnification (Gaylord and Aschoff). Diabetes 297 Schmidt (one with no lesion of the parenchyma), and Lepine (one with recent fibrosis), have met with a similar condition, but it would not appear to be as common as Opie's experience would suggest. Bosanquet has pointed out that the possibility of the degeneration being a secon- dary change in diabetes has not been excluded. He quotes a case in which he found it apart from diabetes in association with extensive arteriosclerosis of the pan- creatic vessels, in a woman who died after an operation for gall-stones. According to Weichselbaum and Stangl, who have studied the islands of Langerhans in thirty-five cases of diabetes, the lesion most frequently met with is simple atrophy of the cells together with vacuolisation and lique- faction of the cell-protoplasm. Sclerosis of the cell islets was only met with in four out of their second series of seventeen cases. Herzog has studied three cases of diabetes in which the islands were the seat of marked sclerotic changes, and Schmidt has met w4th two in which there was interacinar pancreatitis so seriously involving the cell islets that many of them were converted into connective- tissue balls resembling fibrosed glomeruli, as they were also in one of Herzog 's cases. In a case reported by Lepine the islands were surrounded, and in places partly destroyed, by a new-growth of fibrous tissue. Gentes has also described a case of diabetes with chronic interstitial pancreatitis invading the islands of Langer- hans. An acute inflammation, limited to the cell islets, was met with by Schmidt in the case of a child of ten whose urine contained 6.8 per cent, of sugar, and focal necrosis of the pancreas involving the islands of Langer- hans was seen by Opie in one case. Absence of the islands, or diminution of their number, has been reported by several observers. Ssobolew failed to find them in six cases of diabetes, and stated that in nine others they were abnormally few. Herzog found a 298 The Pancreas: Its Surgery and Pathology diminished number in three out of five cases. Weichsel- baum and Stangl reached the conclusion that the number of islands may be diminished in diabetes, and, since the pancreas is almost always atrophic, the total number is still further curtailed. Opie, however, points out that the distribution of the islands varies in different parts of the gland, and that, while they may be almost absent in some parts, thqy may be numerous in others, and partic- ularly in the tail of the organ. Before arriving at any conclusion as to the relative number of islands in any portion of the pancreas, it is, therefore, necessary to compare it with sec- tions prepared from corresponding parts of the gland from nor- mal individuals of the same age. This condi- tion appears to have been fulfilled by Weichselbaum and Stangl, but they did not, however, separate cases in which the islands showed lesions from those in which they were apparently normal. Opie has compared the size and distribution of the cell islets in the head, body, and tail of the pancreas in eight cases of diabetes, and found that the figures obtained showed no constant departure from the normal. A striking diminution in the number of islets was seen in two cases, and in one of these, a child of fourteen in whom the diabetes was hereditary, he sug- gests that it might be due to a congenital defect of the gland. He concludes that " while diminution in the size of Fig. 112. — Chronic interstitial pan- creatitis with fibrosis of the islands of Langerhans from a case of diabetes (X 40). Diabetes 299 the gland, together with absolute and relative diminution in the number of interacinar islands, may occasionally explain the occurrence of diabetes, with our present knowledge it is unjustifiable to assume the existence of such functional deficiency when no lesion can be demon- strated by the methods at our disposal." Most of the published cases of diabetes in which the condition of the islands of Langerhans is reported have been collected by Sauerbeck. He found that in one hun- dred and seventeen out of one hundred and fifty-seven there was some abnormality. If the purely quantitative changes are rejected, as being of too indefinitive a char- acter, there remain ninety-eight (62 per cent.) in which qualitative changes were observed. Most modern observers, who have systematically investigated the islands of Langerhans in diabetes, have eith,er accepted the view that there is a causal relation- ship between the disease and lesions of the islands, or suspended judgment until further evidence is available. Hansemann, however, definitely states his conviction that no such relationship exists. He investigated thirty-four cases, and found that the islands were present in all. In some, where nearly the whole parenchyma had been destroyed by fat or interstitial fibrosis, he states that, although they were diminished in number, they were unchanged. In six cases he found the islands invaded by what he regards as hyaline connective tissue, but since they were not all affected and there was an accompany- ing interstitial fibrosis of the gland, it appeared to be a matter of chance whether the islands were involved or not, but he admits that he has not met wdth a case in which fibrosis affected the islands without diabetes being present. Chauft'ard and Ravant met with swelling and increase of size in the islands of Langerhans, without glycosuria, in thirteen cases of enteric fever, two of pneumonia, and 300 The Pancreas: Its Surgery and Pathology one of erysipelas. They do not regard the condition as pathological, however, but consider it as a hypertrophic reaction. Salisbury Trevor observed similar changes in the cell islets in pneumonia and infective endocarditis. Herxheimer, studying the cell islets in the cirrhotic pancreas so often found in diabetes, states that he found evidence of their new formation from the small ducts, but he regards the whole pancreas as controlling sugar metabolism, and thinks that diabetes is due to a functional lesion of the gland, which may or not be accompanied by visible morbid changes. In man he considers that the islets alone are inadequate for the prevention of diabetes, while in animals they appear to be sufficient. This statement is, however, not supported by any convincing evidence, and is in fact directly con- troverted, as regards human diabetes, by a case described by S. G. Scott, in which, although only the islands of Langerhans remained, no trace of sugar was found in the urine. Even if the connection of the islands of Langerhans with diabetes is granted, a certain number of cases remain in which no lesion whatever of the pancreas has been dis- covered by competent observers. Opie met with four in his nineteen cases. Williamson eight in twenty-two, Ssobolew two in fifteen, and in twenty- three examined by Schmidt there was no change in eight, and in eight others the alterations were so slight as to be considered secondary to the diabetic condition. It may be contended that the specific diabetic disturbances of the pancreas are not necessarily connected with visible anatomical alterations of the islets, or in any other tissue of the gland, and, although it is possible that in some instances there may be minute alterations in the molecular arrangement of the cells which cannot be discovered with the micro- scope, we are at present unacquainted with any other method of demonstrating minute morbid changes, and Diabetes 301 most perforce aV)i(le by the results obtainable by the means at our disposal. It is ^jossible, however, that at the present time we include under the term diabetes several conditions, having somewhat similar symptoms, and all characterised by the presence of glycosuria, which are not due to the same cause. There is no doubt that those in which lesions of the pancreas are present form a larger class than is generally supposed, but it may prove that diseases of other organs or tissues are responsible for some, and that the cases in which the pancreas appears to be normal after death have a separate origin. At the same time it is well to bear in mind, as von Noorden has insisted, that scientific medicine has been so long under the influence of morbid anatomy that it is often difficult to realise that important disturbances of function may occur when microscopical examination reveals no dis- tinctive pathological changes. The association of arteriosclerosis, gout, syphilis, and alcoholism with diabetes is probably to be explained by the fibrotic and degenerative changes which each is capable of setting up in the pancreas. Cirrhosis of the liver is often found to coexist with chronic interstitial pancreatitis in diabetes, and both probably originate from the same causes. The apparently infective cases of diabetes may be due to the effects produced upon the pan- creas by the entry of micro-organisms from the duodenum, and the history of digestive disturbances, which is not uncommon in diabetes, suggests that a chronic duodenal and gastric catarrh may in some instances give rise to the conditions favourable for the onset of a pancreatic lesion. Chronic pancreatitis has been repeatedly observed in association with acromegaly, and it appears probable the diabetes, transient glycosuria, and alimentary glycosuria that have been seen, not infrequently, to accompany this disease are to be referred to pathological changes in the pancreas. 302 The Pancreas: Its Surgery and Pathology We have seen that painting the pancreas with adrenaHn, and a variety of other substances, gives rise to glycosuria, but that the action of the suprarenal extract appears to be of a specific and peculiar nature. Blum has produced temporary glycosuria by injecting suprarenal extract into the veins and subcutaneous tissues of animals, and the same result, but to a more marked degree, has been found by Herter and Richards to follow injections of adrenalin into the peritoneal cavity. These experiments suggest that there is some connection between the glycolytic action of the pancreas and the suprarenal bodies, and the anatomi- cal picture presented by cases of so-called "diabete bronze" tends to favour this view. Bronzed diabetes is closely related to the condition de- scribed by von Recklinghausen under the name of hsemo- chromatosis, in which the epithelial cells of the various glands of the body, and particularly of the pancreas and liver, show deposits of a reddish-yellow iron-containing pigment, the smooth muscle-fibres of the blood- and lymph -vessels, and of the walls of the gastro-intestinal tract, contain fine granules of a bright yellow, iron-free pigment, and there is hypertrophic cirrhosis of the liver, but, unlike simple hasmochromatosis, it is associated with a rapidly fatal form of diabetes mellitus. Although bronzing of the skin is present in the majority of cases, it is not constant. When present it is usually general and uniform, but is not accompanied by pigmentation of the mucous membranes, as in Addison's disease. Hanot and Chauffard, who first described the condi- tion in 1882, found, in one of their cases, of which they made a careful study, that there was advanced chronic interstitial pancreatitis, and other observers, who have had the opportunity of investigating the disease, have also found that the pancreas was affected. Where a microscopical examination of the gland has been made interstitial fibrosis has been found, the connective-tissue Diabetes 303 spaces being much enlarged, and they, as well as the cells, have contained a deposit of reddish-yellow f)igiTient. Margain has recently reported that, in a fatal case of bronzed diabetes he examined, some of the islands of Langerhans were preserved, but that their cells were crowded with pigment. Hanot and Chauffard believed that the diabetes was the primary factor in the disease, the changes in the liver and other tissues being due to diabetic alterations in the blood, and to the accompany- ing endarteritis, but Marie Acard, Dutourier and Jeanselme, and Anschiiltz think that the tissue changes result from a deposit of pigment in them, and that the pigment arises from a dissolution of haemoglobin from some un- known cause. According to this view, the diabetes is a secondary phenomenon due to the changes in the pancreas. Opie, who has examined a case of haemochromatosis, is of opinion that it is a distinct morbid entity, associated with chronic interstitial inflammation, notably of the liver and pancreas, and that, when the pancreatitis has reached a certain grade of intensity, diabetes mellitus ensues, and is usually the terminal event. He finds that the pancreatic inflammation is of the interacinar type, and that the islands of Langerhans are implicated in the lesion. A few cases have been described in which there has been no glycosuria, although the whole of the pancreas has been apparently destroyed by malignant disease or inflam- matory processes, but none are of recent date, and in most instances the proof of total destruction rests upon macroscopical examination alone. In view of the abun- dant experimental and clinical evidence now available, that the pancreas is essential for carbohydrate metabo- lism, the proof that its absence can be unattended by glycosuria must be exceptionally strong, although it must be admitted that a single well authenticated in- stance would call for a revision of the views at present generally held. 304 The Pancreas: Its Surgery and Pathology Minkowski, as we have seen, showed that if only part of the pancreas be removed in animals diabetes does not result, but the ability of the organism to perform its nor- mal functions in carbohydrate metabolism is impaired, so that large doses of sugar give rise to temporary glyco- suria. The capacity of the body to deal with carbohy- drates is normally limited, and is not the same for all varieties of sugar. Glucose appears to have the highest limit (150 to 200 grams in one dose), tevulose can be taken in somewhat similar amounts without producing glycosuria (140 to 160 grams in a single dose), cane-sugar also can be taken in doses of 150 to 200 grams, but milk- sugar has a much lower limit (80 to 100 grams). Accord- ing to V. Noorden, the assimilative capacity of normal individuals varies very considerably for maltose, for while there are some who can tolerate considerable quantities, there are others who possess a very low assimilation limit. According to him, this accounts for the appearance of sugar in the urine of some persons after even a moderate amount of beer has been taken. Pentoses are only assimilated with difficulty, and even the ingestion of so small a quantity as 30 to 50 grams is followed by the appearance of almost half in the urine. For starch no limit is known, for if as much as 400 or 500 grams are consumed in a few hours alimentary glycosuria does not occur. Pathologically alimentary glycosuria occurs in certain nervous and brain troubles, in diseases of the liver, and as the result of lesions of the pancreas, as well as in some cases of acute febrile disease, and in acute and chronic alcoholic intoxication. The glycosuria met with in the two last groups has been attributed to disturbances of the liver, but v. Noorden considers that it arises from real, though transitory, disturbances of the pancreas, for the glycosuria is much more marked than is that met with in liver disease, the influence of glucose much ex- Diabetes 305 ceeds that of laevulose, whereas in liver disease, as a rule, the organism reacts much more strongly to the latter, and glycosuria can be produced by an excess of starchy food. The relation of the pancreas to alimentary glycosuria in man has been investigated by Willie, who gave 70 to 100 grams of grape-sugar to eight hundred patients, suf- fering from a variety of diseases, in the morning before food had been taken. The urine was examined before the test, and at intervals of two hours afterwards. Of the eight hundred individuals, seventy-seven subse- quently died and were examined post-mortem. Alimen- tary glycosuria had been found in fifteen of these, and in ten there were present grave lesions of the pancreas, either primary, or secondary to growths in the stomach, liver, or gall-bladder. Since pancreatic diabetes presents no characteristic clinical symptoms by which it can be recognised with certainty, its treatment differs in no essential respect from that by which it is sought to stay the progress of diabetes in general. Reliance must be placed mainly upon diet and the control of the hygienic condition of the patient. Although numerous drugs have been sup- posed to have a beneficial effect in cases where the pan- creas was believed to be diseased, there is no evidence that a pancreatic lesion can be directly controlled by this means, except possibly, to a certain extent, by mercury in syphilitic cases. As it is probable, however, that some cases of pancreatic diabetes may arise from an infection reaching the gland by way of the intestine, drugs calcu- lated to allay catarrh of the duodenal mucous mem- brane and control the intestinal flora may be of some service in preventing the progress of the disease. It is also possible that gastro-enterostomy and cholecysten- terostomy, by draining the affected areas and putting them to rest, might in similar cases have a beneficial effect. 3o6 The Pancreas: Its Surgery and Pathology The satisfactory results following the administration of thyroid extract in myxoedema and sporadic cretinism have naturally suggested that the use of pancreatic extracts, or of the fresh gland, might be equally effectual in the treatment of pancreatic diabetes. In a few cases it has been claimed that some amelioration of the symp- toms has been produced in this way, but the majority of observers are agreed that, although some improvement in the digestive powers may result, the glycosuria and other symptoms of the diabetic condition are unin- fluenced. In order to avoid destruction of the ferments contained in the extract by digestion in the alimentary tract, subcutaneous injection has been resorted to, but with equally unsatisfactory results. Clinical experience, however, in these respects only confirms the experimental results obtained by Minkowski, Thiroloix, and others, who found that in depancreatised animals the administra- tion of fresh pancreas or pancreatic extracts by the mouth, subcutaneously, into the peritoneum, or into the veins had no effect in controlling the glycosuria. It is possible that the action of the pancreas in carbohydrate metabo- lism may be a function of the living gland, and that, for this reason, extracts and preparations made from the dead organ may fail to be of use in diabetes, but the subcutaneous implantation of the pancreas of the lower animals into diabetics, contrary to what might be expected from experimental work on the grafting of pancreatic tissue beneath an animal's own skin previous to extir- pating the gland, has not proved of any service as a therapeutic measure. Basing their treatment on the effects of secretin as a stimulant of the pancreas, Moore, Edie, and Abram have employed an acid extract of duodenal mucous membrane in diabetes. They found that, when this was given by the mouth, the sugar in the urine gradually diminished in some cases, and finally disappeared in a few. In others, although Diabetes 307 there was an improvement in the digestion, no effect on the sugar output was produced. Bainbridge and Bed- dard, however, have not noticed any amelioration of the symptoms in cases they have treated by this method, and suggest, as the result of their experience, that any improvement that takes place is to be attributed to the diet and not to the secretin. J. R. Charles has also failed to notice any good following the use of secretin in three cases of diabetes. The method is, as yet, in an experi- mental stage, and the number of cases in which it has been tried are too few to prove whether it is really efficacious in genuine cases of pancreatic diabetes or not. Even should it be shown eventually that acid duodenal extract can exert some influence in controlling glycostiria, it is open to question whether the treatment may not, in the end, do more harm than good, for the artificial stimtda- tion of the diseased tissue that may remain in cases of pancreatic diabetes, although it may at first induce in- creased activity, is likely to eventually bring about fatigue, and cause more rapid degeneration than would have occurred if it had been let alone. The intravenous injection of secretin has been shown by Starling to give rise to acute inflammation of the intestines, and even to gastric ulcers, in animals, for the pancreatic juice is not met and neutralised by the acid gastric contents which normally cause the flow. This objection does not apply to the administration of secretin by the mouth, as the secretion will be gradual, and correspond to the acidity of the gastric juice reaching the intestine. Cohnheim's work upon the effects of a mixture of pan- creatic and muscle extracts in glycolysis has suggested the use of such a mixture in diabetes, and it has been employed for this purpose by Crofton. The difficulty that attends all methods of treating diabetes is vastly increased by the obscurity of the etiol- ogy, but where it follows, or is associated with, evidences 3o8 The Pancreas: Its Surgery and Pathology of pancreatic or gall-stone trouble, the origin of the disease is plain, and, in the early stages, appropriate means of treatment can be adopted with a fair hope of success. We have already referred to four cases of cholelithiasis in our own experience that were accompanied by glyco- suria, and in three of which operation was followed by disappearance of sugar from the urine. Gifford Nash's case of acute pancreatitis with glycosuria may also be cited as an example of the satisfactory results attending timely operative interference. It is essential, however, that the disease should not be too far advanced, and for this reason we strongly recommend that all cases of pan- creatitis and gall-stones likely to give rise to a pan- creatic lesion should be operated on with as little delay as possible after they have been diagnosed. As we shall show in a subsequent chapter, the early recognition of these conditions is not now a matter of great difficulty, if all the signs and symptoms to be obtained by a careful consideration of the clinical and pathological features of the case are taken into account, so that difficulty of diag- nosis cannot be urged as a valid excuse for cases of pan- creatitis being permitted to proceed untreated to a stage at which diabetes supervenes. Literature Abram: Lancet, Jan. 27, 1906. Acard: Thesis, Paris, 1895. Aldehoff: Zeitsch. f. Biol., xxviii, 293. Anschiitz: Deutsch. Arch. f. klin. Med., 1899, Ixii, 411. Bainbridge and Beddard: Bio-chemical Journ., Sept., 1906. Bloch: Quoted by Oser, Nothnagel's "Encylop. of Pract. Med." Blum: Deutsch. Arch. f. klin. Med., 190 1, Ixxi, 146. Blumenthal: Zeitsch. f. diatet. u. physik. Therapie, 1898, i, 250. Bosanquet: Lancet, 1905, i, 903. Bouchardat: Compt. rend, acad., xx, 1085. Bright: Med. Assoc. Trans., 1833, p. 18. Capparelli: Arch. ital. de Biol., 1894, xxi, 398. Charles, J. R. : Bristol Medico-Chirurg. Journ., Sept., 1906. Chauffard and Ravant: Arch, de Med. Exp. et d'Anat. Path., March, 1901, p. 175. Chopart: "Diabetes with Calculus Formation," quoted by Klebs, P- 547- Diabetes 309 Cohnheim: Zeitsch. f. physiol. Chem., 1903, xxxix, 336; IVjid., 1904, xlii, 401; Ibid., 1906, xlvii, 253. Cowley: London Med. Journ., 1788. Crofton: Amer. Journ. of Med. Sci., April, 1902. Philadelphia Med. Journ., Jan. i, 1902. American Medicine, Jan. 18, 1902. Dewitt: Journ. of Exp. Med., 1906, viii, 193. Diamare: Internal. Monatsschr. f. Anat. u. Phys., 1899, xvi, 155. Diamare and Kuliabho: Zeit. f. Phys., xviii, 14. Dutourier: Thesis, Paris, 1895. Elliotson: Med. Chir. Trans., 1833, xviii, 67. Fles: Holland. Arch., 1864, iii, 187. Fleiner; Berliner klin. Woch., 1894, xxxi, 38. Frerichs: "Leberkrankheiten," 1858; " Ueber der Diabetes," 1884. Gentes: Thesis, Bordeaux, 1901. Guleke: Archiv. f. klin. Chirurgie, Ixxxvii, Heft 4. Hammerschlag and Kauffmann: Quoted by Lenn6, Therapeut. Mon- atsch., 1902, S. 182. Hanot and Chauffard: Rev. de. m^d., 1882, ii, 385. Harley: Journ. of Anat. and Phys., 1891, p. 201; Brit. Med. Journ., 1892. Hartsen: Arch. f. Holland. Beitrage z. Naturheilkunde, 1884, iii, 319. Hansemann: Zeitschr. f. klin. Med., 1894, S. 191 ; Verhandl. derDeutsch. path. Gesellsch., 1902, iv, 187. Hedon: "Physiol, du Pancreas," Paris. Herter and Richards: Medical News, Feb. i, 1902, Herxheimer: Virchow's Archiv., 1906, clxxxiii, 2. Herzog: Virchow's Archiv., 1902, clxviii, Hft. i, 83. Hoppe-Seyler: Deutsches Arch. f. klin. Med., 1893, Iii, 171. Jeanselme: Bull, et mem. d. h6p. de Paris, 1897, ^^^' ^79- Laguesse: Compt Rend. Soc. de. Biol., 1893. Lancereaux: Bull. acad. de. m6d., 1877, p. 1224. Lemoine and Lannois: Arch, de m6d. exp., 1891, iii, 33. Lepine: Compt. rend, de la Soc. d. Biol., 1903, Iv, 161; Lyon Medicale, 1903, xliii, 623; Journ. de Phys. et de Path, gen., 1905, vii, i. Lorand: Compt. rend, de la Soc. de Biol., 1904, Ivi, 488. Margain: Rev. de Medecine, March 10, 1905, p. 214. Marie: Semaine m6d., 1895, xv, 229. Mering and Minkowski: Klebs' Arch., 1899, xxvi, 37; Semaine med., 22 mai, 1889; Arch. f. exper. Path. u. Phar., 1890, xxvi, 371. Minkowski: Berliner klin. Wochenschr., 1890, Nr. 8; Centralbl. f. Pathologie, 1892; Arch. f. exper. Path. u. Phar., 1893, xxxi, 85. Moore, Edie, and Abram: Bio. Chemical Journ., Sept., 1906. Munk: Tagebl. der 43 Naturforscherversammlung, 1869, S. 112. Nash, Gifford: Lancet, November 11, 1902. Noorden: "Diabetes Mellitus," tr. Boardman Reed, 1906. Opie: "Diseases of the Pancreas," 1903. Pavy: "Carbohydrate Metabolism and Diabetes." Recklinghausen: Virchow's Arch., 1864, xxx, 362; Tageblatt der 62 Versamme. deutsch. Naturforsch. u. Aertze in Heidelberg, 1889, . S. 324. Rennie: Zent. f. Phys., 1905, xvii, 23. Sandmeyer: Zeitsch. f. Biol., 1895, xxxi, 13. Sauerbeck: Virchow's Archiv, 1904, clxxvii, Suppl. Heft 1; Verhandl. der deutsche path. Gesellsch. Centralb. f. path. Anat., 1904, XV, 217. Schafer: Lancet, ii, 1905. Schmidt: Munchener med. Woch., 1902, xlix, 51. Schulze: Arch. f. mik. Anat., 1900, Ivi, 491. 3IO The Pancreas: Its Surgery and Pathology Seegen: "Der Diabetes mellitus," Berlin, 1893. Silver: Trans. Path. Soc, 1873, xxiv; Ibid., 1878, xxix. Ssobolew: Virchow's Arch., 1902, clxviii, 91; Centralbl. f. allg. Path. u. path. Anat., xi, 202. Starling: Tr. Path. Soc, Ivi, 255. Strauss: Zeitsch. f. klin. Medicin., xxvi, 27. Thiroloix: Thesis, Paris, 1892. Trevor, Salisbury: Practitioner, April, 1904, p. 574. Tuckett: Journ. of Physiol., 1899, xxv, 63. Umber: Zeitsch. f. klin. Med., 1900, xxxix, 13. Underhill: Journ. of Biolog. Chem., i, i, 113. Weichselbaum and Stangl: Wien. klin. Wochensch., 1901, xiv, 968; Ibid., 1902, XV, 969. Weintraud: Arch. f. exp. Pathol., xxxiv, 303. Williamson: "Diabetes Mellitus," 1898. Windle: Dublin Journ. of Med. Sci., 1883. Wright and Joslin: Journ. of Med. Research, 1901, i, 360. CHAPTER XII GENERAL SYMPTOMATOLOGY AND DIAGNOSIS The varied and important part the pancreas takes in the digestive processes that go on in the intestine, and the equally important influence that it appears to exert upon the internal metabolism of the body, would natur- ally suggest that any departure from the normal would lead to such disturbances of function that the symp- toms of diseases of the pancreas would be so marked as to make the diagnosis easy. But this is far from being the case for several reasons: First, it is seldom that the pancreas is diseased without other organs also being involved; for example, pancreatitis is very frequently associated with cholelithiasis; gastro-intestinal catarrh and catarrh of the bile and pancreatic ducts often coexist ; ulcers or tumours of the stomach or duodenum may extend to, and involve, the pancreas; and affections of the liver, colon, or lymphatic glands may give rise to disease in the pancreas. Secondly, the digestive functions of the pancreas can be carried out, more or less completely, by other agencies, the stomach can deal with proteids, the salivary and intestinal glands have the power of digesting starches, and the bile and intestinal secretions can emul- sify fat; moreover, the intestinal bacteria, as we have seen, possess the power of breaking down various food materials, and so interfering with the pathological altera- tions in the stools that might be expected in pancreatic diseases. Thirdly, a considerable portion of the gland may be necrosed and cast off, or otherwise disabled, and yet the remaining portion may apparently be sufficient to carry on the functions of the organ. Fourthly, in 311 312 The Pancreas: Its Surgery and Pathology some cases the pancreas may be the organ primarily at fault, and yet the most prominent symptoms may be caused by another organ that is involved secondarily; for instance, cancer of the head of the pancreas gives rise to intense jaundice and distension of the gall-bladder, suggesting to the uninitiated a primary affection of the liver or bile-ducts, but the symptoms are in reality due to gripping of the common bile-duct by the growth, and simple chronic pancreatitis may also cause jaundice for the same reason. Again, a tumour of the pancreas may compress the intestine and produce intestinal obstruction, or may press upon the neighbouring ganglia and cause most violent pain, mimicking that met with in spinal disease, aneurysm, etc. Thus it will be seen that in diseases of the pancreas very conflicting combinations of symptoms may arise, which may lead to great difficulty in diagnosis, unless some well-defined guiding principles can be established by which it may be determined, with a considerable degree of certainty, that the pancreas is, or is not, the organ primarily at fault. This we hope to show is not impossible. The signs and symptoms which are present, to a greater or less extent, in most pathological conditions of the pancreas may be classified as follows: (i) physical signs, (2) digestive symptoms, (3) metabolic symptoms, (4) symptoms produced by artificial means. I. Physical Signs. — (a) Tumour. — The situation of the pancreas, behind the stomach and in front of the spinal column, places it in a very unfavourable position for palpation, and, normally, if the patient be at all stout, it can only be felt indistinctly; but when the patient is thin, and especially in cases of gastroptosis, it can be readily defined, if the muscles are relaxed and a warm flat hand is applied firmly to the epigastric region. It is commonly stated in text-books that acute and chronic inflammation, and even abscess of the pancreas, rarely, General Symptomatology and Diagnosis 313 if ever, cause perceptible enlargement of the organ, but this is not correct, for in many cases a distinct swelling may be felt, which, in acute cases, is made up of the enlarged pancreas, with surrounding effusions of blood and inflammatory fluid and matted omentum ; in sub- acute cases the swelling is due to suppuration ; and in chronic inflammation it arises from tumefaction of the gland itself. In cancer of the head of the pancreas the only tumour that is ordinarily felt is that due to the en- larged gall-bladder, which can be readily palpated in a considerable proportion of cases. Tumours of the body or tail, as well as some growths of the head of the gland, can be readily distinguished, and by distending the stomach with gas, either by means of bicarbonate of soda and tartaric acid given in separate doses, or by pumping in air through the stomach-tube, the relation of the stomach to the tumour can be satisfactorily made out. Resonance on percussion, owing to the position of the stomach, unless this organ is empty, communicated non-expansile pulsa- tion, and very slight movement on deep inspiration are characteristic of swellings of the pancreas. In cystic diseases of the gland a tumour is frequently, at first, the only symptom ; the position and relation of such a tumour depend on the part of the organ from which it springs, as we shall show later. It will thus be seen that the absence of a tumour does not negative serious disease of the pancreas, but the presence of a swelling, when taken with other symptoms, affords valuable evidence in favour of a pancreatic lesion. (6) Fever. — A rise of temperature is, as a rule, associated with acute and subacute pancreatitis, but only rarely with any of the more chronic forms of inflammation. Cystic disease, calculus, and new-growth do not generally give rise to fever. In acute pancreatitis the temperature may be high, but in some cases, as in the hasmorrhagic form, it . is usually subnormal. The temperature in 314 The Pancreas: Its Surgery and Pathology suppurative pancreatitis is generally irregular, and may assume a hectic type, but occasionally it is subnormal. A persistent temperature of ioi°F. to 102° or io3°F., associated with rigors, was observed by one of us in a case of pancreatitis with abscess formation. In cancer of the head of the pancreas the temperature is generally subnormal, although occasionally there may be fever from attendant complications, such as cholangitis and abscess of the liver. It will thus be seen that fever as a symptom of disease of the pancreas is extremely variable, and, alone, is no guide, though when associated with digestive, metabolic, and other signs, it may be of con- siderable assistance in making a differential diagnosis. (c) Pain and Tenderness. — These symptoms, although important when present, are so variable that even their complete absence is no proof that the pancreas is normal. Both pain and tenderness are, as a rule, absent in malig- nant disease of the head of the pancreas, but in exceptional cases of carcinoma and sarcoma of the head, body, or tail the pain may be excruciating. This is due either to pressure on, or involvement of, the great sympathetic ganglia, or to pressure on, or invasion of, neighbouring viscera, particularly the stomach and duodenum. Small scirrhus tumours are, as a rule, characterised by absence of pain, while large growths are often marked by constant and extreme agony. In the various forms of pancreatitis pain and tenderness in the epigastrium are generally well marked. The more acute inflammations are characterised by excessive tenderness on pressure, the presence of a tender spot just above and to the right of the umbilicus, rigidity of the recti, and pain of an agonising character. The pain in hsemorrhage into the pancreas is intermittent, being at times severe and of a colicky character, then diminishing or disappearing, to return later with increased intensity. In chronic pancreatitis pain and tenderness, although usually present, may be but little marked. In General Symptomatology and Diagnosis 315 some cases, however, the pain is paroxysmal and severe, and epigastric tenderness is pronounced. Cysts are frequently painless and free from tenderness, but in some instances both pain and tenderness are well marked. Even in abscess of the pancreas pain is not a constant symptom. It may be absent, as in the case reported by Stibler, but in the majority of cases both pain and tender- ness are pronounced. Calculus of the pancreas may exist for years undetected, and unsuspected, without caus- ing any pain. If, however, the calculus reaches the ori- fice of the pancreatic duct, or is impacted in the ampulla of Vater, severe paroxysmal pain, resembling a gall-stone seizure, will occur and be as- sociated with jaundice. As to the character of the pain, it may be continuous or paroxysmal, and may be limited to the epigastrium or radiate around either side of the thorax. Pain in the back, under the left scapula, or between the scapulae, is more frequent than pain be- neath the right scapula in pancreatic disease, thus serving to distinguish it from gall-bladder pain. " Coeliac neural- gia" was a term long ago applied to epigastric pain such as is associated with some forms of pancreatic disease, and such pain may radiate to the cardiac region and resemble angina pectoris, both in its intensity and in its effect upon the circulation. It will thus be seen that, while pain is a guide to diagnosis, it is not pathognomonic of any special form of pancreatic disturbance, except acute pancreatitis. Fig. 113. — Most frequent site of the tender spot in inflam- matory affections of the pan- creas. 3i6 The Pancreas: Its Surgery and Pathology {d) Pressure Symptoms. — Owing to involvement of the portal vein, ascites is seen at times in the later stages of cancer of the pancreas, and, when there is also pressure on the inferior vena cava, oedema of the lower limbs will occur. Pressure on the portal vein may also cause en- largement of the spleen, and give rise to hemorrhoids. In those rare cases where the duodenum is surrounded by, or partly surrounded by, the head of the pancreas, malig- nant disease, or even inflammation of the head of the gland, may lead to obstruction of the passage of the stomach contents, causing gastric dilatation and vomiting, as in pyloric stenosis. The stomach, duodenum, and colon may also be pressed upon by cysts or new-growths of the pancreas, and be seriously displaced; the stomach, for instance, may be pushed upwards beneath the diaphragm or downwards below the umbilicus. Distension of the gall-bladder, with jaundice, is so frequently found in cancer of the head of the pancreas that it is now a well-recognised sign of the disease, but it must be remembered that in some cases of chronic pancreatitis a similar sequence of events may occur, the presence or absence of the sign in this instance being determined by the relation of the common duct to the head of the pancreas. In some cases the hepatic duct may be pressed upon when the common duct is free, as in a case coming under our observation, where, owing to a prolongation upwards of the pancreas being involved in a chronic inflammation of the gland, there was jaundice without distension of the gall-bladder. The pressure of a pancreatic cyst upwards, on to the under surface of the diaphragm, may cause dyspnoea, from interference with the functions of the heart or lungs, and, in cases of inflam- matory effusion into the lesser peritoneal sac, there may be pressure on the pericardium, through the diaphragm, leading to distressing cardiac symptoms. Occasionally hydronephrosis may be produced by the pressure of pan- General Symptomatology and Diagnosis 317 creatic tumours on one or other ureter, and pressure on, or involvement of, the solar plexus may give rise to agon- ising pain. (e) HoBmorrhage. — In inflammatory disease and malig- nant growths of the pancreas there is a well-marked hsemorrhagic tendency, which is not only seen at opera- tion, but, in advanced cases, may be manifest by bleeding from mucous surfaces and by haemorrhages into the skin or subcutaneous tissues, so that the patient bruises very readily. Profuse and uncontrollable haemorrhage from the mucous surfaces was the cause of death in the case of pancreatitis above referred to, where there was a pro- longation of the head of the pancreas upwards on to the hepatic duct, and in several cases of cancer of the pan- creas operated on by one of us, bleeding has cost the patient his life. For instance, this occurred in a case sent by Professor Clifford Albutt, which was operated on in 1888. Cholecystotomy was followed by persistent oozing of the blood from the interior of the gall-bladder, and from the stitch punctures, which resisted all the then known remedial measures, and proved fatal on the ninth day. In another case of cancer of the head of the pan- creas, sent by Dr. W. Scatterty, of Keighley, a cerebral, haemorrhage on the tenth day produced a fatal result. In neither of the cases was there any peritonitis, or other cause than the haemorrhage, to account for death. That the hsemorrhage in these cases is not entirely dependent upon the jaundice, but is associated with the changes induced in the blood by the pancreatic lesion, is shown by the fact that patients with equally profound jaundice, but in whom there is no disease of the pancreas, do not bleed to anything like the same extent. Thus, in a patient seen with Dr. T. Churton in 1889, the jaundice wasiquite as deep as in either of those just quoted, but there was no haemorrhage, although the man lived several weeks, and ultimately died from suppurative cholangitis and exhaus- 3i8 The Pancreas: Its Surgery and Pathology tion, for the obstruction was dependent on cancer of the common bile-duct above the entrance of the pancreatic duct. The tendency to haemorrhage, both at operation and after, can be successfully counteracted by the admin- istration of calcium chloride, in 30-grain doses, thrice daily, for from twenty-four to forty-eight hours before operation, and by enema, in 30-grain doses, twice daily for forty-eight hours afterwards. The following cases illustrate the efficiency of this procedure subsequent to operation, and the danger that may arise from its absence or too early disuse : A woman, aged thirty-eight years, was suffering from deep jaundice, associated with gall-stones in the common duct and chronic pancreatitis. There was no bleeding at the time of the performance of duodeno-choledocho- tomy, as calcium chloride had been administered for several days before operation. The drug was inadver- tently omitted after operation, and on the third day violent haemorrhage occurred, which was arrested by opening up the wound and packing with gauze, and at the same time giving calcium chloride in 60-grain doses twice, and afterwards in 30-grain doses for several days. No recurrence of bleeding occurred and a good recovery was made. In the case of a male patient, aged forty-two years, suffering from cirrhosis of the liver, gall-stones in the common duct, and chronic pancreatitis, no bleeding occurred at operation, owing to the previous administra- tion of lime salts. In consequence of the absence of haem- orrhage the calcium chloride was left off the second day after operation. Bleeding occurred very freely on the sixth day, in the form of general oozing, which was per- manently arrested by the free administration of calcium chloride, after which recovery occurred. In a case of suppurative catarrh of the pancreas in a gentleman, aged sixty-five years, the same freedom from haemorrhage was found at operation, after the administra- tion, for some days, of calcium chloride, which was not continued, as the rectum was intolerant of injections. On the seventh day free bleeding occurred, which was General Symptomatology and Diagnosis 319 arrested by giving thirty grains of calcium chloride every two hours by the mouth. (/) Jaundice. — The now well-recognised relation be- tween gall-stone trouble and pancreatic disease would lead one to expect that jaundice would be a frequent accompaniment of diseases of the pancreas, but the symp- tom is by no means constant. The relation of the com- mon bile-duct to the duct of Wirsung, and to the head of the pancreas, is generally the determining factor. If, as is the case in 38 per cent, of bodies (Hellyj, the common duct passes behind the head of the pancreas, either an acute or a chronic pancreatitis, or even a cancer of the pancreas, may run its course without the appearance of jaundice; but if the common duct lies in a deep groove, or is embedded in the head of the gland, as occurs in 62 per cent, of cases, either pancreatitis or growi;h of the head of the organ must necessarily compress the bile-duct and lead to jaundice of greater or less intensity. It may perhaps be only a coincidence, but it is noteworthy that in 62 per cent, of our cases of chronic pancreatitis asso- ciated with cholelithiasis bile-pigments were found in the urine before operation, and in 38 per cent, there w^as neither jaundice nor bile-pigment in the luine. Chronic pancreatitis, not associated with the presence of gall- stones in the common duct, was accompanied by jaundice in 16 per cent, of our cases, the icterus in these instances being probably due either to compression of the common duct by the swollen head of the pancreas, or to an ascend- ing catarrh from the duodenum, which simultaneously involved the pancreatic and biliary passages. In pan- creatic calculus jaundice may occur if the stone lodges in the ampulla of Vater. Acute hasmorrhagic pancreatitis may be accompanied by slight jaundice, especially when it results from the impaction of a small gall-stone in the duodenal outlet of the ampulla of Vater. 320 The Pancreas: Its Surgery and Pathology Deep jaundice, with a distended gall-bladder, is signifi- cant of cancer of the head of the pancreas, whereas if the cause of the jaundice be gall-stones in the common duct the gall-bladder is nearly always contracted and cannot be felt. The jaundice met with in cancer develops without pain, slowly and insidiously, but steadily. As the cachexia increases the patient's skin assumes a ghastly slaty appearance in many instances, so that instead of the saffron yellow colour of cholelithiasis there is seen the so-called "black jaundice." (g) Emaciation. — Ever since pancreatic disease has been recognised, emaciation has been regarded as a striking symptom. In some cases of chronic inflammation there is a very marked loss of flesh, which is rapidly regained after a cure of the condition has been effected by opera- tion. Thus, a patient who was operated on by one of us for the removal of a calculus obstructing the pancreatic duct had lost eight stones in a little over two years, but rapidly regained his normal weight after the operation. In another, and similar case, the patient had lost flve stones in three months, but gained three stones in the same period after his condition had been relieved by opera- tion. It is not surprising that emaciation should occur in cancer of the pancreas, in which, indeed, it is most marked, or in diabetes of pancreatic origin, but it may also be met with in cystic disease, as in Kuster's case, in which the patient lost two stones five pounds in four months, and it is also seen in calculus disease. The dis- turbance of digestion may afford a sufficient explanation in some cases, but in malignant disease and cases of atrophy of the gland, interference with the metabolic functions of the pancreas are probably, in part, respon- sible for the rapid wasting that is met with in these con- ditions. 2. Digestive Symptoms. — (a) Dyspepsia and Alteration of A p petite. —Dys-pe-ptic disturbances are very constantly General Symptomatology and Diagnosis 321 associated with affections of the pancreas. They take the form of anorexia, pain, and fulness after food, flatu- lence with offensive eructations, heart-burn, nausea, dis- taste for fats and for meat. In the case of a woman aged twenty-eight, to be referred to later under " Chronic Pancreatitis," where a biliary fistula was established in order to relieve the jaundice and by drainage to cure the pancreatitis that was causing pressure on the common bile-duct, all the above symptoms were well marked, both before the operation and when the fistula was dis- charging, and the patient had such a loathing for food that she became greatly emaciated. Within twelve hours of a cholecystenterostomy being performed, by which the pancreatic juice and bile were diverted into the duo- denum, she expressed herself as hungry, a sensation she said she had not felt for many months. During the month succeeding the operation she put on flesh rapidly, and three months later had gained two stones in weight. In several cases where dyspeptic symptoms have been pronounced, both in simple and malignant disease of the pancreas, the administration of pancreatic preparations after meals has been found to give marked relief, and the patients have gained in weight. (6) Nausea and Vomiting. — These symptoms are fre- quently associated with acute pancreatitis, and, in some instances, the vomiting may be so violent as to suggest intestinal obstruction. In other forms of pancreatic disease vomiting is not a common symptom, and, when present, is often due to neighbouring organs — stomach or duodenum— participating in the trouble or being pressed upon. There is nothing specially characteristic in the vomited matter, unless, as rarely occurs, extremely offensive pus and altered blood are brought up, as in cases where a pancreatic abscess has ruptured into the stomach. In one of our cases of erosion of the pancreas by chronic gastric ulcer the pus vomited was so offen- 322 The Pancreas: Its Surgery and Pathology sive that the nurses in attendance were made sick by the odour. The case was treated successfully by gastro- jejunostomy. In acute pancreatitis altered blood, the so-called "black vomit," is seen at an earlier stage than in any other peritoneal condition. (c) Fceces. — The condition of the fseces furnishes, in many instances, what might almost be termed a charac- teristic sign of pancreatic disease. In these cases the motions are exceedingly bulky, soft, greasy, and pale. They contain undigested fat and muscle fibre, and are extremely offensive. Patients frequently state that they suffer from diarrhoea, but investigation will show that this is hardly correct, for, although the stools are bulky and soft, they are not liquid in consistency. The symp- tom is a very noticeable one, and, when it occurs in cases of jaundice, may nearly always be taken as indicating that the pancreatic functions are being interfered with, either by an interstitial pancreatitis or some other form of pancreatic disease. It is more common in inflamma- tory conditions than in cancer, probably because the appetite is more interfered with by the latter than the former, so that in pancreatitis the full effects of the lack of digestive powers are seen. For the same reason the symptom is more apt to be noticed in the earlier than in the later stages of pancreatic affections, unless large amounts of milk are being given, when the bulk of it may pass away in the form of spurious diarrhoea. In some cases there is constipation, the motions being still very bulky, however, and, as a rule, pale. The bulk of these motions, out of all proportion to the amount of food taken, is to be attributed partly to the abnormal quantity of undigested food materials that they contain, and partly to the excessive fermentation that takes place in the intestines. Their frequency is due to their bulk, but is no doubt contributed to by the excess of irritating by-products they contain. The ques- General Symptomatology and Diagnosis 323 tion of their lack of colour has already been discussed (page 225), but it may be repeated that it does not neces- sarily arise from the absence of bile, as precisely similar stools may be seen in cases of pancreatic disease when there is obstruction of the biliary passages, and when there is no obstruction whatever to the free flow of bile into the intestine. Fig. 114. — Microscopical characters of the residues met with in the stools in case of pancreatic disease and biliary obstruction: a, Striated muscle fibres; 5, fat globules; c, free fatty acid crystals; d, combined fatty acid (soap) crystals. Steatorrhcea or fatty stools have long been recognised as a symptom of disease of the pancreas. Kuntzmann, in 1820, described the case of a man who died from chronic induration of the pancreas, with complete obliteration of the common bile-duct and pancreatic duct, where large stools containing undigested fat were seen. In his record of seven cases of disease of the pancreas, Bright, in 1833, noted an excess of fat in the faeces of three. Fles, in 1864, 324 The Pancreas: Its Surgery and Pathology reported the case of a diabetic, who had eaten much bacon and fat meat, with stools containing such a quan- tity of fat that it could be skimmed off the surface by the ounce. The fat disappeared when an emulsion of calf pancreas was administered, but reappeared as soon as the emulsion was omitted. The patient died of phthi- sis, and post-mortem the pancreas was found to be re- placed by fibrous tissue, with scarcely any recognisable trace of gland substance left. In many of the earlier recorded cases of steatorrhoea the pancreatic disease was associated with jaundice, and there was some doubt as to the part the lack of bile played in producing the condi- tion. More recent observations have shown, however, that fatty stools occur in diseases of the pancreas independently of jaundice, and that in such cases the steatorrhoea is to be attributed directly to the pancreatic lesion. Fitz, of Boston, in an address before the Congress of American Physicians and Surgeons in May, 1903, gave in tabular form the relationship between visible faecal fat, jaundice, diabetes, and pancreatic disease in twenty-nine cases collected from the literature of the subject. He found that in about three-fifths the steatorrhoea was attributable to disease of the pancreas unassociated with either jaundice or diabetes, that in two-fifths there was pan- creatic disease with either jaundice or diabetes in about equal proportions, and that in a few instances there was a combination of pancreatic disease, diabetes, and jaun- dice. In the following case, reported by Oser, the appearance of steatorrhoea in pancreatic disease, before the onset of jaundice, is well shown: A woman, aged thirty-nine years, had had diarrhoea since the summer of 1892. The patient became emaci- ated, yet the appetite remained good. Faecal evacuations appeared regularly every night, were unusually copious, of the consistency of thick porridge, and of cadaverous General Symptomatology and Diagnosis 325 odour, chocolate-coloured, and always abundantly cov- ered with fat rings. On January 11, 1893, he saw the patient for the first time and found steatorrhoea. The investigations of the stools gave the following results: " Large in amount and of the consistency of thick porridge ; in the sediment were scattered white particles. Micro- scopic examination showed: (i) very numerous frag- ments of striated muscle in the main with well-preser\-ed structure; (2) numerous fat acid needles and fat drops; and (3) bacteria and detritus. After drying the stools for several days on the water-bath in order to determine the amount of fat, there were obtained 4.6325 grams of solid substance, in which 2.1265 grams were fat, repre- senting 45.8 per cent, of the dried residue. The other extract consisted almost entirely of neutral fat." On January i8th he found in the epigastrium a distinct, hard, round tumour, w^hich was diagnosed as carcinoma of the head of the pancreas. In March jaundice developed. At the beginning of April an exploratory laparotomy was undertaken and the diagnosis confirmed. Fatty stools have been most frequently described in connection with cancer of the pancreas, but they have also been met with in fibro-adenoma (Biondi), calculi (Gould, Chari, Reeves, Cowley, Capparelli, etc.), cysts (Gould, Goodman, Bull, etc.), syphilitic atrophy (Demme), fibrosis (Kuntzmann) , fatty degeneration (Reeves, Motta) , abscess (Harley), and diabetes (Silver, Le Nobel, Hirsch- feld, etc.). In most instances reliance has been placed upon naked-eye observations, and in only a few have actual determinations of the amount of fat present been made. Ziehl states that, in a case of cancer of the pan- creas with jaundice that he investigated, the fat formed about 50 per cent, of the dry weight of the faeces, and Demme found from 64 to 73 per cent, of fat in the asbes- tos-like fasces of a case of congenital syphilis, with atrophy of the pancreas, that he examined. In some cases, where fat is parted with in abundance, the greasy bulky motions, occasionally coated with oil which 326 The Pancreas: Its Surgery and Pathology may float on the surface of the urine passed at the same time, are sufficiently striking to attact the attention of the patient himself, but in others, although a large amount of fat may be passed, it can only be recognised on chemical examination, as our own experience has frequently dem- onstrated. It is therefore essential, if full advantage is to be taken of the occurrence of steatorrhoea as a symptom of disease of the pancreas, that the stools should be submitted to careful quantitative analysis. The method we have described in a previous chapter gives satisfactory results for clinical purposes, and has the advantage of occupying much less time than the processes usually adopted; further, it has the additional advantage of giving not only the amount of unabsorbed fat present in the stools, but of, supplying information as to quantities of saponified and unsaponified fat that the motion contains, with but little extra trouble. We have now employed this method of investigation in a large number of cases, and found that in many it has given results of the greatest value in diagnosis. As a rule, it has been found that when the functions of the pancreas have been seriously interfered with there has not only been an excess of "total fat," but that the relation between the "neutral fats" and combined "fatty acids" has been disturbed, the former being in excess, whereas in cases of simple jaundice or biliary obstruction, although the amount of " neutral fat" may also have been abnormal, the combined "fatty acids" formed the larger proportion of the fat in the dry faeces. In some cases of malignant disease of the pancreas as much as 93 per cent, of the dry weight of the fasces has been found to be fat, and even in chronic pancreatitis we have found from 80 to 82 per cent. The average amount in malignant disease, however, has been 77 per cent. In chronic pancreatitis more than 60 per cent, has been uncommon, and in some instances an amount within the limits for a normal mixed General Symptomatology and Diagnosis 327 diet has been met with. The "neutral fats" in cancer of the pancreas have ranged from 69 per cent, to 31 per cent., and the combined "fatty acids" from 36 per cent, to 3 per cent., with an average for the former of 50 per cent., and for the latter of 27 per cent. The "neutral fats" and combined "fatty acids" are normally present in equal proportions, but in chronic pancreatitis an aver- age of 32 per cent, of the one and 18 per cent, of the other has been obtained. In our cases of biliary obstruction not associated with pancreatitis, on the other hand, the average amount of "neutral fat" has been 18 per cent., and of combined "fatty acid" 23 per cent., the "total fat" being 42 per cent. In some cases of undoubted pancreatic disease, how- ever, we have found that there was no excess of fat in the stools, and that even where an excess was present the relation between the "neutral fats" and combined "fatty acids" was not markedly disturbed, or else that there was a more or less marked excess of combined " fatty acid, ' ' contrary to what might have been expected. These variations have usually been met with in cases of pan- creatitis, and generally in the earlier stages of the disease. The absence of an excess of fat in these cases is to be explained (i) by the food containing an abnormally small proportion of fatty material owing to the distaste of the patient for fat ; (2) by the fat being of a readily digested and easily absorbed kind {e.g., milk, etc.); and (3) by the action of the fat-splitting ferment of the stomach. The relatively high proportion of combined "fatty acids" may be accounted for (i) by the action of fat-splitting bacteria in the intestines; (2) by the examination being made at a comparatively early stage in the disease, when the flow of pancreatic juice is not diminished, or may be actually increased; and (3) by an associated enteritis hurrying the contents of the intestine onwards to the large bowel before they have had time to be completely ab- sorbed. 328 The Pancreas: Its Surgery and Pathology Microscopical examination of the iseces may sometimes be of assistance when a chemical examination, for various reasons, is impracticable. In serious pancreatic disease the stools show numerous fat globules and many free fatty acid crystals. The latter, together with crystals of combined fatty acids, are also seen in jaundice, but no fat globules are usually met with in this condition, unless it is combined with disease of the pancreas. Microscopi- cal examination, however, is only of use, as a rule, in con- firming an opinion based upon the naked-eye characters, and is far inferior as a diagnostic aid to a chemical investi- gation. In interpreting the results of an examination of the fseces for undigested fat it has to be borne in mind that, in addition to diseases of the pancreas and jaundice, an excess of fat may arise from (i) an abnormal quantity being taken in the food, for the capacity for digesting and absorbing fat is limited and probably varies somewhat for different individuals; (2) from diseases of the intes- tines and mesenteric glands that interfere with absorp- tion, such as sprue, tuberculosis, etc. If these conditions can be excluded steatorrhoea is suggestive of disease of the pancreas, especially if jaun- dice is absent, and even if there is obstruction of the biliary passage an excess of "neutral" fat over combined "fatty acid" points to there being some interference with the fat-splitting functions of the pancreas in the majority of cases. Disappearance of the excess of fat in the stools on the administration of preparations of pancreas after meals tends to confirm a diagnosis of pancreatic mischief. The presence of azotorrhoea as a valuable symptom in diseases of the pancreas was first recognised by Fles in 1864. He found in the stools of a diabetic, who was proved post-mortem to be suffering from chronic intersti- tial pancreatitis, large numbers of undigested muscle fibres, which disappeared when calf's pancreas was administered General Symptomatology and Diagnosis 329 daily, and reappeared when it was omitted. Le Nobel subsequently reported a similar case. Harley found muscle fibres in the fasces, in large quantities, in a case of pancreatic abscess, Kuster in a case of pancreatic cyst, Lichtheim in a patient with pancreatic calculi, and v. Ackeron and Oser have described cases of cancer of the pancreas in which the stools contained an excess of striated muscle fibres. In twenty out of twenty-four of our cases of cancer of the pancreas, included in the table on page 214, numerous muscle fibres were found microscopically, but they were only present in sixteen out of fifty-six cases of chronic pancreatitis. An abnormally large number was also observed in one out of eight cases in which there was jaundice but no disease of the pancreas, and in tw^o out of twelve cases in which there was a stone in the gall- bladder or common duct but no pancreatic disease. It will thus be seen that while an excess of undigested mus- cle fibres is frequently met with in malignant disease, it is not such a common symptom in pancreatitis. It is usually only seen in advanced cases of cirrhosis, or in other lesions where a great part of the secreting tissue has been destroyed, and the formation of pancreatic juice is very seriously interfered with. Azotorrhoea is not so readily noticed as steatorrhoea, and attention is hardly ever drawn to it by the patient. In some instances the undigested muscle can be recog- nised by the naked eye, but, in most, a microscopical examination is necessary. Microscopical examination is, however, likely to prove misleading unless a considera- ble number of preparations are examined and muscle fibres are found in all ; moreover, it must be remembered that if much meat enters into the diet, or there is an enteri- tis by which the food material is hurried through the in- testine, an excess of muscle fibres may be met with when there is no disease of the pancreas. It must also be borne in mind that gastric digestion is necessary as a prepara- 330 The Pancreas: Its Surgery and Pathology tion for the work of the pancreatic juice on meat, for if the connective tissue binding the muscle bundles to- gether has not been attacked by the gastric juice, the pancreatic secretion can only act from the surface, eating its way slowly inwards, and consequently undigested muscle fibre will be passed in the stools. As the result of his observations on the different beha- viour of tissue elements to the gastric and pancreatic juices, Schmidt found that the nuclei of cells were digested by the pancreatic but not by the gastric secretion. He therefore concluded that if undigested tissue nuclei reap- pear in the faeces they afford evidence that the functions of the pancreas are being unsatisfactorily performed, and he suggested that, under appropriate conditions, this might serve as a test of pancreatic efficiency. Schmidt advises that small cubes of fresh, "marbled" beef, about 0.5 to 0.75 cm. thick, should be hardened in alcohol, placed in little silk-gauze bags, and preserved in alcohol until required. Before use they are to be well washed in water for several hours, placed in a wafer, and given with the food at noon for several days. The bags can be easily recovered from the fseces, on rubbing them up in water, and, after they have been washed, the con- tents can be examined fresh, after treatment with acetic acid or methylene-blue, or they can be hardened, cut, and stained. He states that the nuclei are never preserved in disorders of the liver, intestine, or stomach, but that they are found intact in destructive lesions of the pancreas, and in animals after the pancreas has been removed. .. The disadvantages of the method appear to be that the demonstration of the nuclei is not always easy, single nuclei remain unchanged in intense diarrhoea, even when there is no disease of the pancreas, and the nuclei may disap- pear as the result of putrefactive changes in the intestine, particularly if the material is retained in the bowel more than thirty hours. Schmidt therefore insists that all, General Symptomatology and Diagnosis 331 or at least most, of the nuclei should be preserved un- changed before it is inferred that the pancreas is diseased. Partial extirpation of the pancreas was found not to pro- duce diagnostic changes, and it is therefore improbable that the method is of great value in any but the most advanced and serious pancreatic lesions. The fasces are normally neutral or faintly alkaline to litmus, but in many cases of pancreatic disease, we have found that they were distinctly acid. This alteration of reaction, although by no means constant or pathogno- monic of pancreatic lesions, may sometimes serve as a confirmatory sign. The specimen should be examined as fresh as possible, and the sample to be tested should be taken from the centre of the faecal mass, not from the surface. It can be applied to moistened litmus paper on a glass slide, which is then examined on the reverse side, or a fragment may be added to a little neutral lit- mus contained in a test-tube, which is centrifugalised, after it has been well shaken. One of the most serious difficulties arising in connection with the diagnosis of diseases of the pancreas is the differ- entiation of the jaundice due to cancer of the head of the gland from that occurring in chronic pancreatitis and common-duct cholelithiasis. A chemical examination of the faeces for stercohilin may, in many instances, afford very considerable assistance and supply valuable confirmatory evidence. The method of investigation we have employed has been described on page 213. Ob- struction of the common duct in most cases of cancer of the head of the pancreas, at the time they usually- come under observation, is, we have found, generally complete, or almost complete, and the fseces are therefore free from stercobilin or contain but faint traces. In common-duct cholelithiasis and chronic pancreatitis, on the other hand, it has been our experience that the obstruction is rarely absolute, so that the faeces give a 332 The Pancreas: Its Surgery and Pathology distinct, although often subnormal, reaction for sterco- bilin. In the table on page 214 it will be seen that twenty-two out of twenty-four cases of malignant disease of the pancreas gave no stercobilin reaction, in two traces were found, and in one only was there a well-marked reaction. In eighteen cases of chronic pancreatitis, with jaundice and obstruction of the common duct, there was a well-marked reaction for stercobilin in all but six, and in these six traces were found. Stercobilin was also present in the faces of eight cases of common-duct chole- lithiasis with jaundice but no affection of the pancreas, in considerable amounts in five, and in small quantities in three. Blood may be noticed in the motions occasionally, but it is not a regular symptom until the hemorrhagic ten- dency occurs later in the disease, or unless there happens to be a malignant growth ulcerating into the intestine. Abscesses of the pancreas and pancreatic cysts have been known to rupture into the bowel and their characteristic contents have been found in the motions, or even in the vomited matter, as occurred in one of our cases. A necrotic pancreas has been passed through the intestine, and cases have been reported by Leichtenstern and Min- nich in which pancreatic calculi have been passed per anum, {d) Sialorrhosa Pancreatica. — An increased flow of saliva has been noted by some observers in disease of the pancreas, and particularly in cases of pancreatic calculi (Holzmann, Capparelli, and Guidiceandra) and cysts of the pancreas (Battersby and Ludolph). It has been sup- posed to be due to a reflex excitation of the salivary glands. The occasional association of pancreatitis with parotitis suggests that there is some obscure connection between the buccal and abdominal "salivary glands," but excessive salivation is such a very rare occurrence in pancreatic disease that it is possibly accidental and General Symptomatology and Diagnosis 333 cannot be relied upon as an aid to diagnosis. We have observed it in two at least of the cases that have come under our notice, and have recently met with a third in which it was one of the symptoms that most seriously troubled the patient. In this instance the salivation ceased in a most striking manner within forty-eight hours after he had been operated on for chronic pancreatitis. 3. Metabolic Symptoms. — (a) Diabetes and Glycosuria. — Glycosuria is by no means a common symptom of pan- creatic disease, and cannot be relied upon as a diagnostic symptom. When present it indicates a serious, although not necessarily hopeless, condition. The relation of diabetes to disease of the pancreas has been discussed in the chapter on diabetes, and it will be recollected that both experimental and clinical observations have shown that glycosuria occurs only when the greater part of the pancreas has been removed or destroyed. The appear- ance of sugar in the urine, along with other signs of disease of the pancreas, therefore points to a wide-spread and advanced lesion. This is particularly the case in chronic interstitial pancreatitis of the interlobular type, which is the form that follows obstruction of the ducts by calculi, and is produced by ascending catarrhal inflammations from the duodenum. Interacinar pancreatitis gives rise to glycosuria at an earlier stage, but appears to be a much less common disease than the interlobular variety. Malignant disease of the pancreas rarely gives rise to diabetes, and then only, as a rule, when the whole organ has been destroyed by the growth. [b) Maltosuria and Pentosuria. — The sugar met with in diseases of the pancreas is usually dextrose, but occa- sionally maltose has been found, and very rarely there would appear to be pentosuria. Neither maltosuria nor pentosuria can be regarded as pathognomonic of pancreatic diseases, and they are of such rare occurrence as to be of no practical importance. 334 The Pancreas: Its Surgery and Pathology (c) The ''Pancreatic'' (Cammidge's) Reaction in the Urine.— The original method of performing this reaction, as described by one of us in 1904 (see page 245), has to a large extent been superceded by the improved method (page 252). The former, in our hands, had proved ex- ceedingly useful in many anxious and doubtful cases, but, as the interpretation of the results it yielded were largely dependent upon the experience of the observer, it was difficult for those who had not the opportunity of frequently performing the test to satisfactorily apply it in practice. The improved method has, we hope, over- come this difficulty, and also removed some of the possible sources of manipulative error in inexperienced hands. We have regularly employed the improved reaction since the early part of 1905, and, as the table on page 255 shows, have found that a positive reaction may be expected in all cases where there are active inflammatory changes in the pancreas. Acute pancreatitis can thus be distin- guished from intestinal obstruction, and other conditions with which it is liable to be confused, and chronic pan- creatitis, associated with obstruction of the common duct by gall-stones, or secondary to duodenal catarrh, can be differentiated from simple cholelithiasis and jaundice, etc., for which a distinct method of treatment may be required. It is always advisable to control the urine examination by an investigation of the fasces, for if the results agree the chances of a mistaken opinion are con- siderably reduced, and are probably very small, if the analyses have been conducted by a competent observer. A chemical examination of the fasces is particularly useful in suspected cases of malignant disease of the pancreas, for, although no reaction is obtained by the improved method in about 75 per cent, of such cases, a crystalline deposit, indicating an associated inflammatory lesion, is met with in the remaining 25 per cent. Clinically it is often a matter of extreme difficulty to General Symptomatology and Diagnosis 335 differentiate chronic inflammation of the pancreas from cancer, but if the results of a complete examination of the urine and faeces, carefully and conscientiously per- formed, are considered in conjunction with the history and symptoms, the chance of an erroneous diagnosis is materially reduced, even in those cases where the spread of the growth is giving rise to secondary inflammatory changes in the adjacent gland tissue. In our experience a characteristic "pancreatic" reaction in the urine has always been associated with evidence of disease of the pancreas at operation, or post-mortem, in all cases where it has been possible to investigate the condition of the gland, and information kindly supplied to one of us by others regarding cases examined for them confirms our opinion of the clinical value of the test. Confirmatory evidence is also afforded by the way in which the reaction disappears in patients who have suffered from pancrea- titis, after steps have been taken to deal with the condi- tion by operative means, and the uniform manner in which gall-stones in a common duct passing through the head of the pancreas have been associated with a positive reaction in the urine, whereas when the duct has passed behind the pancreas the presence of calculi has not given rise to any reaction. A striking demonstration of the diagnostic value of the test was afforded by a case already referred to in the chapter on diabetes. The urine from this patient, who suffered from a duodenal growth, gave no reaction when first examined, but later gave a well- marked result, which was shown at operation to be due to an invasion of the pancreas by the growth. {d) Test for Fat-splitting Ferment in the Urine (Opie) . — This test has only been employed by Opie in one case of acute haemorrhagic pancreatitis, and there he obtained evidence that the urine contained a fat-splitting ferment. We have had no experience with it in acute pancreatitis, and are not acquainted with any published accounts of 336 The Pancreas: Its Surgery and Pathology cases in which it has been tried. If future experience should confirm Opie's observation, and show that fat- splitting ferments are constantly present in acute pan- creatitis, a most useful addition will have been made to our means of diagnosing the condition. (e) Indicanuria. — An increased excretion of indican and ethereal sulphates in the urine has been considered, by some writers, as an indication of disease of the pan- creas. There are others, however, who have advanced reasons why the excretion of these substances should be diminished. There is no doubt that in many cases an abnormal amount of indican is found in the urine, but in our experience there is just as frequently no excess, and it is probable that the condition of the urine with regard to this substance is dependent upon factors of which the condition of the pancreatic secretion is only one, and that not the most important. Indicanuria is not, therefore, of any great value in the diagnosis of diseases of the pancreas, and the same may be said of an increase in the proportion of ethereal sulphates. (/) Rediiction in the Excretion of Phosphates. — A reduced amount of phosphates in the urine when on a milk diet, owing to non-splitting of the nuclein constituent of the caseinogen, has been noted in cases of pancreatic disease associated with a diminution or absence of the secretion. In order that this test may be applied it is necessary that the patient should be placed upon a milk diet for several days, and that a regular estimation of the phos- phates in the total excretion of each twenty-four hours should be made. Since the excretion of phosphates is also diminished in pneumonia and other acute febrile diseases, in chronic and acute nephritis, and in gout and during pregnancy, these must first be excluded, and it is advisable . that the diminution should be shown to be dependent upon failure of the pancreatic secretion by watching the General Symptomatology and Diagnosis 337 effects produced by the administration of preparations of pancreas with the food. (g) Oxaluria. — A well-marked deposit of calcium oxa- late crystals has been found in 63 yev cent, of our cases of chronic pancreatitis, and, although we do not lay much stress upon it as an aid to diagnosis, such a deposit may be regarded as tending to confirm evidence obtained by other means. Qi) Lipuria. — The presence of fat globules in the urine has been noted in a few cases of pancreatic disease. It is such a rare occurrence, however, and may be due to so many different causes, that it is to be regarded rather as a curiosity than as a diagnostic sign of any practical value. {i) Fat Necrosis. — The recognition of fat necrosis by the surgeon who opens the abdomen to relieve symptoms associated with peritonitis in its upper part, is of the utmost importance, as, in practically all cases, it may be taken to indicate a grave lesion of the pancreas, probably haemorrhagic, gangrenous, or suppurative inflammation. It is said not to occur generally with suppurative inflam- mation, but in a case, to be referred to subsequently, most extensive fat necrosis was found with a subdiaphragmatic abscess of pancreatic origin. It has also been said that the presence of extensive fat necrosis is a fatal sign, but this is not invariably so, for Truhart has collected ten cases in which the diagnosis was made, and yet an immediately fatal issue did not follow, and we have had the opportunity of observing one case in which complete recovery took place after operation in a patient with acute pancreatitis in whom fat necrosis was well-marked and diffuse. 4. Special Symptoms Obtained by Artificial Means.— (a) Alimentary Glycosuria. — If 100 grams of grape-sugar be given in a quarter of a litre of water or tea to a normal individual in the morning, fasting, an examination of the urine two or three hours subsequently will show that it has not given rise to glycosuria, but if, for any reason. 7,2)8 The Pancreas: Its Surgery and Pathology the metabolic functions of the pancreas are at fault a more or less marked reaction for sugar will be obtained. The production of alimentary glycosuria in this way does not, however, necessarily indicate that there is a gross lesion of the pancreas, for it may result from toxic changes such as are probably present in acute febrile diseases and alcoholism, and a similar result may be obtained in cases of neurasthenia, traumatic necrosis, acute diseases of the brain and meninges, in many forms of mental debil- ity, especially mania and paralysis, in exophthalmic goitre, and in some diseases of the liver, not necessarily associated with obvious changes in the pancreas. The observations of Wille, however, show that in some 65 per cent, of cases in which alimentary glycosuria occurs a grave pancreatic disease is present. Many investiga- tors have made use of cane-sugar in applying this test, but it is not so suitable for the purpose as dextrose, or fruit-sugar, as the results are not so simple and easy to interpret, for the glycosuria following the administration of cane-sugar may be due to anomalies of fermentation and absorption in the intestine. (6) SahWs Test. — This well-known method of diagno- sis depends upon the fact that if iodoform be enclosed in gelatin capsules, hardened in formalin, and given by the mouth, it is almost unaffected by gastric digestion, but is readily dissolved by the pancreatic secretion. If there- fore pancreatic digestion is normal, iodine should appear in the urine and saliva in from four to eight hours; the absence of the reaction, or its delayed appearance, if the motor functions of the stomach be normal, indicates, according to Sahli, an impairment of pancreatic digestion. The great and apparently insurmountable difficulty in this method is to strike a degree of hardness for the cap- sules suitable in all cases. Formalin has the property of making all tissues on which it acts proof against digestion, tryptic as well as peptic, and it is therefore necessary General Symptomatology and Diagnosis 339 that the capsules should only be acted on long enough to protect them against digestion in the stomach, but not for a sufficient length of time to prevent their solution by the more active pancreatic secretions. This in itself is a difficult matter, but when it is remembered that nor- mally there are individual variations in peptic and tryptic digestion the difficulty is still further increased. Ex- perience has proved that these objections are not merely theoretical, for Fromme, Wallenfang, and Sahli himself have stated that, from a retarded reaction, the diagnosis of disturbed pancreatic function cannot always be made with certainty. A prompt reaction, however, appears to exclude any serious lesion of the gland. The results of Sahli 's test have therefore only a negative value in diagnosis. (c) Test Meals. — Since 1879, when Van den Valden drew attention to the fact that free hydrochloric acid was absent from the stomach contents in carcinoma of the pylorus, it has been recognised that a chemical exami- nation of the gastric secretion was of considerable assis- tance in the diagnosis of malignant disease of the stomach. But in April, 1905, Moore, Alexander, Kelly, and Roaf extended this proposition by stating that an examination of the stomach contents, obtained about one hour after the administration of Ewald's test meal of a pint of tea without milk or sugar and a round of dry toast, showed striking diminution or entire absence of the hydrochloric acid normally present after such a meal, in cancer situated in other parts of the body, such as the breast, uterus, tongue, etc. Subsequent observers have not obtained quite such remarkable and constant results as those quoted by Moore and his fellow-workers in their original paper, but there is no doubt that a marked diminution in the amount of hydrochloric acid in the stomach con- tents is frequently met with in such cases, and that, taken in conjunction with other symptoms, it may often be of assistance in diagnosis. 340 The Pancreas: Its Surgery and Pathology The differentiation of malignant disease of the pancreas from chronic pancreatitis is, in some instances, so difficult that any method that holds promise of assistance is worthy of a careful trial. We have only had the opportunity of examining a test meal from three cases of cancer of the pancreas, and have obtained the following results: Case I. (No. 662.) Phloroglucin-vanillin reaction for free HCl , Negative Total acid (as HCl) 0.007% Physiologically active HCl 0.000% Case II. (No. 749.) Phloroglucin-vanillin reaction for free HCl Negative Total acid (as HCl) 0.05% Physiologically active HCl 0.00% Case III. (No. 780.) Phloroglucin-vanillin reaction for free HCl Negative Total acid (as HCl) 0.009% Physiologically active HCl 0.000% It will thus be seen that not only was there absence of free hydrochloric acid in all these cases, but no evidence of physiologically active hydrochloric acid (as estimated by Willcox's method) could be found. For the sake of comparison we may quote another case, which, although the results of the "pancreatic reaction" pointed to sim- ple inflammation, was so much like malignant disease, both clinically and at operation, that a guarded prognosis was given. The patient, however, rapidly improved after the operation and is now quite well : Case IV. (No. 639.) Phloroglucin-vanillin reaction for free HCl Deep crimson Total acid (as HCl) 0.13% Physiologically active HCl 0.13% In this instance there was an abundance of free hydro- chloric acid, and the percentage of physiologically active General Symptomatology and Diagnosis 341 acid was not suVjnormal, in marked contrast to their entire absence in the other three cases. Our experience is as yet too small to permit of any dogmatic statement being made, but it is sufficient to show that the results of a chemical examination of a test meal, when taken in con- junction with other evidence, may prove of considerable help in diagnosis. With such a number of signs and symptoms as those above enumerated it is difficult to understand how the idea has gained so firm a hold that disease of the pancreas is, as a rule, unrecognisable during life. Although, in any particular case, one may not meet with all of them, there is usually such a combination that, with care, no difficulty need be experienced in arriving at a satisfactory conclusion. Different diseases of the pancreas, as one would expect, present very various groupings of symp- toms, but in all digestive, metabolic, and physical signs can be found that will indicate the true source of those alterations in well-being of which the patient complains. In every case the past history of the patient should be carefully gone into for evidence of chronic dyspepsia, gall-stone attacks, and recent infectious diseases, such as typhoid fever, influenza, etc., all of which may be followed by diseases of the pancreas. Then, having investigated the present condition of the case as regards loss of weight, alterations of appetite, especially an objection to meat and fat, nausea, vomiting, and pain in the upper abdominal region radiating under the left scapula, the presence or absence of jaundice should be noticed, particularly in the sclerotics, and the pancreatic region should be examined for a tumour and for tenderness on pressure. It should now be possible to determine whether the pancreas is probably diseased or not, and to decide whether the assistance of the clinical pathologist is required to further investigate the case. In all cases of suspected pancreatic trouble it is advis- able that a complete chemical and microscopical ex- 342 The Pancreas: Its Surgery and Pathology amination of the faeces and urine should be made by a competent observer, for the results of such examina- tions will, in the large majority of cases, clinch the diagnosis. The fseces should be examined as fresh as possible, for their reaction may quickly change, and the presence of an acid reaction is suggestive of pancreatic trouble. An excess of unabsorbed fat in the motions points to disease of the pancreas, particularly if this is chiefly due to a high proportion of neutral fat. Azotor- rhoea, along with steatorrhoea, tends still further to con- firm the diagnosis. Complete absence of stercobilin from the f^ces is suggestive of malignant disease of the pan- creas, while its presence points to obstruction of the com- mon duct by gall-stones. In investigating the urine a thorough examination should be made for albumin, sugar, acetone bodies, indican, bile, and urobilin; the urea, chlorides, and phos- phates should be estimated quantitatively; and the centrifugalised deposit examined for calcium oxalate crystals. Sahli's test and Opie's test for fat-splitting ferments may be tried, and in all cases it is advisable to perform Cammidge's "pancreatic" reaction. The presence of sugar in the urine, along with evidence of disease of the pancreas, is of great value, and, even when sugar is absent naturally, its discovery after the administration of a test dose of loo grams of glucose tends to confirm the diagnosis of disease of the pan- creas. The acetone-bodies point to abnormal tissue waste, such as is met with in serious cases of pancreatitis and cancer of the pancreas, and an excessive amount of in- dican, although not diagnostic of disease of the pancreas, may indicate the site of the infection from which a chronic inflammation has arisen. Bile-pigments in the urine show that there is obstruction of the biliary passages, and a pathological excess of urobilin points to catarrh of the bile-duct, and possibly of the pancreatic duct, which may or may not be associated with a floating biliary General Symptomatology and Diagnosis 343 calculus. A diminished excretion of phosphates, when the patient is on a milk diet, is indicative of a pancreatic lesion, and a diminution of chlorides relative to urea is met with in serous effusions such as occur in malignant disease. A well-marked deposit of oxalate crystals tends to confirm a diagnosis of chronic pancreatitis. Sahli's test, when negative, is strongly against there being serious disturbance of the functions of the pancreas, and the presence of fat-splitting ferment in the urine probably indicates acute pancreatitis. We have found the "pancreatic" reaction of very great assistance in all cases of suspected pancreatic disease, but we are not prepared to contend that it is pathognomonic or infallible, for even the most com- monly relied upon tests are, it is well known, liable to prove misleading at times, unless due regard is paid to possible sources of fallacy, and the results are in- terpreted in the light of clinical evidence. As we have pointed out, it is not impossible that inflammatory and degenerative changes in other pentose-containing tissues of the body may at times give rise to a positive " pancre- atic" reaction, but, having regard to the relatively large proportion of this substance contained in the pancreas, disease of that organ is the most likely, and probably the most common, cause of such a result. In making a diagnosis in suspected cases of pancreatic disease, or of jaundice in which it is sought to determine whether there is a gross obstruction to the free flow of bile and pancreatic juice into the intestine, and whether this obstruction, when present, is of a simple or malignant nature, it is important to remember that the "pancreatic" reaction is only one factor that has to be taken into account, and that, although it may afford valuable evidence, for or against, it is necessary that its indication should be con- sidered with the clinical symptoms, and that, whenever possible, they should be checked by a chemical analysis of the faeces. 344 The Pancreas: Its Surgery and Pathology Finally, if there is a suspicion of malignant disease, Ewald's test breakfast may be given, and the stomach contents examined in an hour for free and physiologically active hydrochloric acid. Literature V. Ackeron: Berliner klin. Wochensch., 1889, Nr. 14. Battersby: Gaz. med. de Paris, 1844, pp. 219, 617; Arch. gen. de med., 1844. Biondi: Riforma med., 1896, ii, Nr. 9, p. 97. Bright: Med. Associat. Trans., 1838, p. 18. Bull: New York Med. Journ., 1887, p. 376. Cammidge: Arris and Gale Lect., Lancet, March 19, 1904. Brit. Med. Journ., Oct. 25, 1905. Tr. Roy. Med. Soc. 1906. Capparelli: Arch. ital. de Biol., 1894, xxi, 398. Chari: Wiener med. Wochenschr., 1880, xxx, 139. Cowley: Lond. Med. Journ., 1788. Demme: Wiener med. Blatter, 1884, Nr. 51. Fitz: Congress of Amer. Phys. and Surg., May, 1903. Fles: Holland Arch., 1864, ii, 187. Fromme: Miinch. med. Wochensch., 1901, Nr. 15. Goodman: Phila. Med. Trans., xxii, 6, 1878. Gould: Anat. Museum of Boston. 1847, P- ^47- Gould: Soc. for Med. Inform., 1847, P- 217. Lancet, 1891, ii, 290. Brit. Med. Journ., 1894, i, 1191. Hirschfeld: Zeitsch. f. klin. Med., 1896, xxxi, 212; 189 1, xix, 249. Holzmann: Miinchener med. Wochensch., 1894, Nr. 20. Kuntzmann: Hufeland's Journal, 1820. Ktister: Berliner klin. Woch., 1887, S. 154; Deutsch. med. Woch., 1887, S. 189 u. 216. Leichtenstern : Handb. d. spec. Therap. von Penzoldt-Stintzung, 1896, iv, 203. Lichtheim: Berliner klin. Wochenschr., 1894, Nr. 8. Ludolph: Dissertation, 1890. Minnich: Berliner klin. Wochenschr., 1894, S. 187. Moore, Alexander: Kelly and Roaf, Lancet, April 29, 1905, p. 1120. Le Nobel: Maly's Jahresb., 1886, S. 449. Opie: "Diseases of the Pancreas," 1903. Oser: Nothnagel's "Encycl. of Pract. Med." ("Dis. of Pancreas and Liver"), 1903. Reeves: Monthly Journal, March, 1854. Robson, Mayo: Hunterian Lecture, 1904. Address before Canadian Med. Assoc, 1904. Polyclinic Lecture, Lancet, 1900. Address before Amer. Surg. Assoc, 1902. Sahli: Deutsch. med. Wochenschr., 1899, Nr. i. Deutsch Arch. f. klin. Med., 1904, Ixi, 383. Berliner klin. Wochenschr., 1902, Nr. 16 u. 17. Schmidt: "Test Diet in Intestinal Disease." Silver: Trans. Path. Soc. of London, 1873, xxiv. Truhart: Pankreas-Pathologie, Wiesbaden, 1902. Wallenfang: Inaug. Dissert., Bonn, 1903. Willcox: Lancet, June 10, 1905, p. 1566. Wille: Deutsch. Arch. f. klin. Med., 1899, Ixiii, 546. Ziehl: Deutsch. med. Wochenschr., 1883 Nr. 37. CHAPTER XIII INJURIES The pancreas is more securely protected from direct violence than almost any other abdominal organ. In- juries are therefore comparatively rare, and, when they do occur, are almost invariably accompanied by more or less damage of other viscera. Garre was only able to Inferior vena cava Aorta Stomach Left adrenal Right adrenal Right kidnej' Kidney Colon Fig. 115.- — Diagram showing the relations of the pancreas. meet with thirty recorded cases, and in but eight was the pancreas the only organ injured. Recovery occurred in three cases. In addition to the anterior wall of the abdomen, there lie in front of the pancreas the transverse 345 346 The Pancreas: Its Surgery and Pathology colon, the stomach, and the omentum, while behind are the aorta and inferior vena cava, the second lumbar verte- bra, with the adjoining portions of the first and second vertebrae in the middle line, and the psoas and the quad- ratus lumborum muscles and the thick mass of the erector spinas, with many fascial strata, laterally. The pancreas thus occupies almost the centre of the body, and any harmful influence coming from without must first encoun- ter other more superficial structures. Injuries of the pancreas may be divided into: (i) Lac- erations due to direct violence; (2) bullet wounds ; (3) penetrating wounds and stabs. (I) LACERATIONS DUE TO DIRECT VIOLENCE In the majority of these cases the force producing the injury has been directed from before backwards at the epigastrium, and has been of such severity that other organs within the ab dominal cavity have been damaged at the same time. A rent in the liver, a tear of the kidney or spleen, a rupture of the duodeno- jejunal flexure or duode- num, a laceration of the stomach, or extensive in- jury to the peritoneum have all been observed. In most of the earlier re- corded cases the injury to the pancreas was only recognised after death, and was accompanied by one or other of the lesions mentioned, but even in some of these the injury to the pancreas was so much greater Fig. 116. — Traumatic rupture of the pancreas and bruising of the duo- denum, followed by hasmatemesis and fat necrosis (Santos). Injuries 347 than that inflicted on other organs that it was proVjably the cause of the fatal issue. The number of recorded examples of injuries of the pancreas is probably not a correct indication of the frequency with which they occur, for when an injury of the gland forms only a part of a multijjle visceral disorganisation due to violence it is apt Fig. iiy.- — Showing the rupture in the duodenum. The ends have been filled with cotton-wool and separated from one another for the pur- pose of demonstration. to be overlooked, even after death, as the hsemorrhage from a ruptured liver, spleen, or kidney is so profuse as to rapidly fill the peritoneal cavity. Laceration of the pancreas is most commonly due to the patient being run over, and fatal cases arising from this cause have been reported by Travers, Stoerk, Cooper, 348 The Pancreas: Its Surgery and Pathology Pressel, and Hale White. In all of these the injury of the pancreas was accompanied by fracture or laceration of ribs, liver, kidneys, or other organs. In a fatal case recorded by Wilks and Moxon, however, where the pan- creas was so crushed opposite the spinal column as to be Fig. ii8. — Shows the rupture in the pancreas and its peritoneal cover- ing. divided into two parts, the laceration was unaccompanied by other abdominal injuries, and the specimen shown in Fig. 119, from St. Bartholomew's Hospital Museum, was also taken from a patient in whom post-mortem the pancreas was the only organ found to be injured. In the latter .case the patient had been crushed between two Injuries 349 vans, and on admission only complained of slight pain in the epigastrium. Twenty-four hours later he became collapsed, but recovered. Subsequently he vomited, became seriously collapsed, and died three days after the receipt of the injury. At autopsy a rupture of the pan- creas dividing it into two nearly equal portions was found ; there was fat necrosis in the neighbourhood, but no injury of the duodenum, liver, spleen, kidneys, or other abdominal viscera. Fatal cases of laceration of the pancreas following a kick have been described by Jaun, Uninjured pancreas Spleen Rupture and hsemorrhage into pancreas Fig. 1 19.— -Rupture of the pancreas, without any other abdominal in- jury (St. Bartholomew's Hosp. Museum, 2276 D). Leith, and Groeningen. In Jaun's case there was no other abdominal injury. An injury of the pancreas that caused death on the fourth day was produced, in a case reported by Wagstaff , by a fall from a cart on to the left side. Post-mortem the other abdominal viscera were found to be uninjured. Goldmann, Villiers, and Hale White have reported instances in which fatal injury of the pancreas, accompanied by more or less extensive damage of other abdominal organs, has followed a blow in the upper epigastric region. In most instances the 350 The Pancreas: Its Surgery and Pathology chief injury has been opposite the spinal column, the pancreas being probably caught and compressed thereon by the crushing force. Although injury of the pancreas is usually brought about by severe violence, the tissues of the gland are so soft and easily bruised that slight injuries have more effect upon it than upon firmer organs. The following case, observed by one of us, illustrates the serious results that may follow from a comparatively slight blow over the region of the pancreas : A butler slipped and fell forward against a knifeboard projecting from the end of the table at which he was working. The blow was comparatively slight and the man did not even fall to the ground. Pancreatitis fol- lowed on what was, at the beginning, probably a mere bruising of the pancreas, but which was succeeded by slight bleeding into the gland, and, this effusion becoming in- fected, acute haemorrhagic pancreatitis resulted. An exploration for the cause of the pancreatitis resulted in the discovery of a large collection of highly blood-stained fluid in the lesser peritoneal sac, some of which had burst through a small laceration in the omentum into the greater peritoneal sac. There was general periton- itis present at the time of operation, and though drainage was freely adopted, both from the front and back, the patient did not survive many hours. A case of considerable interest and importance has been described^ in which pressure from a tourniquet caused severe bruising of the pancreas. The patient had an aneurysm of the abdominal aorta, not far from its bifurcation, and an attempt was made to treat it by the application of a tourniquet nearer the heart. In the few hours during which the patient survived the application of the instrument, no symptoms referable to the pancreas were observed, but post-mortem the gland was found to be much bruised where it lay across the vertebral column. ^ Lancet, Feb. 4. 1905. Injuries 351 In a second class of injuries due to crushes and blows, although the pancreatic lesion is less profound than in those just referred to, it is the main result of the injury, and slowly gives rise to symptoms that may be relieved by operation. After the shock of the accident has passed off, the patient may appear to have quite recovered, but in a longer or shorter time, varying from a few days to several weeks, an abdominal tumour appears, which on being explored proves to be a distension of the lesser sac by blood-stained fluid. The first case of this kind was recorded by Kiilenkampff , and others have been reported by Senn, Kiister, Karewski, etc. Cases following a blow have been described by Ross, Hadra, Lloyd, Randall, Karewski, and one of us, by Littlewood after a kick, by W. H. Brown after a crush, and by Cathcart and Sheen after the patients had been run over. Coombs and Nash have tabulated the records of twenty-five cases, including a few in which the swelling has followed vomiting or some other form of straining. " The sequence of events in such cases is probably that the traumatism causes a laceration of the posterior layer of the lesser sac and of the pancreas, to which it is intimately adherent. Blood, and possibly some pancreatic secretion, are then poured into the lesser sac and peritonitis results. The foramen of Winslow is sealed by adhesion, and the lesser cavity of the peritoneum, now a closed sac, is distended with serous fluid mixed with blood and pancreatic secretion. When the fluid is evacuated the pancreas continues to pour its secretion into the lesser sac through the rent in its peritoneal invest- ment. ' ' (Mayo Robson and Moynihan, p. 5 1 . ) The rela- tion of these tumours to the pancreas, and the presence in their contents, and in the fluid issuing from the drainage- tube, of ferments, has led to their being generally spoken of as pancreatic cysts, but they are now regarded as one of the varieties of pseudo-cysts of the pancreas, and will be considered in that connection in a subsequent chapter. 352 The Pancreas: Its Surgery and Pathology Symptoms and Diagnosis. — In the first class of cases, where the injury to the pancreas is great, but, as a rule, only constitutes one of the results of the accident, there has been profound collapse coming on rapidly or, as in the cases reported by Jaun, Wagstaff, and Leith, after an interval of an hour or more. The immediate symp- toms of shock are due to the hemorrhage, but when they appear after a latent interval are to be ascribed to the complications arising from the injury to the pancreas. According to Leith, the absence of all external signs of injury of the abdomen is surprising and noteworthy. The absence of any definite signs pointing to pancreatic injury renders an accurate diagnosis impossible in such cases, and it has usually only been at the post-mortem examination that the source of the haemorrhage, etc., have been traced. When, however, from the gravity of the collapse, the site of the injury, and the presence of dulness in the right or left flank, it is evident that there is some serious visceral lesion with internal hemorrhage, speedy operation is indicated, and may prove successful in saving the patient's life, even when there is extensive laceration of the pancreas, as in the case recorded by Randall. Where the symptoms of shock are less severe, and the patient recovers, but later develops a cystic swelling in the region of the pancreas, the diagnosis is less difficult, and operation may be undertaken with every prospect of a successful issue. The nature of the contents of such a cystic swelling may afford evidence of its origin from an injury of the pancreas by the discovery of ferments. The fluid may be clear, or turbid from the presence of blood, and is alkaline in reaction. It has a specific gravity of i.oio to 1.0120, and usually contains albumin and nucleo-proteids. Starch-splitting and fat-splitting ferments are present, and can be recognised by the tests described on page 264. As a rule, it has no digestive Injuries 353 power for proteids, for it has not been activated by con- tact with enterokinase, but feeble proteolytic powers may be shown, such as are possessed by the juices of many tissues. The accelerating action of salts of calcium, etc., on the proteolytic activities of pancreatic juice, dem- onstrated by Delezenne, may account for the presence of a more marked digestive power in some instances. Where the swelling forms quickly, direct laceration of the pan- creas is indicated, but where it forms slowly, may be after some months or even years, it is probably due to pancreatitis following the injury, with effusion into the lesser sac. The appearance of a swelling which rapidly reaches a certain bulk, and then remains stationary, suggests the outpouring of fluid into a preformed sac produced by closure of the foramen of Winslow by ad- hesions previous to the injury. The swelling occupies the epigastric, umbilical, and left hypochondriac regions. The stomach and transverse colon can be detected in front, and the descending colon behind and to the left. Treatment. — The immediate shock and collapse con- sequent on the accident having been treated by the means usually adopted in such cases, a satisfactory reac- tion on the part of the patient raises the question of oper- ative interference. Where there is evidence of internal hemorrhage this should be undertaken at once, and an attempt made to secure the bleeding points. Experi- mental work has shown that wounds of the pancreas can be sutured, and that healing speedily takes place, so that if a laceration of the gland be found it should be dealt with by accurately coapting the edges by sutures, care being taken, however, to avoid puncturing the main duct of the gland. The cases in which such surgical intervention is possible are, unfortunately, few, as the injury to the pancreas is most frequently only part of a wide-spread destruction involving the liver, kidneys, spleen, stomach, or intestine. In the case reported by 23 354 The Pancreas: Its Surgery and Pathology Randall, referred to previously, operation six hours after the accident was followed by recovery. The patient had been crushed between the pole of a van and another vehicle in the epigastric region. He had violent pain, became faint and collapsed, and vom- ited. On admission to hospital an hour later, he was found to be still collapsed, with a small weak pulse of 90 and a temperature of 95° F. He was treated with stimulants, warmth, and a hypodermic injection of half a grain of morphin. Midway between the ensiform cartilage and the umbilicus there was a bruise nearly two inches in diameter and very tender to the touch. The abdomen moved with respiration, but was very tender, especially above the navel. The liver dulness was nor- mal, but there was marked dulness in the right flank. The urine showed no abnormality. At the operation no gas was found in the peritoneum, but there was much clotted and fluid blood. No lesion of the liver, stomach, or bowels could be found. A large tear, directed verti- cally, and running from the stomach to the liver, was seen in the gastrohepatic omentum, and there was a tear in the peritoneum over the pancreas. Under this was a tear in the body of the gland two inches long, running from the right and below tipwards and to the left, leaving a loose, tongue-like process of gland substance, the base of which was half the width of the organ. On passing a finger through the tear it came directly upon the aorta. There was free oozing, but no large blood-vessels could be found. The rent in the pancreas was sewn up by four silk stitches, and, after the lesser peritoneal sac had been cleansed, the greater part of the tear in the gastro-hepatic omentum was sewn up with a continuous catgut suture, space being left for drainage. The abdomen having been thoroughly flushed with hot saline solution, an iodoform gauze packing was inserted down to the pancreas and the rest of the abdominal wound closed. For the first three days the patient was constantly sick and in much pain. He later developed delusions, but slowly recovered, gaining in weight and strength, and was discharged, ten weeks after the operation, with the wound healed. On one occasion the urine contained a small quantity of Injuries 355 sugar, but it quickly disappeared. No special features were noticed in the motions. The discharge from the wound consisted of a viscid, slightly turbid fluid, which caused excoriation of the margins of the wound. A fortnight after the operation it was noticed that the skin over the abdomen and lower part of the thorax resembled that produced by an extensi\'e burn of the first and third degrees. A somewhat similar case in which recovery followed prompt operation has been reported by Karewski. The patient had been run over after being struck in the abdo- men by the shaft of a vehicle. He was able to walk home, but abdominal pain coming on shortly afterwards an exploratory laparotomy was performed. There was a large quantity of blood, especially in the region of the gastrocolic ligament, and the head of the pancreas was found to be crushed. The patient lost 400 grams of pure pancreatic juice daily through the fistula and diminished in weight, but the secretion diminished on a fatty anti- diabetic diet and the fistula was eventually closed. The treatment of the cystic swellings following injury of the gland will be discussed under Cysts of the Pancreas. (2) BULLET WOUNDS Cases of bullet wounds of the pancreas have been described by Otis, Sanitas, Niemann, Bertram, Von Bra- mann, Hahn, Nini, Borchardt, Simmonds, Mann, Korte, Slavsky, Carnell, Jephson, Becker, and Kindt. Rarely the pancreas has been the only abdominal organ injured, but in the majority of instances other organs have also been involved in the injury. Otis, in his surgical report on the American Civil War, relates three cases. In one the pancreas and spleen were both lacerated, and the splenic artery was divided. The patient lived a month. In the second the patient lived fifteen days, and the sto- mach, as well as the pancreas, was found to have been 356 The Pancreas: Its Surgery and Pathology wounded. In the third the lung, the liver, and the pan- creas were all injured. The patient lived but twelve days. The late President McKinley was wounded in the stomach, pancreas, and left kidney. Ninni reports a case of revolver wound of the abdomen in which there was a wound of the pancreas, six of the small intestine, and one in the colon at the hepatic flexure, but in which the patient recovered after operation, and left the hospital thirty-five days after admission. In a case of revolver wound of the pancreas, associated with double perfora- tion of the stomach, recorded by Kindt, widely dissemi- nated fat necrosis was found post-mortem, and a similar condition has also been noted in two other cases after death, but has not been recorded as present in any of the cases at operation. The organ most commonly injured with the pancreas has been the stomach (nine cases), then the liver (seven cases), the lesser omentum (four cases), the diaphragm (three cases), the spleen, small intestine, and large intestine each in two, and the lung, kidney, heart, and portal vein each in one instance. Symptoms. — There are no pathognomonic signs of injury of the pancreas in bullet wounds of the abdomen, and even suggestive symptoms are usually absent. The probable course of the bullet, as indicated by the site of entry and exit, is usually the only guide. When the abdomen is opened and neighbouring viscera are found to be wounded, particularly the posterior wall of the stomach and lesser omentum, it is essential that a careful search should be made for any injury of the pancreas. Treatment. — Operation should be undertaken as speed- ily as possible in all cases. Any bleeding points should be secured, and a careful but rapid search made for injury to the stomach, intestine, liver, etc. The wound of the pancreas may be sutured, but, if there is much laceration, it may be necessary to resect a portion of the gland and unite the clean-cut edges by sutures. Care Injuries 357 must be taken to avoid the main duct, the superior mesenteric artery, and the portal vein. Complete dis- organization of the gland can only be treated by plugging and drainage, for it is practically impossible to remove it, and the attempt is not justifiable on physiological grounds. Where suture is possible drainage should always be adopted, for there is invariably a certain amount of leak- age, and if an exit is not provided for the exuding secretion, local disturbances and peritonitis may result. It is note- worthy that in two cases where an injury of the pancreas was sutured, but no drainage was provided, a localised destruction of tissue was foimd post-mortem. Drainage has usually been provided through the abdominal wound, but a posterior opening, such as Jephson adopted in his case, and was also carried out by one of us in another instance, is probably more efficient. It is frequently stated that wounds of the pancreas are almost always fatal, but this is not necessarily the case, if suitable operative measures are quickly taken. Of the twenty- one cases of injury of the pancreas due to gunshot wounds of which we have found records, fifteen were operated on and nine of these recovered (Bramann — two, Hahn, Nini, Borchardt, Slavsky, Jephson, Otis, and Becker). Of the six in which death occurred, the injury of the pancreas was not discovered in three, so that in nine out of twelve instances it may be considered that the operation saved the patient's life, for all but one of the cases in which operation was not resorted to died. (3) PENETRATING WOUNDS Penetrating wounds due to stabs or cuts with a knife or bayonet have been reported by Kleburg, Laborderie, Caldwell, Dargau, and Kiittner. In the cases described by all but Kiittner, the pancreas protruded through the wound, and was either returned or the projecting portion removed. Recovery followed in all. In Ktittner's case 358 The Pancreas: Its Surgery and Pathology there was an abdominal wound, 14 cm. long, through which protruded the stomach, transverse colon, and several loops of small intestine. At the operation, a quarter of an hour after the receipt of the injury, the anterior wall of the stomach was found to have been injured, the left border of the liver was notched, the lesser omentum was cut, and the pancreas had been transfixed just to the left of the tuber omentale. Venous and arterial blood were welling up from the wound in the pancreas, but this was checked by two deep and one superficial catgut suture in the parenchyma of the gland. The bursa omentalis was plugged and the stomach wound sutured. The patient recovered, although, at first, he was gravely ill with symptoms of threatened collapse and subphrenic abscess. Treatment. — The treatment of penetrating wounds of the pancreas is exactly on the same lines as those described for gunshot wounds, and what has been said under that heading equally applies here. SEQUELS The sequels of injuries of the pancreas may be seen in the gland itself, or are shown by changes in the neigh- bouring tissues. The effects produced by injury of the pancreas, with closure of the foramen of Winslow, and the conversion of the lesser peritoneal sac into a cystic cavity, have already been referred to. The injury may also produce a true cyst of the pancreas, either from bruising and tearing of the duct, causing stenosis and an accumulation of secretion behind the point of injury, or the duct may be compressed and distorted by the scar tissue resulting from the injury to the neighbouring gland substance. Cases in which inflammatory changes in the gland have resulted from injury have been recorded by Wandesleben, Rolleston, and others. In Rolleston's case an abscess in the head of the pancreas, with fat Injuries 359 necrosis in the subperitoneal cavity, was found at the post-mortem, eighty days after a blow in the abdomen which had given rise to pain, vomiting, inaction of the bowels, and collapse simulating intestinal obstruction. Hansemann, Fitz, and Prince have described cases of necrosis of the pancreas after injury. A wound of the abdominal wall has, in some cases, been followed by pro- trusion of the pancreas. The possibility of such an occurrence has been doubted, but there is now indisput- able evidence that such a complication may occur. The original case of "prolapse of the pancreas" recorded by Laborderie was proved by microscopical examination of the protruding tissue to be in reality a case of prolapsed omentum, but well authenticated cases have been since reported by Otis (two), Kleberg, Caldwell, Dargau, Allen, Thompson, and Pereira-Guimaraes. The failure of Labor- derie to recognise the nature of the tissue in his case by naked-eye inspection emphasises the necessity of a care- ful microscopical examination in every instance. Literature Allen: American Weekly, 1876, p. 305. Becker: Zent. f. Chir., 1905, Nr. 5. Bertram: Inaug. Dissert., Jena, 1893. Borchardt: Berlin, klin. Woch., Jan., 1904, Nr. 3. V. Bramann: Arch. f. klin. Chir., 1899-1900, Ix, 482. Brown, W. H.: Lancet, 1894, i, 21. Caldwell: Transylvan. Journ. of Med., 1828, i, 116. Carnell: Annals of Surgery, xli, 724. Cathcart: Edinburgh Med. Journ., July, 1890. Coombs and Nash: Lancet, 1901, i, 1826. Cooper: Lancet, 1839. Dargau: Med. and Surg. Report., Aug. 22, 1874. Delegenne: Brit. Med. Journ., Dec. 22, 1906, p. 1785. Fitz: Truhart, "Pankreas Path.," S, 332. Goldmann: Quoted by Korte, Deut. Chir., 1870. Groennigen: Rep. Berlin Garrison Hosp., 1890. Hadra: New York Med. Rec, 1896. Amer. Journ. Med. Sci., 1897, i. III. Hahn: Deut. Zeit. f. Chir., Iviii, 1900-1901. Hansemann: Truhart, "Pankreas Path.," 1887, obs. 160. Jaun: Indian Annals of Med. Sci., 1855. Jephson:_ Quoted by Connel, Annals of Surg., xli, 724. Karewski: Deutsche med. Woch., 1890, No. 46. Med. Press, March 6, 1907. 360 The Pancreas: Its Surgery and Pathology Kindt: Gaz. des Hopital, April 4, 1905. Kleburg: Archiv. f. klin. Chir., 1868, S. 523. __ Korte: Verhandl. der Freien. Chir. Verein, xiii, 87. KiilerLkampff : Berliner klin. Wochenschr., 1882, Nr. 7. Kuster: Berlin, klin. Wochenschr., 1887, S. 154. Deut. med. Woch., 1887, S. 189 u. 215. _ Ktittner: Beitr. z. klin. Chir., 1901, xxxii, 244. Laborderie: Gaz. des Hopital, 1856, No. 2 and 9. Leith: Lancet, 1895, i, 770. Edinb. Med. Journ., Nov., 1895. Littlewood: Lancet, 1892, i, 871. Lloyd, Jordan: Lancet, Nov., 1892. Brit. Med. Journ., 1892, ii, 105 1. Mann: American Med., Oct. 19, 1901. Nini: Cent. f. Chir., 1901, No. 41, p. 1024. Otis: "Med. and Surg. History of the War of the Rebellion," ii, 2, 159. Pereira-Guiamaraes : Quoted by Oser, Nothnagel's "Encycl. of Pract. Med.," 1903, p. 269. Prince: Boston Med. and Surg. Journ., 1882, p. 28. Pressel: Inaug. Dissert., Berlin, 1895. Randall: Lancet, Feb., 1905, p. 291. RoUeston: Brit. Med. Journ., 1892, ii, 895. Rose: Deut. Zeitschr. f. Chir., xxxiv, 3, 36. Sanitas: "Rep. of German Army in War," 1870-187 1. Senn: Trans, of Amer. Med. Assoc, 1886. Amer. Journ. of Med. Sci._, 1885. Sheen: Clin. Journ., Nov. 8, 1899; Ibid., March 29, 1905, p. 381. Simmonds: Miin. med. Woch., 1898, No. 6, S. 169. Slavsky: Roussky Vratch, July 31, 1904. Stoerk: "Annus Medicus," 1836. Thompson: Quoted by Senn, "Surgery of the Pancreas," p. 34. Travers: Lancet, 1827. Villiere: Bull. d. 1. Soc. Anat., 1895. Wagstaff : Lancet, 1895, i. Wandesleben: Wochenschr. f. d. gesam. Halk., 1S45. White, Hale: Guy's Hosp. Reports, liv. Wilks and Moxon: "Path. Anat.," second and third edits. CHAPTER XIV INFLAMMATORY AFFECTIONS OF THE PANCREAS Catarrh and Suppurative Catarrh of the Pancreas If we were to base our opinions on the post-mortem records of the past, inflammatory affections of the pan- creas would have to be reckoned among the rarest of dis- eases, but recent clinical observations and operative experience show that such conclusions would be far from accurate, and that inflammatory aft'ections of the pan- Fig. 1 20. — "Enlarged and hard pancreas" — probably a case of chronic pancreatitis (Baillie). creas, or its ducts, are very much more common than is generally supposed. Historical References. — When studying the subject of pancreatitis, in the light of modern pathological know- ledge, it behoves us to bear in mind that the older pathol- ogists had noticed and described the naked-eye appear- 361 362 The Pancreas: Its Surgery and Pathology ances of nearly all the conditions that are engaging so much of our attention at the present time. Tulpius, so far back as 1672, describes a diffuse pancreatitic abscess of pyaemic origin, and Matthew Baillie, physician to St. George's Hospital, in a work on "Morbid Anatomy," published in 1799, describes what he calls a hard pancreas with the lobules distinct, but which is what we now should call a case of chronic interstitial pancreatitis. He also figures in the same volume a case of pancreatic calculi, most carefully dissected, showing the relation of the bile and pancreatic ducts. Portal in 1804 described a case of acute suppurative pancreatitis following on an attack of gout in the feet, and Percival, in 18 18, described a well- marked case of pancreatic abscess associated with jaun- dice. The following is a quotation from a paper by Dr. W. J. Mayo before the American Surgical Association, the executive committee of which approved of it for pub- lication: " Balzer in 1879 first described acute pancrea- titis with fat necrosis. Little attention was attracted to the subject, however, and it was not until Fitz ten years later wrote his classical papers that the medical world really became aware of the inflammatory diseases of the pancreas. Fitz soon after pointed out the fact that many supposed cysts of the pancreas due to trauma- tism were really accumulations of fluid in the lesser cavity of the peritoneum and the omental bursse. A proper understanding of chronic pancreatitis has been largely due to Robson, who first noticed the disease in connexion with his operative work upon the biliary tract. In fact, the surgical study of the inflammatory diseases of the pancreas may be said to be the result of an inquiry into the causation of some of the complications of gall-stone disease." Classification. — Pancreatic inflammations may be catar- rhal, in which the inflammatory trouble is in the ducts, or parenchymatous, in which the substance of the pancreas Inflammatory Affections of the Pancreas 363 is involved. The former resemble the different forms of cholangitis, with which, indeed, they are frequently associated; the latter bear more resemblance to inflam- matory affections of the appendix, " suppurative and gangrenous appendicitis." The following shows the classification at a glance : (A) Catarrhal Inflammations: (a) Simple catarrh | , '^chronic. (b) Suppurative catarrh. (c) Pancreo-lithic catarrh. (B) Parenchymatous Inflammations: Acute : (a) Hsemorrhagic pancreatitis. . Ultra-acute, in w^hich the hasmorrhage precedes the inflammation, the bleed- ing being profuse, and both within and outside the gland. I 2. Acute, in which inflammation precedes I the hasmorrhage, which is less profuse I and is distributed in patches through I the gland. (b) Gangrenous pancreatitis. (c) Suppurative pancreatitis (diffuse suppura- tion) . Subacute : Abscess of the pancreas (not diffuse suppuration) . Chronic : (a) Interstitial pancreatitis. f I . Interlobular, I 2 . Interacinar. (6) Cirrhosis of the pancreas. Etiology. — The etiology of pancreatitis may be classified under predisposing and exciting causes. 364 The Pancreas: Its Surgery and Pathology Among the predisposing causes are : (a) Obstruction in the ducts, the result of gall- stones, duodenal catarrh, pancreatic calculi, cancer of the papilla or of the head of the pan- creas, ulcer of the duodenum, followed by cicatricial stenosis of the papilla, ascarides and lumbrici, etc. (b) Injury either from a bruise, as by manipula- tion in operating, or from a crush, as by a blow in the epigastrium, or from wounding by a sharp instrument. (c) Haemorrhage into the gland. (d) General ailments, such as typhoid fever, in- fluenza, and mumps. (e) Certain anatomical peculiarities in the pancreas or its ducts. (/) Atheroma, or fatty degeneration, of the blood- vessels. Back-pressure from disease of the heart, lungs, etc. (g) New-growth, e. g., cancer or sarcoma. The chief exciting causes are : 1 , Infection conveyed : (a) From the blood, as in syphilis or pyaemia. (6) From the duodenum, as in gall-stone obstruc- tion or gastro-intestinal catarrh. (c) By extension inwards from adjoining organs, as in gastric ulcer or cancer eroding the pan- creas. 2. Irritation, as in alcoholism (doubtful). The anatomy of the pancreas, with its ducts opening into a portion of the intestine never free from organisms, is the key to the etiology of pancreatitis, but even so, were it not that the common bile-duct and the pancreatic duct are so closely related the pancreas would probably generally escape. It is well known that even aseptic Inflammatory xA.ffcctions of the Pancreas 365 ligature of the common bile-duct opens the way to the presence of organisms within the bile-ducts, and we have very definite proof that a gall-stone in the common duct is very shortly followed by infective cholangitis, which may, in unfavourable circumstances, become suppurative cholangitis and lead to abscesses in the liver or to other secondary troubles. But in 28.5 per cent, of cases (Tes- tut) the common bile and pancreatic ducts open together into the ampulla of Vater, which itself opens into the duodenum, and, according to Helly, in 62 per cent, of cases the common bile-duct is intimately embraced by the pancreas, so that when a gall-stone passes down the bile-duct it must, of necessity, in a large proportion of cases, compress the pancreatic duct and cause a damming back of its secretion, which, arguing from analogy as well as from practical experience of the troubles that follow, means damming back an infected secretion. Thus it is brought about that in many cases of common-duct chole- lithiasis, w^here the calculus reaches the pancreatic por- tion of the duct, and remains there for some time, catar- rhal inflammation of the pancreas occurs. If the stone passes after a short period the pancreatitis may subside and leave no trace, or the swelling of the pancreas may persist and, for a time, keep up pressure on the com^mon bile-duct, leading to a persistence of the jaundice, though there is no concretion left to cause obstruction, nor any evidence of disease of the liver beyond the jaundice due to the mechanical obstruction. Thus may be explained some of the cases of chronic jaundice with so-called biliary catarrh. If, however, the gall-stone obstruction persists for some time, and the patient's health is feeble or becomes seriously deteriorated, what was at first merely a simple catarrh may become a suppurative one, and as the same process involves the liver and the pancreas the ducts of b)Oth become filled with pus. We have now suppurative 366 The Pancreas: Its Surgery and Pathology catarrh of the pancreatic ducts associated with a suppura- tive cholangitis, a very serious, and generally a fatal, condition. If the suppurative catarrh persists unrelieved, it may lead, not only to abscesses in the liver, but also to ab- scesses in the pancreas, and possibly, in case of survival, to subacute pancreatitis, as in cases to be described under abscess of the pancreas. If the suppurative catarrh takes on an acute course the condition may become one allied to, and unrecognisable from, pyaemia, as in a case to be related later. If the infective catarrhal condition persists, and does not assume the more dangerous suppurative form, or even if simple obstruction of the pancreatic duct persists from any cause with only mild infection, we may have an almost analogous condition to the one occurring in the liver that produces cirrhosis. In this more chronic form interstitial pancreatitis occurs, which in an early stage may be arrested by the removal of the cause, as will be shown when considering chronic pancreatitis. The chronic pancreatitis is of the interlobular variety, and conse- quently does not involve the islands of Langerhans until a late stage, when the organ may become cirrhotic and diabetes supervenes. If a small gall-stone happens to descend into an unusu- ally large diverticulum of Vater and to lodge there, it will make a through channel from the common bile-duct, as shown in the diagram (Figs. 60 and 61), and this Opie has shown to be a cause of acute pancreatitis, the bile being forced direct into the pancreas. In one case under the care of Dr. Halsted this condition occurred and acute hsemorrhagic pancreatitis ensued. Opie states that he has produced acute hcemorrhagic pancreatitis in dogs by injecting bile into the pancreatic duct. Other irritating substances, suspensions of bacteria, and various acids and alkalies have the same effect, and have been considered in detail in the chapter on pathology. Inflammatory Affections of the Pancreas 367 It is quite clear, therefore, that gall-stones in the com- mon duct are a frequent, in fact, by far the most frequent, cause of the various forms of pancreatitis, but the ana- tomical conditions just mentioned, though evidently potent, are certainly not necessary for the production of acute pancreatitis, which may, as is well known, occur apart from cholelithiasis. Any gall-stone or stones im- pacted in the pancreatic portion of the duct, or even filling the ampulla of Vater, may be efficient causes of the trouble. It may be asked, Why should not every case of common duct cholelithiasis be complicated by pancreatic inflamma- tion ? This is readily explained by the fact that in a cer- tain percentage of cases the common bile-duct and the pancreatic duct open by separate orifices into the duode- num, while in another percentage the duct of Santorini is either the principal outlet for the pancreatic secretion or is of such a size that it can act as an efficient outlet even if Wirsung's duct becomes obstructed. The condi- tion described by Opie, where the ampulla of Vater is very large and a small gall-stone becomes impacted at its orifice, is only rarely found ; otherwise acute pancreatitis would be more common owing to overwhelming of the pancreatic ducts by infected bile. Besides gall-stones the other factors mentioned may lead to obstruction of the pancreatic ducts, to infection of the pent-up secretion, and to the different varieties of pancreatitis, the rationale of the process being similar to the one sketched above. It is possible that infec- tion may extend upwards from the duodenum without preliminary obstruction, apparently by continuity of mucous membrane, catarrhal pancreatitis being then a sequel of gastro-duodenal catarrh. In case of injury, in whatever way inflicted, it seems not unreasonable to think that the soft glandular sub- stance will readily yield and so set free the auto-destruc- 368 The Pancreas: Its Surgery and Pathology tive secretion of the gland, which by dissolving the walls of the blood-vessels will lead to further haemorrhage, and then to the collection of a quantity of easily decom- posable material that only needs infecting to become acutely dangerous. The contiguity of the stomach and intestines furnishes the possibility of infection, though if infection does not take place the injury may be repaired as in other organs. This probably explains acute pan- creatitis supervening not immediately but some days after an injury. Haemorrhage into the pancreas, so-called pancreatic apoplexy, arising from diseased vessels, or in some other way, by disrupting the gland, may lead to pancreatitis, as in the case recently reported to the Societe de Chirurgie by M. Guinard.^ That haemorrhage into the pancreas does not always give rise to pancreatitis is shown by the presence of old blood-stains in cases dying from other causes, and it is reasonable to argue that some other factor is necessary. M. Guinard is of opinion that in his case the mercurial treatment played a part in the etiology of the condition, for, just as mercury produces saliva- tion, it is possible that it may act upon the pancreas in an ^ A man aged thirty-five, for several days had been reUeved by in- jections of the benzoate of mercury for a specific orchitis. After the tenth injection he suddenly felt a sharp stabbing pain in the epigastric region; it was so acute that several hypodermics of morphine failed to give relief. During the following days the patient had fetid diarrhoea, and a srnall tumour appeared in the epigastric region. It was diag- nosed as a gumma, and the mercurial infections were continued ; but the patient continued to have great pain, with continued diarrhoea and complete intolerance for food; he lost flesh to an alarming extent, and became cachectic. When M. Guinard was called in he found him almost moribund. To the left of the Hnea alba, between the umbilicus and the costal margin, there was a tumour of the size of a man's fist, smooth, shining, almost fluctuating, dullish on percussion. An exploratory puncture gave issue to sticky blood. From the sudden onset of the attack, the intense pain, the rapid cachexia, and the absence of fever, M. Guinard diagnosed haemorrhagic pancreatitis, and performed lap- arotoniy. He found a large blackish retrogastric tumour, from which on incision a large quantity of fluid blood mixed with clots escaped. It was a haematic cyst of the pancreas. A drain was put in, and cure was rapid. On awakening from the anesthetic the epigastric pains had disappeared. The contents of the cyst were found to be absolutely aseptic. Inflammatory Affections of the Pancreas 369 analogous way, giving rise to mercurial pancreatism with intense congestion of the gland and interstitial haemor- rhage. It has also to be remembered that the patient was a syphilitic subject. Infection of the disorganised tissues probably plays an important part in the conver- sion of a simple haemorrhage into the acute fulminating inflammatory condition in many instances, and the amount and situation of the blood have also, no doubt, an im- portant influence on the result. In general ailments, such as typhoid fever, influenza, etc., the well-known predilection of typhoid bacilli for the biliary passages would afford an easy explanation of their access to the pancreas, and, though it is difficult to prove, in several cases of catarrhal inflammation of the pancreas a history pointing strongly to influenza and to typhoid fever as the cause has been obtained. In one case the relationship was proved by the discovery of typhoid bacilli. As to mumps and pancreatitis there seems to be some peculiar and intimate relationship between the salivary glands of the mouth and the abdomen, and in the case of a young adult coming under the observation of one of us some years ago, it seemed highly probable that a metastasis occurred about the third day of the disease, when the pain and distress almost completely left the face and were followed by violent epigastric pain and alarming symptoms of depression, accompanied by sick- ness and fever, which then rapidly passed off, after three days' anxiety, and were followed by orchitis. M. Simo- nin gave the result of his observations on 652 cases of mumps treated in the military hospital of Val de Grace. In ten cases, or 1.3 per cent., there were symptoms of pancreatitis which occurred from the first to the twelfth day of the disease and lasted from two to seven days, the principal symptom being epigastric pain and tenderness, with sickness and vomiting. 24 370 The Pancreas: Its Surgery and Pathology Auche has reported two cases of pancreatitis compH- cating mumps. The first was in a lad, aged twelve years, who woke dur- ing the third night of his illness complaining of pain in the epigastrium of a continuous nature, with exacerba- tions ; in half an hour vomiting occurred, at first of food, later of bile. The pain was confined to the epigastrium, midway between the umbilicus and xiphoid cartilage, extending as far as the left costal arch. Owing to the tenderness it was impossible to ascertain if there was any intra-abdominal swelling. During the ensuing day the pain was slightly less, the exacerbations were less frequent. Vomiting occurred four or five times, but only on taking fluids. The bowels acted once; the motion looked as if it did not contain fat. Next day, the fifth since the parotid glands were swollen, the pain was still less, and less frequent, but it was sufficiently severe to prevent deep palpation; a motion passed was normal, as was also the urine. Bilious vomiting continued. Next day, the third since the onset of abdominal symptoms, vomiting ceased, and liquid food was well borne. On the fourth day no swelling could be felt on deep palpation. The temperature had fallen from 38.9° C. to 36.8° C. On the fifth day the patient seemed perfectly recovered. The second case was a boy, aged nine years. On the fifth day of an attack of mumps, suddenly epigastric pain and vomiting supervened. Next day pain and vomiting continued. When these symptoms had lasted three days the patient was seen for the first time; the pain was limited to the left side of the epigastrium, vom- iting had' occurred once during the day, the liver could be felt below the costal arch. Calomel was ordered. The patient was only seen once. A similar case, also in a boy, has been described by Jacob. In this instance pain in the abdomen was com- plained of on the fourth day of the disease, and on examin- ation a tender swelling was found in the epigastrium. The first case in which a post-mortem examination had been made was described by Lemoine and Lapasset in 1905- Inflammatory Affections of the Pancreas 371 In this case a soldier, a native of Algiers, aged nineteen years, was admitted into hospital on OctoVjer 2, 1902, suffering from mumps. The only previous illness was malarial fever. The case ran a benign course, the tem- perature not rising above 100.2° F. and becoming nor- mal on the fifth day. On the eighth day the illness appeared to have terminated, but on the evening of the tenth day there was a rigor with a rise of temperature to 103.8° and pain and swelling in the right testicle. The orchitis rapidly subsided, the temperature became nor- mal on the fourteenth day, and the patient was again pronounced convalescent, but when visited on the morn- ing of the fifteenth day he complained that he had not slept and that he vomited several times during the night. Although the temperature was normal he appeared to be prostrated. The conjunctivae showed a slightly icteric tinge. The upper abdomen was tender, especially in the epigastrium and the region of the gall-bladder. Here the patient had a feeling of weight. The spleen was enlarged and tender. The pulse was slow (52) and the axillary temperature was 97.3°. An aperient was given, but it was vomited and produced no action of the bowels. On the sixteenth day the icteric tint involved the whole skin and the discolouration of the conjunctivae was more marked. Vomiting had become more frequent and the intolerance of the stomach was absolute, all liquids being rejected. The hepatic and splenic regions were very painful, with a maximum in the region of the gall-bladder, which organ appeared to be enlarged. The temperature and pulse remained the same. An injection of 1000 grams of saline solution was ordered in order to provoke ditiresis , as the kidneys had ceased to act since the previous day. In the evening the general condition seemed to have improved. Vomiting had occurred only once during the day. Eighty grams of brownish-red urine, containing I gram of albumin per litre and abundance of biliary pig- ments, had been passed. The patient continued to be very prostrate and spoke of his approaching death. In the night the vomiting recurred and was uncontrollable. At first the vomit was black, then it gradually became sanguineous. There was also constipation. On the seventeenth day the haematemesis was incessant and the 372 The Pancreas: Its Surgery and Pathology patient lost consciousness. The pulse rose to 120 and the temperature to 101.5° ^'^^ the jaundice increased in depth. Death occurred on this day. At the necropsy all the tissues had an icteric tinge. The liver did not appear to be enlarged, but it was much con- gested ; on section it looked like a " cardiac liver." Some lobules seemed to be in a state of incipient degeneration. The gall-bladder was voluminous and oedematous. This oedema extended as far as the beginning of the common bile-duct, where an enlarged gland pressed on the latter and appeared mechanically to produce the oedema of the gall-bladder and the icterus. The bile in the gall-bladder was brownish, thick, and very viscid. The stomach con- tained black fluid ; its mucous membrane was spotted with fine ecchymoses which extended as far as the first part of the duodenum. The pancreas was greatly enlarged, oedem- atous, and congested. It weighed 190 grams and was of a reddish-gray colour. All the region around the pancreas, the duodenum, and the hilum of the liver contained a large number of swollen lymphatic glands. The spleen was enlarged and weighed 1200 grams. The kidneys were a little congested and the capsules were adherent in places. Microscopic examination of the liver showed proliferation of the connective tissue surrounding the biliary canaliculi and a number of nodules composed of embryonic cells. The spleen also contained nodules of embryonic cells and its capsule and trabecule were thick- ened. In the kidneys lesions were limited to the convo- luted tubes, the cells of which showed signs of granulo- f atty degeneration and the lumen of which was filled with epithelial debris. The cells and acini of the pancreas were abnoiTnally large, but the islands of Langerhans were diminished, being compressed by the turgescent tubes. The nuclei of the pancreatic cells stained badly and many of them were vesicular. In this instance the delayed onset of the pancreatic symptoms is remarkable, for they did not occur until the fifteenth day, whereas in most instances they appear to occur about the third or fourth of the illness. In addi- tion to the epigastric pain, which was the principal symp- Inflammatory Affections of the Pancreas 373 torn of the pancreatitis, the gall-bladder was tender and the spleen was tender and enlarged. There was also grave icterus with hasmatemesis, prostration, and sub- normal temperature. An outbreak of epidemic parotitis, accompanied in four cases by symptoms suggestive of metastatic pancreatitis, has been described by Dr. Edgecombe, of Harrogate. The urines from two cases in this epidemic were examined by one of us. In the first no striking abnormality was found and the "pancreatic" reaction was negative. We are informed by Dr. Edgecombe that it was a simple case of uncomplicated parotitis. In the second the urine was acid in reaction, specific gravity 1.030, no albumin, no sugar, a well-marked reaction for acetone and for diacetic acid was obtained, there was a fair amount of indican, a slight pathological excess of urobilin was present, but no bile-pigment was detected, the "pancreatic" reaction (by the improved or C -method) showed many fine crys- tals, soluble in 33 per cent, sulphuric acid in ten to fifteen seconds. The history of this case, as supplied to us by Dr. Edgecombe, is as follows: April 2 2d a moderate amount of parotid swelhng, with no constitutional dis- turbance ; April 2 5th the patient was delirious ; April 2 7th there was severe epigastric pain with vomiting, a tem- perature of 100° F., and a pulse of 96; April 28th the vomiting continued, tenderness and swelling in the epigas- trium were detected; April 29th vomiting less, but still pain and swelling in the epigastrium, temperature 102° F., pulse 105; April 30th pain in the epigastrium disap- peared, but still slight swelling and tenderness, tempera- ture and pulse normal. It will be noticed that in these two cases the results of the "pancreatic" reaction coin- cided with the clinical symptoms and course of the disease. Among the blood infections have to be mentioned "pyemia," which presents no special peculiarity in the pancreas, and "syphilis," which may affect the pancreas 374 The Pancreas: Its Surgery and Pathology either avS a tertiary affection in the shape of gumma or as a congenital affection, as first described by Birch- Hirschfeld. It produces an interstitial pancreatitis of the interlobular type and the islands of Langerhans are unaffected. That the spread of ulceration inwards from the stomach may produce an indurative pancreatitis, or even suppu- ration in the pancreas, can be readily understood, for the ulcer must be constantly bathed with septic matter and the eroding action, when once it has passed through the stomach wall, may assume great activity. The effect of the spread of ulceration is also well exemplified by the case (described later) of pancreatic abscess apparently due to gastric ulcer bursting into the stomach and setting up acute gastritis, for which gastro-enterostomy was per- formed with a good result. Whether alcohol can act directly in producing cirrhosis is a matter of great doubt, the probability being that it sets up a gastro-intestinal catarrh which by extension gives rise to the chronic infective process, or another explanation may be in the fact that alcohol causes vascu- lar degeneration, a well-recognised cause of chronic inter- stitial pancreatitis. With regard to cirrhosis, the most chronic form of inflammation of the pancreas, which is, though slow in its progress, almost necessarily fatal from diabetes, the cause is probably a long-continued catarrh setting up interlobu- lar and interacinar pancreatitis, which is originally due to infection. Vascular degeneration is ascribed as a cause of chronic pancreatitis in old or in prematurely aged persons. CATARRH OF THE PANCREAS From the foregoing remarks on the etiology of inflam- mation of the pancreas it will be seen that catarrh of the pancreas is a disease as well worthy of recognition as is Inflammatory Affections of the Pancreas 375 catarrhal jaundice, which in the same way has until recently been thought to be always dependent on catarrh of the bile-ducts. It is held that biliary catarrh is known to exist, as it can be so readily recognised by enlarge- ment of the liver and jaundice, but that catarrhal pan- creatitis is beyond recognition. We hope to be able to prove that these views regarding diagnosis will need revis- ing, for catarrh of the pancreas can also be usually verified by digestive and metabolic signs, and by swelling of the gland, which can, in some cases, .be recognised by palpa- tion through the abdominal wall, but in others only by manipulation of the pancreas through the opened abdo- men. Just as catarrh of the bile-ducts may, and usually does, pass off if the cause be removed, so may pancreatic catarrh entirely clear up under appropriate treatment. Should the cause continue, the catarrh will become chronic and an interstitial pancreatitis ensue which may end in cirrhosis or atrophy of the gland— a condition which prob- ably always has a fatal termination from diabetes. It will be seen that chronic interstitial pancreatitis is in many cases simply a sequence of pancreatic catarrh, and as the latter is curable by appropriate treatment, and the former, when well advanced, is only capable of relief and probably not of complete cure, it is of the ut- most importance that we should recognise catarrh of the pancreatic ducts at an early stage, and if in a short time it fails to yield to medical treatment, that we should per- form an exploratory operation with a view to remove the cause, whether that be gall-stones or some other removable condition ; but if the cause be not discovered, or if when found it proves to be incapable of removal, then drainage of the bile-ducts, either by cholecystenterostomy or cholecystotomy, will nearly always aflord relief: (i) by removing the infected bile and thus ridding the system of poison which tends to deteriorate the blood; and (2) by removing the pressure of pent-up bile from the pan- 376 The Pancreas: Its Surgery and Pathology creas, thus rcHeving tension. Still another beneficial effect will result in some cases where the obstruction is at the papilla, for the pancreatic ducts will then also be drained indirectly through the bile-ducts. In certain cases a mere manipulation of the gland without drainage has been followed by recovery and apparently by cure. An explanation of this result may possibly be that an obstruction in the shape of concretions or adhesions may have been inadvertently removed during the manipula- tion; but in one case thus treated without drainage glycosuria has subsequently developed, which possibly might have been prevented by draining the ducts in the first instance. An example of the beneficial effects pro- duced by draining the ducts in catarrhal pancreatitis is afforded by the following case : A patient, aged thirty-eight, after being subject to indigestion for years had biliary colic in July, 1899, and passed gall-stones, which were found in the motions. Subsequently the attacks of pain were frequent and se- vere, necessitating the use of morphia. They were usually accompanied by icterus, which, though slight, probably never quite disappeared. When seen on November, 1903, he had lost flesh and was prevented from carrying on his professional duties. The metabolic and digestive signs of pancreatic catarrh were well marked. At the opera- tion on November 23, 1903, no gall-stones were found, though the gall-bladder was thickened and adherent to contiguous organs. The pancreas was firmer than usual, though not very much swollen. Cholecystotomy led to recovery, though the drainage of the bile-ducts had to be continued for three months. The patient is now well. In this case the pancreatic catarrh had evidently been set up by the passage of gall-stones through the common duct. The pancreatitis had, however, persisted, and was not only keeping up painful symptoms, but leading to obstruction of the bile-ducts and to interference with nutrition. This case would formerly have been called Inflammatory Affections of the Pancreas 377 catarrhal jaundice, whereas it was really due to catarrhal pancreatitis, as proved by the digestive and metabolic signs and later by operation. Just as post-mortem evidence is not easy to obtain in simple catarrh of the liver, so it is reasonable to antici- pate that pathologists will rarely find gross lesions of the pancreas, even if opportunity for a post-mortem examination occurs, in cases of catarrhal pancreatitis. The micro-photographs in Fig. 121 are from a case of Fig. 121. — a, Catarrh of the pancreas during incipient stage of inter- stitial pancreatitis (X 30) ; h, a. portion of the same more highly mag- nified, showing the round-celled infiltration (X 190)- early interstitial pancreatitis where death occurred in an aged patient twelve hours after operation, appar- ently from a cerebral attack that came on during anaesthe- sia. A gall-stone was impacted in the common duct and pressed on the pancreatic duct, the pancreas being found swollen at the time of operation. There was a well-marked pancreatic urinary reaction before operation. At the necropsy no gross lesion of the pancreas could be seen, but microscopically there were small-celled infiltration, 378 The Pancreas: Its Surgery and Pathology congested vessels, in fact, an incipient interstitial pancrea- titis, and it is probable that careful investigation would demonstrate similar microscopic lesions in cases where no gross changes are manifest. It is not necessary to consider separately the sympto- matology of acute and chronic catarrh of the pancreas, as the symptoms and signs, though less in degree, are practically the same as those of chronic pancreatitis, under which subject they will be' fully dealt with. SUPPURATIVE CATARRH This disease bears the same relation to simple pan- creatic catarrh that simple catarrhal jaundice does to suppurative chol- angitis and, like the latter, it is an extremely serious and frequently a fatal disease. In all the cases we have seen gall- stones have been the cause, but why some patients should have sim- ple catarrh ending in chronic inter- stitial pancreatitis and others should at once develop an acute suppuration of the pancreatic duct it is difficult to say, unless one may surmise that in the latter class the infection may be of a more virulent character and the patient's tissues less able to withstand the attack. The disease tends towards death from septi- Fig. 122. — Camera lucida drawing of suppurative catarrh of the pancreatic ducts : I, Exudation and cells, 2, duct; 3, acini; 4, pus; 5, detached epithelium (A. J. Chal- mers, Journ. of Ceylon Branch of Brit. Med. Assoc, vi, part 2, 1904). Inflammatory Affections of the Pancreas 379 csemia, or if the process be less acute, or the vital powers more resistant, it may possibly end in a localised abscess. Suppurative catarrh of the pancreatic ducts is generally, if not always, combined with suppurative cholangitis. The following four cases illustrate the serious nature of the disease, but show that it is not necessarily hopeless if treated early. If the suppurative catarrh be diffuse and involve the ducts throughout the liver and pancreas, the associated septicaemia is very serious, as the following case, seen with Dr. Hector Mackenzie, proves: Mr. W , aged sixty-five years, seen on January 4, 1904. He had had attacks of gall-stones seven years before and two seizures during the last two years, both of which were followed by jaundice. His present illness started on November 23d with severe pain, followed by jaundice. On December 20th a very severe attack of colic was followed by more intense jaundice and enlarge- ment of the liver with irregular temperature and ague-like attacks. The patient had had albimiinuria for seven or eight years. On examination there was tenderness above and to the right of the umbilicus and he had severe pain. A specimen of the urine was examined, and found to give a marked pancreatic reaction, and to contain calcium oxalate crystals. On opening the abdomen on January 7th, firm adhesions were encountered, and on detaching the omentum, phlegmonous cholecystitis was discovered, with gangrene of the fundus of the gall-bladder; pus escaped freely, but the peritoneal cavity w^as saved from being soiled by means of sponge packing. The common duct was enormously dilated and embraced by the swollen pancreas, but no gall-stones could be felt. On opening the common duct a large quantity of pus and bile escaped. By means of the scoop passed into the common duct and the fingers passed behind the pancreas a number of ' gall-stones were extracted, but a hardness could be felt at the papilla which could not be removed. On laying this open after incising the duodenum, a gall-stone was removed from the ampulla of Vater, and pus was imme- diately seen to flow from the duct of Wirsung. The duo- 380 The Pancreas: Its Surgery and Pathology denum was then closed, the gangrenous upper part of the gall-bladder was removed, and the common duct and gall-bladder were drained. The patient bore the operation well, and from that time onwards had no more fever, but for the fortnight during which he lived his temperature was persistently subnormal. He had no peritoneal symp- toms and the bowels were moved freely from the second day onwards. Calcium chloride had been given before operation, and at the operation he lost no blood. None was given subsequently to operation, as the rectum was intolerant of injections, and on the eighth day there was rather free oozing of blood from the drainage track, which had to be treated by gauze packing, after which the cal- cium chloride was resumed, and no more bleeding occurred. On the eleventh day the patient became somnolent and declined to take food. From this time he got gradually weaker, and died comatose on the fourteenth day in a condition almost resembling that associated with acute atrophy of the liver. If the suppurative catarrh takes on a very acute form, the development of abscess in the liver and pancreas may occur, and the condition become one of pyaemia, when the chance of recovery will be very remote, as in the following case: The patient, a woman, aged sixty-five years, seen with Sir William Broadbent and Dr. Bousfield, was suffering from deep jaundice, suppurative cholangitis, pancreatitis, and parotitis of pysemic origin ; rigors with a temperature of 105° occurring daily, or even twice a day, the acute symptoms having come on within a fortnight, though there had been a history of gall-stones for years. The common and hepatic ducts were filled with gall-stones, which were removed through an incision in the common duct, and a large quantity of extremely offensive pus and bile was evacuated. At the same time the right parotid gland (the seat of inflammation) was incised. The bile was examined bacteriologically, and found to contain the bacillus coli in large numbers ; next in numbers were strep- tococci and another rather fine bacillus, which appeared to grow anaerobically only, and there was a fine spore- Inflammatory Affections of the Pancreas 381 bearing organism, probably the bacillus putrificus coli. The urine gave a well-marked pancreatic reaction. The patient, who had also heart disease and albuminuria, appeared to be doing well for twenty-four hours, when she died suddenly, apparently from cardiac thrombosis. If the suppurative catarrh assumes a subacute form, it may end in a simple pancreatic abscess, which can be successfully evacuated, as in the following cases: Mrs. P., aged sixty-one, gave the history of having been subject to biUary colic for three or four years, though there had been no jaundice till two and a half years ago, since which time the attacks of pain had always been accompanied by rigors and by deepening of the jaundice. Within a short time of being first seen, the symptoms had become aggravated, and the loss of flesh had become extreme. The patient was so ill that the question of can- cer of the pancreas was raised, but the pancreatic reaction in the urine and a chemical examination of the faeces defi- nitely pointed to inflammation and not to growth. At the operation the pancreatic portion of the common duct was packed with large gall-stones and the head of the pancreas was markedly swollen. On passing the scoop through the opening in the common duct, a stone the size of a cherry was extracted from the pancreatic portion of the duct, it being covered with offensive pus. This had apparently lodged in a cavity in the head of the pan- creas. A profuse discharge of bile and offensive pancrea- tic fluid with pus continued to pass for a week, after which . the discharge became gradually less. She made a good recovery, and was well two years later. In general subacute pancreatitis, starting as suppura- tive catarrh with the formation of a locaHzed abscess, the pancreas may be so damaged that after the abscess has been cured by drainage, the extensive interstitial pan- creatitis may ultimately lead to the death of the patient at a longer or shorter interval, as in the following cases : Mr. H., aged forty, had suffered from continuous fever, with exacerbations associated with rigors, that recurred 382 The Pancreas: Its Surgery and Pathology almost daily. He gave the history of failing health for nine months, and of having had gall-stone attacks much longer, but the acute symptoms associated with jaundice had only been present for a fortnight. The pancreatic reaction was found in the urine. At the operation, on October 11, 1900, he was far too ill to bear a prolonged search, and as the adhesions were very firm, it was felt desirable only to drain the bile-ducts through the gall- bladder, though a marked swelling of the pancreas made it appear probable that an abscess might be present, A large quantity -of muco-pus drained from the gall-blad- der and a number of gall-stones were re- moved. The abscess of the pancreas dis- charged subsequently through the drainage- tube, after which the pancreatic swelling subsided. The patient made a slow though steady recovery, and returned home early in December. Though he was able to get out and to take food, he never fully regained his strength, and died in February of the fol- lowing year. At the necropsy the pancreas was found to be much enlarged and to be the seat of interstitial pancreatitis. The cavity where the abscess had been was occupied by a little pulpy material, but no further collection of pus was formed, nor were any gall-stones discovered in the bile-ducts. A microscopic examination of the pancreas showed advanced interstitial pancreatitis. The preceding cases are most instructive in that they illustrate one of the dangers of cholelithiasis, which might he avoided by appropriate treatment at an earlier stage, for the removal of gall-stones before the onset of deep '^' Fig. 123. — Microphotograph of the pancreas from a case of chronic sup- purative pancreatitis (X 40). Inflammatory Affections of the Pancreas 383 jaundice and infection of the bile and pancreatic ducts is with due care and in skilful hands almost devoid of danger. So far as we know, these conditions were first described in the Hunterian Lectures for 1904 as separate a.nd distinct diseases. The reasons given and the cases related show a justification for separating catarrhal inflam- mation about to be described. Simple catarrh of the pancreas can be treated most successfully if taken in time, but, as will have been noticed, suppurative catarrhal pan- creatitis is quite as serious as acute phlegmonous pancrea- titis, and unless treated surgically must be almost neces- sarily fatal. Literature Auche: Journ. de Med. de Bordeaux, Oct. 29, 1905. Baillie: "Morbid Anatomy," 1799. Birch-Hirschfeld: Arch. d. Heilkunde, 1875, xvi, 174. Edgecombe: Brit. Med. Journ., Feb. 16, 1907. Fitz: Med. Record, 1889, xxxv, 197. Guinard: Brit. Med. Journ., March 16, 1907, p. 656. Halsted: Quoted by Opie, "Diseases of the Pancreas," 1903. Helly: Arch. f. mik. Anat., Bd. Hi, p. 773. Jacob: Brit. Med. Journ., June 23, 1900, p. 1532. Lemoine and Lapasset: Bull, de la Soc. Med. des Hopit., July 7, 1905. Mayo: Address to the Amer. Surg. Associat. Opie: "Diseases of the Pancreas," 1903. Percivai: Tr. Assoc. King's Coll. Ireland, 18 18 p. 128. Portal: "Anat. Med.," 1804, p. 353. Robson, Mayo: Hunterian Lectures, Lancet, March 19, 26, April 2, 1904. Simoni: Bull, de la Soc. Med. des Hopit., July 30, 1903. Smith: Lancet, July 28, 1901. Testut: "Traite de Anat. Humaine," viii, 1894. Tulpius: "Observat. Med.," 1672, p. 328. CHAPTER XV ACUTE PANCREATITIS AND SUBACUTE PANCREATITIS ACUTE PANCREATITIS Symptoms. — Acute pancreatitis is usually ushered in- by a sudden pain in the superior abdominal region, accom- panied by faintness or collapse, and followed, sooner or later, by vomiting. There is usually some epigastric swelling with tenderness from the first, and if the warm flat hand be placed over the epigastrium, and retained there without movement for a time, it will be found that the swelling is diffuse and not simply dependent on a dis- tended stomach or colon, though later, when peritonitis is established, the hollow viscera become inflated. It is almost constantly accompanied by constipation, so that it is quite usual for these cases to be mistaken at first for intestinal obstruction. The obstruction, however, is not absolute, flatus passes, and a large enema may secure an evacuation ; if the patient survives for several days diar- rhoea may supervene. The pain may be so severe as to produce syncope or collapse, and though the pain does not quite pass away, it has a tendency to be paroxysmal and to be increased by movement; it is associated with well-marked tenderness just above the umbilicus and between it and the ensiform cartilage. The pain is soon folowed by distension in the superior abdominal region, which may become general, and usually does so in the later stages, and by vomiting, first of food, then of bile, and soon of black, altered blood. The vomiting may be severe and each attack of sickness aggravates the pain. Rarely vomiting may not be a prominent symptom. 384 Acute Pancreatitis and Subacute Pancreatitis 385 Slight jaundice from associated catarrh of the bile-ducts and pressure is usually present and deepens the longer the patient survives. As the impaction of a gall-stone in the ampulla of Vater is probably one of the most frequent causes, the jaundice may become intense, from a complete stoppage of the passage of bile into the duodenum. The aspect is anxious and the face is pinched, resembling the facies of peritonitis, which, in fact, is usually present. The pulse, which is rapid and small, is a better guide than the temperature, which may be normal, subnormal, irreg- ular, or high. In the ultra-acute cases the temperature is usually subnormal, but in the cases that survive for several days the temperature becomes irregular and may be excessive. Delirium comes on in the later stages. The distension and tenderness may prevent an exact examination of the pancreas, which would otherwise be found enlarged. Death usually supervenes from the second to the fifth day from collapse, probably due to absorption of virulent matter, though in the less acute cases life may be more prolonged and recovery may possi- bly occur, as in cases related below, proved by laparotomy and the discovery of extensive fat necrosis, and by others reported elsewhere. Acute pancreatitis thus takes on the form of acute peritonitis starting in the superior abdominal region. If life be prolonged, the condition may become one of subacute pancreatitis, the onset in such a case being usually less grave, though often equally sudden. It is even possible for the trouble to resolve, apparently completely, and then for a relapse to occur, this sequence being repeated on several occasions. The preceding description refers to acute pancreatitis generally and applies to the hsemorrhagic, gangrenous, or suppurative varieties, which are phases of the same infec- tive conditions, though the morbid appearances differ so much. In gangrenous pancreatitis the organ is dry and dark or even black, and there can be little doubt, as 25 386 The Pancreas: Its Surgery and Pathology Opie has remarked, that this condition represents a late stage of the hsemorrhagic form. What has been pre- viously said concerning pancreatic haemorrhage will show that neither clinicians nor pathologists are agreed on this subject, some believing that inflammation precedes the hsemorrhage, — among these being Fitz, who designates the disease " hsemorrhagic pancreatitis," — others holding that the haemorrhage precedes inflammation, which is, in fact, caused by bacterial infection of the haemorrhagic effusion. It is probable that both views may be correct in different cases, for although a primary pancreatitis may be accompanied by haemorrhage, yet this origin is not the only one, and there are many cases in which haemorrhage precedes and, in fact, is the cause of, inflam- mation ; first, owing to the great tendency of the gland to disruption because of its soft structure when haemorrhage does occur; secondly, owing to the setting free of the pancreatic secretion which decomposes and digests the damaged tissues; thirdly, owing to the communication of the gland with the intestine, rendering the access of putrefactive organisms likely; and, fourthly, owing to the great tendency of the damaged gland and the effusion to become decomposed as soon as organisms gain access. From its proximity to the peritoneum, acute peritonitis rapidly follows acute pancreatitis. These two varieties of haemorrhagic pancreatitis may at times be clinically differentiated, the ultra-acute, with a violent and sudden onset, accompanied by collapse and ending fatally with extreme rapidity, being for the most part the ones where the hemorrhage precedes the inflammation, and the somewhat less though still acute cases, where the onset is more gradual, where the symptoms are not ushered in by collapse, and where resolution and relapse are liable to occur, being the ones where the inflammation precedes the haemorrhage. The varieties in which the inflamma- tion precedes the haemorrhage may in the severer forms Fig. 124. — Pancreas and adjacent tissues from a case of acute hemorrhagic pancreatitis with fat necrosis (St. Bartholomew's Hosp. Museum). Acute Pancreatitis and Subacute Pancreatitis 387 approach the subacute varieties of pancreatitis. These views simplify the subject and place the disease of haemor- rhagic pancreatitis in a line with other well-known inflam- mations. The coloured plate (Fig. 124) shows a striking specimen of acute haemorrhagic pancreatitis with fat necrosis, preserved by the Keiserling process, in St. Bar- tholomew's Hospital Museum. The specimen represented in Fig. 125, copied from Nothnagel, is in the Warren Anatomical Museum of Har- vard Medical School. It exemplifies a case of true 0» Fig. 125. — Acute hsemorrhagic pancreatitis (Oser). hsemorrhagic pancreatitis, of four days' duration, in which the inflammation was the cause of the hcemorrhage. A well-marked example of haemorrhagic pancreatitis from the museum of the Leeds Medical School, which was under the care of Mr. B. G. A. Moynihan and was depen- dent on gall-stones, is seen in Fig. 126. A specimen in St, George's Hospital Museum (204 A) is a good example of haemorrhagic or necrotising pancrea- titis (Fig. 127). The case was reported in the "Lancet" of October 19, 1901, p. 1041, Etiology. — In many cases of acute pancreatitis the etiology is obscure, for although the disease is capable of 388 The Pancreas: Its Surgery and Pathology being produced artificially by injection of bile, bile salts, and other substances into the main pancreatic duct, yet ( Fig. 126. — Specimens from a case of acute pancreatitis (Leeds Museum, E E 200): o, Gall-bladder containing stone; b, slough of the pancreas; c, piece of omental fat, showing fat necrosis. in only a few cases has it been produced in a similar way by diseases in the human subject. In some cases septic influences have been causative and Acute Pancreatitis and Subacute Pancreatitis 389 in others blood disorders, but in a considerable number of reported cases gall-stones either directly or indirectly have been instrumental in setting up the disease. If a small gall-stone happens to descend into an unusu- ally large diverticulum of Vater and to lodge there, it will make a through channel from the common bile-duct into the pancreatic duct, and so set up acute pancreatitis, the infected bile being forced direct into the pancreatic duct, as in Dr. Halsted's case, reported in O pie's work on the pancreas. "L. F., male, aged forty-seven years, was admitted to the Johns Hopkins Hospital complaining of abdominal pain and fever. He had suifered with somewhat fre- quent attacks of indigestion, char- acterised by pain after eating, dis- tension, and rarely nausea and vomiting, but otherwise had en- joyed good health. Six months before his present illness he had had an attack of jaundice, lasting about three weeks, and accom- panied by abdominal pain. The present illness began eighteen days before admission, when he was suddenly seized with violent nausea and vomiting, accompan- ied by intense cramp-like pain in the abdomen. The vomiting continued during the first night, and had since only occasionally recurred. The ab- dominal pain, which was not localised, remained severe during four or five days, and at times there were symptoms of collapse. The abdomen was distended and the bowels were constipated until the fifth day, when, with the aid of a purgative, movement occurred. The stool was normal in colour. On the third day elevation of temperature to 101.5° F. was noted. About the seventh day tenderness and slight swelling were noticed in the right hypochon- Fig. 127 . — Acute haem- orrhagic pancreatitis and necrosis of the pancreas (St. George's Hosp. Mu- seum, 204 A). 390 The Pancreas: Its Surgery and Pathology drium. Since this time the patient had an irregular tem- perature (ioo° to 103° F.), with several chills. After the first few days the abdominal pain and tenderness were riot severe, but the distension of the abdomen gradually increased. Jaundice was not noticed. " Physical examination: The conjunctivae have a slight yellow cast. On inspection of the abdomen a distinct prominence is found to occupy the right hypochondriac and right half of the epigastric region, extending into the upper half of the umbilical region. Its lower margin, which descends on inspiration, is felt in the middle line at the level of the umbilicus. Its right border cannot be sharply defined, but in the median line the fingers can be pressed in above it. Over the resistant mass there is dull tympany. The leucocytes number 1800. The urine is clear, its reaction is acid, and specific gravity 1.017. There is no reduction of Fehling's solution. A trace of albumin is present. On the second day after admission a stool passed was of a golden yellow colour. On the third day the leucocytes numbered 19,500, and the tem- perature varied from 99.2° to 101.8° F. During the night the patient was irrational at times. The tempera- ture rose gradually, reaching a maximum of 104° F. A liquid stool of ochre-yellow colour was passed. The urine had a specific gravity of 1.020 and no reaction for sugar was obtained. A diagnosis of suppurative pancrea- titis was made and an operation for its relief was performed under cocaine anaesthesia. A linear longitudinal incision was made below the costal margin within the right mam- millary line. After incising the great omentum between the stomach and transverse colon, an abscess cavity was entered. Grumous, purulent fluid, containing necrotic particles, was evacuated. A rubber drainage-tube, packed about with gauze, was inserted into the wound. After operation the pulse remained weak, and death followed at the end of about four hours. The duration of the fatal illness was twenty-one days. ''Autopsy: Performed three hours after death. The body is that of a large-framed, muscular man, with abun- dant subcutaneous fat. The omentum, which contains a large quantity of fat, is thickly studded with conspicu- ous opaque white areas, usually round, and about 3 mm. Acute Pancreatitis and Subacute Pancreatitis 391 in diameter. Similar opaque white areas are present in the fat of the mesentery, in the subperitoneal fat of the abdominal wall, over the bladder, over the kidneys, and about the colon. The drainage-tube inserted into the abdominal wound passes through a small incision in the great omentum and enters an immense abscess cavity; the foramen of Winslow is closed. The walls of the cavity are very irregular and ragged, and have a necrotic appear- ance, in general opaque and grey, occasionally black. The blackish-grey appearance extends only a short dis- tance below the surface, and where the wall is formed of fat gives place to numerous foci of opaque white colour. The retroperitoneal fat in front of the left kidney and psoas muscle has been eroded, and an extension of the cavity passes behind the jejunum near its junction with the duodenum. To the left of the descending part of the duodenum, occupying the position of the pancreas, and projecting forward into the abscess cavity, is a great mass of black material, necrotic in appearance, extending to the left as far as the spleen. The material is reddish- brown on section, somewhat spongy in texture, soft, dry, and friable. The cavity contains at least 500 c.c. of fluid, reddish-grey material, in which are fat droplets and black necrotic particles. The liver is flaccid in con- sistence. The bile-ducts are slightly dilated, and con- tain thin, yellow bile. The gall-bladder is bound by numerous adhesions to the duodenum and stomach. Its walls are thickened and it is much distended, con- taining viscid yellow bile and more than one hundred brown, faceted calculi, varying in diameter from 0.5 to I cm. The hepatic, cystic, and common ducts are much dilated. On opening the duodenum a stone is felt below the mucous membrane, situated in the common bile-duct near its termination. It is 7 mm. in diameter and resem- bles those present in the gall-bladder. The pancreatic duct unites with the common bile-duct at a point 7 mm. from the duodenal orifice. The pancreatic duct is not distended. The pancreas occupies the posterior wall of the abscess cavity of the lesser peritoneum, and is covered by the mass of reddish-brown, friable material, changed coagulated blood-clot, above described. The organ is of large size, and the glandular tissue is in great part firm. 392 The Pancreas: Its Surgery and Pathology yellowish- white, and well-preserved. The interstitial tis- sue has a dull reddish, in places hsemorrhagic appear- ance, and contains conspicuous opaque yellow areas of irregular shape. Where the anterior surface of the head and body is in contact with the overlying material there is a superficial zone of soft, greyish, necrotic appearance. The other organs present no noteworthy alteration. ''Histological examination: The interstitial tissue of the pancreas is much increased and wide bands of fibrous tissue separate groups of lobules. Numerous irregularly shaped cells filled with brownish-yellow pigment granules, which give the prussian-blue reaction for iron, afford evidence of former hemorrhage. In a few places well- preserved red corpuscles are scattered in the tissue. Foci of necrotic fat are present. Many acini are widely dilated ; their cells are flat and the lumen is much distended, con- taining products of secretion and occasionally one or. more poly nuclear leucocytes. In an area corresponding to the superficial zone of necrotic appearance before mentioned, nuclei no longer stain, and the architecture of the glandular tissue is only obscurely distinguished. A thick band of newly formed fibrous tissue containing an occasional acinus or duct separates the necrotic parenchyma from that which is still intact. The mass covering the pancreas is found to consist of altered blood ; upon and immediately below its surface are numerous polynuclear leucocytes. ''Bacteriological examination: Cultures from the blood contained in the heart, from the lungs, and from the liver, were found to contain the bacillus coli communis, A plate culture from the material covering the pancreas, and forming part of the abscess wall, contained the bacillus coli communis, the bacillus lactis aerogenes, and the bacillus proteus vulgaris. "Anatomical diagnosis: Cholelithiasis; calculus lodged in the common bile-duct near its orifice; slight jaundice. Old haemorrhage within and about the pancreas, with localised necrosis and chronic inflammation; necrosis of the fat of the pancreas, greater and lesser omentum, mesentery, and subperitoneal fat of the abdominal wall; peri-pancreatic abscess limited by the lesser peritoneal cavity. Laparotomy wound." Acute Pancreatitis and Subacute Pancreatitis 393 A somewhat similar case has more recently been re- ported by Bunting in the "Johns Hopkins Hospital Bulletin." "A well-nourished man, aged fifty-one, was subject to attacks of epigastric pain with some constipation, which were usually easily relieved. He was seized with intense epigastric pain, became collapsed, and showed considerable abdominal distension. Intestinal obstruction was diag- nosed and laparotomy was performed. Some peritoneal adhesions were freed and the wound was closed. There was no fat necrosis to attract attention to the pancreas. Death occurred on the following morning. At the autopsy both chronic interstitial and acute haemorrhagic pan- creatitis were found. The pancreas was large, swollen, and mottled with red areas of haemorrhage and opaque areas of fat necrosis. The gall-bladder was distended with bile and the bile-ducts were dilated and firm to the touch, showing that there was obstruction to the flow of bile. On gentle pressure bile did not escape from the papilla, but on increased pressure there was a sudden escape of bile, carrying before it a small yellowish-white mass, which was unfortunately lost. In the bile-ducts a stone about 2 mm. in diameter was foiuid in the apex of the ampulla of Vater, close to the orifice. In the gall- bladder and cystic duct were about 400 soft, light- coloured cholesterin stones, varying from 0.5 to 6 mm. in diameter. There was therefore little doubt that the escaped mass was a stone. The anatomical relations of the ampulla and ducts were such that the obstruction set in progress the mechanism described by Opie, and re- sulted in the injection of bile into the pancreatic duct. This duct joined the common bile-duct 1 1 mm. from the tip of the papilla and was dilated and bile-stained for 4 cm. from its orifice. The common duct was dilated and somewhat hypertrophied. This, in connection with the induration of the pancreas, seemed to indicate that the previous attacks of pain were due to the passage of gall- stones, some of which might have been large enough to block the pancreatic duct in transit and cause the chronic interstitial pancreatitis." 394 The Pancreas: Its Surgery and Pathology But the anatomical conditions just mentioned, though evidently potent, are certainly not necessary for the pro- duction of acute pancreatitis. Any gall-stone or stones impacted in the pancreatic portion of the duct, or even filling the ampulla of Vater, may produce acute pancrea- titis, as in a case under the care of Dr. Fison, of Salisbury. "A man, aged thirty-nine, had a sharp attack of diar- rhoea on March 27, 1904, having been previously consti- pated. The next day, about one and one-half hours after dinner, he was seized with severe epigastric pain followed by vomiting. At 5 p. m. he looked anxious and ill, and the abdomen was tense and tympanitic, but there was no jaundice. The vomiting persisted. There was tenderness over the gall-bladder and to a less degree over the sto- mach, but no enlargement of the liver or any indication of tumour. Temperature, 98°; pulse, no. " The next day the temperature was 97° and pulse 120 ; the vomiting continuing, morphia was given. On the 30th the temperature was 96.8°, the pulse 125, small, weak, and thready, respiration 36. The pain was easier. Urine scanty and dark. ''Operation on evening of the 30th, fifty-four hours after first attack of pain. Very extensive fat necrosis found in subcutaneous tissues and in omentum, mesentery, etc. Large quantity of brown inoffensive fluid in perito- neum. Incision made into tissues around pancreas through mesocolon. Gall-bladder drained through an- other incision, many gall-stones removed. Free drainage of abdomen. After recovery from ansesthetic the vomit- ing persisted, and the pulse remained absent from the wrist up to death, some hours later. "At post-mortem examination a pint of bloody fluid in peritoneal cavity. Base of mesocolon filled with friable offensive material, blackish-brown in colour, and here and there streaked with pus. Pancreas much swollen and weighed seventeen ounces. Haemorrhagic infiltra- tion in centre of body and another in tail; consistency very firm, with swelling of lobules. In the cystic duct were three gall-stones, in the common duct four, and in the hepatic duct four. One gall-stone f inch in length Acute Pancreatitis and Subacute Pancreatitis 395 completely filled the ampulla of Vater, into which the duct of Wirsung opened one-third of an inch from the papilla. The duct of Wirsung did not contain bile. " Urine sent for examination by Dr. Cammidge showed crystals soluble in one-half minute by the 'A' reaction, and a few crystals by the 'B' reaction soluble in the same time. "The following is Dr. R. Salisbury Trevor's report of examination of the pancreas : " The gland is enlarged in all its diameters, the margins being rounded off and producing as a consequence a sausage-shaped contour. In the head, the middle of the body, and the tail are chocolate-coloured areas, which are fairly sharply differentiated from the surrounding parenchyma in which the normal lobulation is visible. The duct of Wirsung is not bile-stained. The portion of common bile-duct attached to the head of the gland appears to be somewhat dilated. Around the gland, as well as in it, are numerous typical foci of fat nec- rosis. "Microscopical examination: Sections have been prepared from the head, body, and tail, in most instances to include the chocolate-coloured areas as well as ap- parently normal parenchyma. The dark coloured areas are due to necrosis of the parenchyma associated with haemorrhage, and in the sections from the head and tail are demarcated off from the neighbouring gland acini by well-marked zones of inflamniatory small-celled in- filtration. In the tail section inflammatory reaction is absent, the necrosed areas merging gradually with the unaffected parenchyma. In the necrosed areas the gland parenchyma is only barely recognisable by a faint alveolar structure, all gland elements having disappeared. The whole of these areas stain badly. In the necrotic por- tions the smaller blood-vessels are filled with more or less hyaline thrombi. Around the necrotic areas in the head and body is a deposit of old blood-pigment, and the ap- pearances rather suggest that here the lesions are of older date than those in the tail. Inflammation is most marked in sections of the head. The remaining gland parenchyma is badly preserved owing to auto-digestion, and the head appears to show a slight grade of chronic 396 The Pancreas: Its Surgery and Pathology interstitial pancreatitis of the interlobular type. Through- out the sections the islands of Langerhans are found with difficulty, and, from comparisons with other sections, their number in the tail sections at all events appears to be diminished. Two of the islands of Langerhans found in the tail sections are very large in size. The cells, however, are rather broken up, and into one of them haemorrhage has occurred. Minute changes are not rec- ognisable owing to bad preservation of the tissue. The epithelium of Wirsung's duct shows distinct signs of a catarrhal change. "Summary. — The condition is one of acute pancrea- titis, with haemorrhage and necrosis (the acute form of haemorrhagic pancreatitis in Mayo Robson's classifica- tion)." Owing to Dr. Fison's kindness we are able to show photographs of the extensive fat necrosis found in his case (see Figs. 87, 88). The following is a case of gangrenous pancreatitis due to gall-stones, which recovered after operation: Mr. S , aged fifty-eight, had for six years been subject to paroxysmal attacks of acute pain, starting in the right hypochondrium and radiating over the ab- domen and through to the right scapula, the attacks being accompanied by vomiting and more or less collapse. On several occasions he had passed small gall-stones. About ten weeks before being seen by one of us he was seized with an attack which did not, as usual, yield to morphia. The liver became enlarged and tender; there was a great amount of flatulence and acidity and a feeling of discomfort generally. After this seizure he had ague- like attacks and jaundice of varying intensity, and from that time a tumour steadily developed in the epigastric and right hypochondriac regions. He rapidly lost flesh and strength, and when he was taken into a surgical home for operation he was so feeble and emaciated that it was questionable whether he would be strong enough to bear it. Jaundice was well marked, and the tumour in the upper abdomen, which was tense, tender, and fluctuating, was still enlarging. He had had diarrhoea six times a Acute Pancreatitis and Subacute Pancreatitis 397 day for several days before admission, and the motions were bulky and pale and contained fat. The urinary pancreatic reaction was well marked. Just before oper- ation he vomited clear fluid, not containing bile. Operation was performed on April 5, 1902, when a pancreatic cyst was ex- posed between the stom- ach and colon, containing four pints of straw-col- oured fluid. Inside the cyst was found a mottled Fig. 128.— Slough of the pan- black slough with grey StS^Tovll Col^'StrgeL" patches, 2^ to 3 mches Museum, 2834 B). long by ij inches broad, and ^ inch thick, evidently pancreas. (See Figs. 128 and 129.) The gall-bladder and ducts contained thirty stones, two the size of walnuts. One of these was found at the junction of the cystic and common duct, and pressing on the latter. The cysts -,^ . of the pancreas and .St "^v ^f.^ • • the gall-bladder were .-**:.', ^- '•'• ' dramed by separate 4,1^4 ■ ' f . • tubes, with the stom- ' .- . J , •■"... ach and the first part * ''' ,' - . ■' '• of the duodenum be- 5"**" '.. -',',, ^ _,:, tween them. On be- ** ^ %r ;,''■- ., ing put back to bed : ' . ' t'.\ : .^ !'■" the patient was quiet, ^ .' "^"f^ .^ - but vomited frequent- ^-^^ ' , •/* ly. He made a steady ', . • recovery without any '^X*- ^ . '' ' untoward symptoms, ^ ;. . . and left for home on May 2, 1903. On Fig. 129.— Microphotograph of ne- Marrh i inn a the r.a erased pancreas shown in Fig. 128 Marcn 3, 1904, tne pa (X 40). tient was the picture of health, and had gained i^ stones in weight. He states that the gall- bladder opening had closed in six weeks and the pancre- atic fistula in nine weeks. 398 The Pancreas: Its Surgery and Pathology Although gall-stones would appear to be the most fre- quent cause of acute pancreatitis, yet the following cases operated on by one of us show that other conditions may cause it. In the case of a young woman suffering from acute suppurative pancreatitis the viscera were found hope- lessly matted together and there was extensive fat necro- sis all over the abdomen. There was no definite history of gall-stones, nor could any be found at the time of the operation. A subphrenic abscess containing masses of necrosed fat and dark pus was evacuated and drained. This only gave temporary relief to the patient, who suc- cumbed on the third day after operation, apparently from septic absorption. In another case, a young man, aged twenty-eight, slipped and fell forward against a board projecting from the end of a table at which he was working. The blow was comparatively slight and the man did not fall to the ground. Acute pancreatitis followed on what was prob- ably a mere bruising of the pancreas, followed by slight bleeding into the gland, but the effusion becoming in- fected, acute hemorrhagic pancreatitis supervened. An exploration for the cause of the peritonitis resulted in the discovery of a large collection of highly blood-stained fluid in the lesser peritoneal sac, some of which had burst through a small laceration in the omentum into the greater sac of the peritoneum. There was general peritonitis present at the time of operation, and though drainage was freely adopted both from the front and through the loin, the patient did not survive many hours. In the one case extension of infection from the duode- num was probably the cause ; in the other, traumatism ; but metastasis from mumps, blood conditions, and in- fection from various diseases, such as typhoid fever, pyaemia, etc., may be the cause, as in cases reported later. Diagnosis. — The diagnosis of acute pancreatitis is at first difficult, as the symptoms are only characteristic of peritonitis starting in the upper part of the abdomen. Acute Pancreatitis and Subacute Pancreatitis 399 Fitz's rule is worth bearing in mind: "Acute pancreatitis is to be suspected when a previously healthy person, or sufferer from occasional attacks of indigestion, is suddenly seized with violent pain in the epigastrium, followed by vomiting and collapse, and, in the course of twenty-four hours, by a circumscribed epigastric swelling, tympanitic or resistant, with slight rise of temperature." In case of laparotomy the presence of extensive fat necrosis is almost pathognomonic. At first the dift'erential diagnosis must be made from intestinal obstruction, perforating duodenal or gastric ulcer, ruptured gall-bladder or bile-ducts, phlegmonous cholecystitis, and gangrenous appendicitis. In consider- ing the difficulty of diagnosing between acute pancrea- titis and intestinal obstruction it has to be borne in mind that the two may coexist, as the swollen pancreas may embrace and strangle the duodenum, or a collec- tion of inflammatory material may seriously compress it. The swelling will, however, be usually less general in pancreatitis than in obstruction, and, even if the bowels will not move, flatus can generally be passed. In case of doubt, exploration may reveal fat necrosis. In perfora- tion of a duodenal or gastric ulcer there will generally have been premonitory symptoms pointing to the disease before the perforation actually occurs, and almost immed- iately an absence of liver dulness may be found. In acute ptomaine poisoning the history, the more general character of the pain, and the presence of diarrhoea will usually help the diagnosis. In phlegmonous cholecystitis the symptoms are usually preceded by a swelling and well-marked tenderness be- neath the right costal margin, at first distinctly localised and only later extending to the epigastrium and umbilical region, where the tenderness is generally found in acute pancreatitis from the beginning of the illness ; moreover, the history of gall-stones will usually be elicited. In 400 The Pancreas: Its Surgery and Pathology appendicitis the tenderness below, and to the right of, the umbilicus and the swelling in that region usually remove the difficulty created by the pain in both appendi- citis and pancreatitis, being frequently felt at first just above the umbilicus. In acute pancreatitis the excru- ciating pain, at first epigastric but later general, the ex- tremely rapid loss of weight, and the irregular tenderness opposite to, and above, the umbilicus are usually charac- teristic. Halsted lays stress on two symptoms— the excessive pain and cyanosis of the face and of the abdom- inal wall. The former symptom is universal, but the latter has not usually been present in the cases we have seen. The urinary test for the ' 'pancreatic crystals" should not be neglected, as a positive reaction has been obtained in all the cases of acute pancreatitis in which we have had the opportunity of employing it. Glycosuria is usually absent, but in two out of forty-one cases of hsem- orrhagic and in three out of forty cases of gangrenous pancreatitis Korte found it present. Treatment. — -The pain at the onset is so acute as to necessitate the administration of morphine, and the collapse will probably demand stimulants, which, on ac- count of the associated vomiting, may have to be given by enema. In the early stages the symptoms may be so indefinite that the indications for surgical treatment are often not clear enough to demand immediate opera- tion, but as soon as acute pancreatitis is suspected, as it may be by the combination of symptoms together with the urinary test, the surgeon must not wait until the collapse has passed off, as that may be dependent on septic absorption which can only be relieved by operation. The simulation of intestinal obstruction will probably lead to efforts to secure an evacuation of the bowels and relief to the distension. Just as in perforative or gan- grenous appendicitis an early evacuation of the septic matter is necessary to recovery, so, in this equally lethal Acute Pancreatitis and Subacute Pancreatitis 401 affection, an early exploration from the front through the middle line above the umbilicus or from behind through the left costo-vertebral angle is indicated, in order, if possible, to relieve tension, to evacuate septic material, to secure free drainage, and to arrest the haem- orrhage which leads to disintegration and necrosis of the pancreas. The after-treatment will be chiefly directed to combating shock and keeping up the strength until the materies morhi, both local and general, can be thrown off. Even if no pus be found no harm should accrue from such an exploration, which can be made in a few minutes through an incision in the middle line above the umbilicus. After establishing the diagnosis by the dis- covery of a swelling in the region of the pancreas with effusion of blood and associated with fat necrosis, a posterior incision in the left costo-vertebral angle will sometimes enable the diseased organ to be very freely drained for the evacuation of pus and gangrenous material without risk to the general peritoneal cavit}^ and with little danger of retained septic matter, as the drainage will be a dependent one. If, however, the inflammatory collection of the tensely distended and inflamed gland be incised from the front, as is advisable in certain cases, gauze packing and gauze drainage may usually be relied on to prevent general infection of the peritoneum. If there are signs of an obstructed common bile-duct, the gall-bladder should be drained, and if gall-stones be dis- covered they should be removed, if this can be done without seriously adding to the length of the operation or imperilling life by adding to the shock ; otherwise they may be left and removed on a subsequent occasion if free drainage of the bile-passages can be secured. We have had six cases of acute pancreatitis under our care and have operated on four, of which two recovered. Of the two cases where operation was not consented to, and where medical treatment alone was carried out, death 26 402 The Pancreas: Its Surgery and Pathology occurred in the first case on the third day and in the second case after a week's illness, attended in both with great pain and incessant vomiting. In a case of gangrenous pancreatitis in a man, aged fifty-eight, a collection of fiuid was opened through the great omentum, above the hepatic flexure of the colon, and a slough of the pancreas was extracted, after which free drainage was established. At the same time the gall- bladder and all the stones within reach were removed, but the common duct was not opened, as the patient was too ill to bear a prolongation of the operation. Fortu- nately, several small calculi worked back through the tube in the gall-bladder and recovery was not delayed and was ultimately complete. The pancreatic reaction was well marked in this case. In another case of a young married woman suffering from acute suppurative pancreatitis the viscera were found hopelessly matted together. There was extensive fat necrosis all over the abdomen. A subphrenic abscess containing masses of necrosed fat and dark pus was evacu- ated. The patient was only temporarily relieved and succumbed on the third day. In this case it would prob- ably have been better to have drained through the costo- spinal angle on the left side as well as from the front, but the patient was so ill that it did not appear to be ad- visable to do more lest death should occur on the table. In a case of traumatic haemorrhagic pancreatitis in a man, aged twenty-eight, drainage through the loin as well as in front was adopted, but did not save life, as at the time of operation peritonitis was already advanced, and involved both the greater and lesser peritoneal sacs. In another case, of a middle-aged medical man, who was seen with Dr. H. P. Hawkins, the diffuse fat necrosis and adhesions of the viscera and omentum into a dense mass presented a formidable obstacle to complete ex- ploration, but as no evidence of any collection of fiuid either in the pancreas or in the lesser peritoneal sac could be obtained, and as no gall-stones could be felt in the gall-bladder or bile-ducts, the peritoneal toilet was per- formed and the abdomen closed, recovery following and ending in complete restoration to health. It is worthy Acute Pancreatitis and Subacute Pancreatitis 403 of note that in this case the diagnosis was confirmed before operation by the urinary pancreatic reaction. A case was reported by Dr. Charles D. Muspratt, of a woman, aged forty years, who had been admitted to the Royal Victoria Hospital, Bournemouth, on December 3. 1903, in a state of collapse and suffering from severe ab- dominal pain with incessant vomiting. The abdomen was opened within twenty-four hours of the onset of acute symptoms and the omentum and intestines in the neighbourhood of the pancreas were found deeply blood- stained with numerous spots of fat necrosis. The pan- creas was almost purple and extremely tense. An incision was made into the dark gland and very free bleeding fol- lowed which was arrested by ligature. Gauze drainage was employed and complete recovery followed. This is apparently the first case in which direct incision of the pancreas was adopted, and the operator is to be congratulated not only on having had the strength of his convictions in treating acute hemorrhagic pancreatitis on the lines of other phlegmonous inflammations, but on the success of such treatment. In a case reported by von Mikulicz in 1903, a patient, under the care of Dr. C. B. Porter, of Boston, was operated on by a deep in- cision into the inflamed gland, with an excellent result. This is apparently the second case in which the pancreas was deliberately incised during acute inflammation with a successful result. Woolsey gives a summary of three cases of this affection successfully dealt with by laparot- omy and drainage. The first two cases were operated on in the early stage — the first on the third day and the second twelve hours after the onset. The first case was a hsemorrhagic one and showed fat necrosis; the second case showed no fat necrosis or bloody fluid, but the latter appeared on the removal of the gauze drain two days after the operation. In the third case there was marked but temporary glycosuria. Dr. C. G. B. Kempe, of Salisbury, on December 11, 1902, excised a portion of the head of 404 The Pancreas: Its Surgery and Pathology the pancreas affected with acute haemorrhagic pancrea- titis. It was done within two hours of the onset of haem- orrhage. The patient unfortunately died from diarrhoea fifteen days later. The argument that the percentage of mortality will be less if the surgeon waits for the for- mation of a local abscess is fallacious, as it takes no con- sideration of the large percentage of those who die before such a favourable result is presented; and, in the second place, many patients never develop a local abscess, the process being diffuse from the onset. The high mortality of early operation in acute cases is due to the fact that in many of these fatal instances intestinal obstruction was suspected and the collapsed patients were subjected to a prolonged search for the seat of the supposed lesion. Of fifty-nine reported cases of operation during the acute stage, twenty-three recovered ; these include the cases just described. Although this is a large mortality, it must be borne in mind that the disease is a lethal one and usually ends in death if not treated surgically. The lessons which one may learn from recorded cases are not to wait until the system is over-weighted with absorbed poison before operating and not to spend too long a time over the operation. SUBACUTE PANCREATITIS Although no hard-and-fast line can be drawn between acute and subacute pancreatitis, yet the less acute onset, the longer course, the limitation of the suppurative proc- ess by lymph barriers, and the much more hopeful out- look as the result of treatment present so many differences that clinical observers will acknowledge that such a divi- sion is desirable, from the point of view of both diag- nosis and treatment. Acute pancreatitis seems to bear the same relation to subacute pancreatitis that a diffuse mas- titis does to a simple abscess of the breast, or a diffuse suppurative parotitis to a simple parotid abscess. Acute Pancreatitis and Subacute Pancreatitis 405 It may have a more or less sudden onset, with acute pain and vomiting, and may be associated with constipation, but collapse is not a marked symptom and is, as a rule, absent. The upper abdominal region does not become so rapidly distended and vomiting is less severe and less prolonged. At other times, and this is generally the case, the onset is more gradual, though the symptoms may be similar. As gall-stones are the usual cause of this form of pancreatitis, a history of intermittent attacks of spasms, at first without and later accompanied by jaundice, will be elicited, and before the onset of pancreatic trouble the symptoms of infective cholangitis, in the shape of rigors with deepening of jaundice and with intermittent fever, will generally be found. Tenderness over the pancreas is well marked, and on account of the tympanites being less than in the acute form, it may be possible to feel the swollen gland, especially under an ansesthetic, and as the case progresses a definite tumour often develops. Con- stipation gives place to diarrhoea, and pus or blood may be noticed in the stools, which have a very foetid odour and usually contain fat and undigested muscle fibres. The pulse is not so seriously affected as in the acute form and the temperature is more irregular. The temperature may reach 104° or 105° F. and yet the pulse may only vary between 70 and no. The morning temperature may be normal and the evening temperature high for several days or even weeks. Rigors or chills usually occur and may be repeated from time to time. The pain occurs in paroxysms, but there is also a constant dull pain at the epigastrium. The patient may lose the more urgent symptoms and appear to be really improving, but the loss of flesh and feebleness continue, and relapses usually occur, leaving the patient each time more and more feeble until death supervenes from asthenia. Albu- minuria is pretty constant, but glycosuria is rarely present. The pancreatic reaction in the urine is, as a rule, well 4o6 The Pancreas: Its Surgery and Pathology marked. If an abscess develop, the pus may form a tumour projecting in the superior abdominal region and forming a tender swelling behind the stomach, or per- haps coming to the surface above or below that viscus; or it may burrow into either loin, forming a perirenal abscess, or passing under the diaphragm it may form a subphrenic abscess. Occasionally the pus may follow the psoas muscle and form a subperitoneal abscess in the iliac region, or even passing over the brim of the pel- vis it may collect in the left broad ligament. Sometimes the abscess bursts into the stomach and is vomited, or into the bowel and is voided per anum, after Fig. 130. — Abscess of the pancreas. which diarrhoea may continue and pus may be seen from time to time as any fresh collection forms and bursts. With the evacuation of the abscess, relief occurs for a time and^ the temperature improves, but relapses usually take place and a mild form of septicaemia persists with a hectic temperature. Death is the usual termination unless an operation be done, though spontaneous recovery may possibly occur after a tedious and prolonged illness should the abscess burst into the bowel or be otherwise safely evacuated. For the diagnosis of subacute pancreatitis in its initial Acute Pancreatitis and Subacute Pancreatitis 407 stages little need be added to what has already been said when considering suppurative catarrh and acute phleg- monous pancreatitis. The presence of a tumour or of a diffuse epigastric swelling behind the stomach will be generally found, or if epigastric tenderness prevents pal- pation, an anassthetic will enable the swelling to be felt. There is usually fever of a septic type. The presence of leucocytosis and the discovery of the pancreatic reaction in the urine will afford valuable aids to diagnosis. As soon as an abscess forms it may reach the surface above or below the stomach, in either loin, in the left iliac region, under the diaphragm, or even in the pelvis, and will require differential diagnosis from other conditions lead- ing to a collection of pus in those situations, such as chronic perforative gastric or duodenal ulcer, suppurative chole- cystitis, splenic abscess, perirenal abscess, spinal abscess, glandular abscess, etc. Treatment. — The subacute form of pancreatitis is more amenable to treatment, as the indications are so much more definite and there is more time for careful considera- tion. Though it has usually been attacked only when an abscess has formed and is manifestly making its way to the surface, yet there is no reason why in some cases surgical treatment should not be adopted at an earlier stage. As in the acute condition, morphine may be re- quired to relieve the pain and to lessen the collapse. Distension, if present, demands attention, and may have to be relieved by lavage of the stomach and turpentine enemata or by the administration of calomel by the mouth. Calomel is also of benefit as an intestinal antiseptic, for which purpose it may be given in small repeated doses followed by a saline aperient. As soon as the constipa- tion is relieved, diarrhoea is apt to supervene, when salol and bismuth, with small doses of opium, may be given. If surgical treatment is decided on, a median incision above the umbilicus will enable the operator to palpate 4o8 The Pancreas: Its Surgery and Pathology the pancreas and to locate any incipient collection of pus, which, if practicable, should then be evacuated by a posterior incision in the left or right costo-vertebral angle. If the posterior incision be thought impracticable, the collection of pus may be removed by aspiration and the cavity opened and packed with gauze, which may be brought forwards through a large rubber tube, which procedure will, in the course of from twenty-four to forty- eight hours, establish a track isolated from the general peritoneal cavity. In abscess of the pancreas, which usually assumes the form of subacute pancreatitis, and which we must distinguish from the acute suppurative pancreatitis where the pus is diffused through the gland or where the ab- scesses are small and multiple, the suppurating process is limited by a pouring out of lymph, so that should the patient survive the initial more acute stage and discovery of the pus-containing cavity be made, the condition is one decidedly amenable to treatment by drainage. The anatomical relation will readily explain the course along which the pus burrows should it burst through its lymph barriers — for instance, in one case an abscess formed and was opened in the right loin of a young, man, aged twenty-four years, that had been mistaken for a perirenal abscess, yet the kidney was quite healthy and the grumous pus had come from the pancreas and passed behind the peritoneum covering the second part of the duodenum. The patient recovered completely. In another case an abscess was opened in the left iliac region that had apparently started from the body of the pancreas and which had burrowed in the same way be- hind the peritoneum. The patient recovered from the operation, but developed trouble in the left side of the thorax and died suddenly several weeks later. In one case the abscess was subphrenic. In another, where the symptoms were rather acute and the patient was ex- Acute Pancreatitis and Subacute Pancreatitis 409 tremely ill, pus was discovered between the liver and the stomach, and although drainage was apparently complete, the patient succumbed in a few days to exhaustion due to the septic process that had been initiated before the abscess was opened. In two other cases, the sequence of suppurative catarrh, abscesses of the pancreas were successfully drained through a tube in the common bile- duct after removing the gall-stones which had obstructed Wirsung's duct. In one of these cases the patient, a woman, aged seventy-two years, remained quite well, but in the other a man, aged forty years, recovered from the operation, but three months afterw^ards died from exhaustion, and at the necropsy the empty abscess cavity was discovered in the head of the pancreas, the rest of the gland being affected with chronic interstitial inflam- mation. In one case — a man, aged thirty-five years^a pancreatic abscess burst into the stomach, setting up acute gastritis, the condition having been proved b}^ an exploratory operation. It was treated by gastro-enteros- tomy to drain away the foul stomach contents. The patient is now quite well, eight years later. In another case, in a young married woman, aged twenty-six years, the abscess apparently burst into the bowel, and, though recovery was tardy, she ultimately got quite well without operation. The diagnosis was made from the symptoms and by an examination of the swollen pancreas under an anaesthetic and subsequently by the presence of a pan- creatic reaction in the urine. It is important in these cases to see that the cause is removed, if that be possible — ■ for instance, gall-stones or pancreatic calculi — so that if recovery occurs there may be no fear of relapse. It will thus be seen that out of eight cases of abscess of the pancreas, one of which was mentioned under acute pancreatitis, six were operated on, with recovery from operation in five, though in one of the cases the relief was only for a few weeks and in another for a few months. 4IO The Pancreas: Its Surgery and Pathology In the eighth case, which was not operated on, the abscess burst into the bowel and was discharged, the diagnosis having been made by an examination of the tumour under an aneesthetic and by the presence of the pancreatic reaction. When inflammation of the pancreas has ended in ab- scess, chronic interstitial pancreatitis will also probably be present, as was shown at the necropsy of one case that died some months subsequently. It is possible that in some cases the interstitial change may be local, though in others it will be general and may then lead to atrophy of the gland and to glycosuria. A search through literature reveals a considerable number of pyasmic abscesses of the pancreas, but those resulting from subacute pancreatitis have been rarely recorded. Besides seven operations for abscess of the pancreas with two deaths above referred to, there have been seven others recorded with three deaths. Thus, of fourteen cases five died, giving a mortality of 35.6 per cent. Literature Babler: St. Louis Courier of Med., Nov., 1905. Baines: Brit. Med. Journ., Feb. lo, 1906. Barling: Brit. Med. Journ., Dec. 22, 1900; Ibid., Feb. 24, 1906. Brennecke: Journ. Amer. Med. Assoc., June 4, 1898. Brown, W. H.: Lancet, Sept. 26, 1903; Ibid., Aug. 13, 1904. Bryant: Lancet, Nov. 10, 1900. Bunting: Johns Hopkins Hosp. Bull., Aug., 1906. Classen: " Krankheit. der Bauchspeicheldruse," Koln, 1843. Cooke: Brit. Med. Journ., May 19, 1906. Deanesly: Lancet, July i, 1899. Deaver: Med. News, March 5, 1904. Deaver and Muller: American Medicine, March 19, 1904. Drasche: Ber. der k. k. Krank., Wien, 1886. Durno: Lancet, Nov. 10, 1906. Earl: Brit. Med. Journ., Nov. 17, 1906. Fison: Lancet, June 4, 1904. Fitz: Medical Congress, Washington, 1903. Fletcher: Prov. Med. and Surg. Journ., 1848. Friederich: Ziemssen's "Handbuch d. spec. Path. u. Therap.," viii, 2, 1878. Fripp and Bryant: Lancet, Dec. 17, 1898. Frison: "Recueil. de Med. militaire," 1876. Fuchs: Deutsch. med. Wochen., 1902, xxviii, 829. Acute Pancreatitis and Subacute Pancreatitis 411 Greisclius: Misc. Acad, curios, 1672, 1673, P- 74- Hahn: Deutsche med. Wochen., 1901, v, 5. Halley: Scottish Med. and Surg. Journ., Jan., 1904. Halsted: Johns Hopkins Hosp. Bull., 1901, Nos. 121, 122, 123. Harvey: Brit. Med. Journ., Nov. 17, 1906. Heaton: Brit. Med. Journ., Dec. 17, 1904. Hogarth and Moynihan: Practitioner, 1903, i, 504. Jeffrey: Lancet, Jan. 20, 1906. Jones, Littler: Lancet, Feb. 18, 1905. Kempe: Brit. Med. Journ., Feb. 27, 1904. Kennan: Brit. Med. Journ., Nov. 14, 1896. Keyser: Lancet, Oct. 19, 1901. Kilgow: Lond. Journ. of Med., i860, p. 1052. Klob: Oester Zeitsch f. prakt. Heilk., i860. Korte: Deutsche, med. Woch., 1901, v, 6. Leusden: Charitd Ana., xxxi, 1902. Lilienthal: Annals of Surgery, Jan., 1906. Lund: Boston City Hosp. Rep., Dec, 1900. Mayo: Journ. of Amer. Med. Assoc, 1902, i, 107. Moore: Trans Lond. Path. Soc, 1882, xxxiii, 186. Morian: Miinch. med. Woch., March 14, 1899. Munster: Lancet, Dec. 30. 1905. Muspratt: Brit. Med. Journ., Feb. 3, 1901; Ibid., Feb. 6, 1904. Norris: Lancet, Dec. 9, 1905. Nothnagel: "Encyclo. of Prac Med.," " Dis. of Liver, Pancreat.," etc., Eng. tr., 1903. Opie: "Diseases of the Pancreas," 1903. Osier: "Practical Medicine." Pauchet: Rev. de Gynec. et de. Chir. Abdom., Nov., 1905. Peiser: Deut. Zeit. f. Chir., 1902, Ixv, 302. Percival: Trans. Ass. K. and Q. Phy. Ireland, 18 18. Perle: Diss. Berlin, 1807. Pitt and Jacobson: Med. Press and Circ, Dec. 14, 1898, p. 619. Portal: Cours d'Anat. Medicale, v, 352. Reynolds: Med. Chron., Aug., 1900, p. 328. Riboli: Gaz. de Sard., 1858." Robson, Mayo: Hunterian Lectures, Lancet, Mar. 19, 26, April 2, 1904. Roddick: Canadian Med. Journ., 1869. p. 385. Selberg: Berlin klin., 1901, xxxviii, 923. Shea: Lancet, 1881, p. 791. Smith: Dublin Journ. of Med. Sci., i86"o, p. 201. Stewart: Annals of Surgery, Aug., 1906. Thomas: Medical Press, Jan. 30, 1907. Toye: Brit. Med. Journ., Jan. 27, 1906. Webber: Lancet, Aug. 5, 1905. Woolsey: Med. News, Dec. 20, 1902; Annals of Surgery, Nov., 1903. Young: Lancet, Feb. 10, 1906. CHAPTER XVI CHRONIC PANCREATITIS In a lecture delivered at the London Medical Grad- uates' College and Polyclinic in July, 1900, attention was drawn by one of us to chronic pancreatitis as a clinical entity. An opinion was then expressed that although it had hitherto been scarcely, if at all, recognised, except as a pathological curiosity, it was not an uncommon disease. Since that statement was made further ex- perience has still more conclusively demonstrated the frequency with which the disease occurs, and how rarely its presence is recognised in clinical practice. Chronic inflammations of the kidneys, liver, and other organs give rise to symptoms which are described in every text-book ; but the no less important and characteristic signs of chronic pancreatitis are, even in the most recent medical and surgical works, entirely omitted, or awarded but scanty and inadequate notice. It is true that the relation of the pancreas to diabetes is now generally recognised, but to wait until glycosuria supervenes before making a diagnosis is to throw away the patient's only opportunity of cure — an opportunity which, if taken sufficiently early, will in nearly all cases restore him to perfect health, and save him from a lingering and painful illness. There appears to be a general belief in the profession that the symptoms of chronic pancreatitis are so over- shadowed by those of other morbid states from which it may arise, or with which it may be associated, that it can rarely be recognised during life as a clinical entity, and that even then it is usually only revealed at operation •undertaken for the removal of gall-stones or for other 412 Chronic Pancreatitis 413 obstructive condition in the bile-passages. Opie, in his work on diseases of the pancreas (p. 163), states that " the lesion is seldom associated with such definite symptoms as to be recognised during life, and that even at autopsy the condition is frequently overlooked." The former part of this statement does not hold good at the present day; for from the inform£ition obtained by a careful ex- amination of the patient, a knowledge of the history of the case, and the results of a chemical and microscopical examination of the excreta, a correct opinion may be formed in a large majority of instances. The latter part of Opie's statement, however, is as true today as when it was written. As evidence of the difficulty that even skilled pathologists have experienced in recognising pan- creatic disease by naked-eye examination alone, one need only compare the statistics compiled by Hale White from the Guy's Hospital post-mortem records for the fourteen years 1884- 189 7, and the results of the micro- scopical examination of the pancreas in a series of con- secutive cases by Bosanquet (page 127). From these it is evident that at present it is impossible to rely on post-mortem records, either ancient or even recent, for precise information as to the frequency with which chronic inflammation occurs, and that "unless a microscopical examination of the pancreas is made, it is frequently im- possible to say in any case whether it is normal or not, since in many instances the external appearance of the gland may be almost unchanged in the presence of con- siderable alteration in its anatomy" (Bosanquet). The surgeon has considerable advantage over the path- ologist in this respect, for he has the opportunity of exam- ining and handling the living pancreas, and, after some experience, he can generally tell the difference between the feel of the normal and diseased gland. A chronically inflamed pancreas is generally swollen and harder than usual. A typical case conveys to the examining hand 414 The Pancreas: Its Surgery and Pathology the impression of a hard waxen cast. This may be recognised as a general swelling of the whole gland, or as a limited swelling involving the head and body of the gland. The swelling and hardness, especially in the early stages of catarrhal inflammation, are, no doubt, due to engorgement with blood and retained secretion ; but as in many cases this will largely disappear after death, the difference noticed in biopsies and autopsies is easily explained. The lobules of the gland have the feeling of being mapped out and differentiated in a manner very different from their state in the ordinary healthy organ. The irregularity and change of consistency is occasion- ally so marked that, to the inexperienced, it may suggest malignant disease. In fact, in some of the earlier cases of chronic pancreatitis that came under our notice this was the idea that suggested itself on examining the pan- creas during operation, and it was only the subsequent uneventful recovery of a number of these patients, fol- lowed by an immunity from any further symptoms of disease — an immunity which in some instances has now extended to a considerable number of years — that con- firmed the suspicion that the condition was purely inflam- matory. The first case occurred in 1890, and the patient, a woman ast. forty-four, is now well, seventeen years later. Since the third case, in 1892, which was confirmed by autopsy and by a microscopical examination, the oppor- tunity has been taken of examining the pancreas in a very large number of cases during operations in the upper ab- domen, and in many of these there was little doubt that disease of this organ contributed to, or was the cause of, the symptoms complained of. The marked, and in many cases striking, relief that followed appropriate operative interference conclusively demonstrated the importance of the condition ; and we can point to a large number of patients, now in perfect health, who before operation were extremely ill, and in many cases supposed to be suffering Chronic Pancreatitis 415 from malignant disease of the pancreas, chronic catarrh of the bile-ducts, cirrhosis of the liver, cancer of the com- mon bile-duct or of the papilla, cancer of the liver, com- mon-duct cholelithiasis, malaria, and other diseases. The number of cases of chronic pancreatitis reported in the journals since attention was called to the subject in 1900 shows that others are now recognising the condi- tion, and that, where suitable treatment has been adopted, Fig. 131. — Chronic interstitial pancreatitis of the interlobular variety (Santos). satisfactory results have followed. Chronic pancreatitis has also been reported as occurring spontaneously in the lower animals. Megnin and Nocard have described the disease in a horse, which, during life, suffered from weak- ness, loss of appetite, constipation, emaciation, and slight icterus. Post-mortem the pancreas was found to be indurated, the duct was dilated and filled with albu- minous material, and the common bile-duct was com- pressed, there was also catarrh of the salivary duct. 41 6 The Pancreas: Its Surgery and Pathology The results of chronic inflammation of the pancreas, as seen on the post-mortem table and following experi- ments on animals, may be divided histologically into: (i) Chronic interstitial interlobular pancreatitis; (2) chronic interstitial inter acinar pancreatitis; (3) cirrhosis of the pancreas. In the interlobular form of chronic interstitial pan- creatitis the normal loose connective tissue between the lobules of the gland is converted into dense sclerotic material, the glandular tissue is com- pressed, and replaced frorn the periphery of the lobule, by newly formed connective tis- sue, the normally ob- scure lobules becoming distinctly defined. In the interacinar va- riety a diffuse network of irregular fibrous tis- sue is found separating the glandular acini, and in some instances pene- trating between the in- dividual cells, while the interlobular tissue is comparatively little affected. The gland is tough rather than hard, and the nodular character seen in the inter- lobular form is lacking. Cirrhosis of the pancreas is the final stage of either inter- lobular or interacinar pancreatitis, but it more commonly occurs as a result of chronic interlobular inflammation, of which glycosuria is a rare sequel, not occurring till the lesion has so far advanced that the glandular acini are Fig. 132. — Fibrosis of the pan- creas (St. Thomas' Hospital Museum, 1413 a). Chronic Pancreatitis 417 almost completely destroyed, and the vascular supply of the islands of Langerhans is seriously interfered with by the pressure of the newly formed fibrous tissue. In the interacinar form, on the other hand, the cell islands are involved at a very early stage, diabetes quickly super- venes, and it may prove fatal before the cirrhosis has be- come very marked. A marked new-formation of fibrous tissue in the pancreas appears to be a comparatively late result of chronic inflammation, and unless this fact is borne in mind a very incomplete conception of the condi- tion is liable to be formed. Experimental liga- ture of the ducts in animals has shown that inflammatory atrophy and degener- ation of the secreting parenchyma precede and accompany the formation of the new fibrous tissue which takes place, and in chronic pancreatitis due to obstruction of the common duct in man the sequence of events is no doubt the same. In the early stages the organ will be engorged with blood and retained secretion, the parenchyma will show cloudy swelling and other degenerative changes, there will be some leucocyte infiltration, and the ducts will be dilated ; but although such an organ may during life be distinctly enlarged and hardened, after death it may appear to be normal or nearly normal to the naked eye, and even microscopically will show no increase of fibrous tissue. The fact that obstruction of the pancreatic duct with 27 Fi-. 133- -Cirrhosis of the (X ca 40). pancreas 41 8 The Pancreas: Its Surgery and Pathology damming back and infection of the secretion gives rise to chronic pancreatitis has already been mentioned. In practice this is by far the most common cause of the disease, and it is found to be due in most cases to the lodgment of a gall-stone in the lower portion of the com- mon bile-duct. The reason for the association of the two conditions is obvious when the anatomy of the parts is considered. The common bile-duct, starting by the junction of the cystic duct and hepatic duct, courses along the free border of the lesser omentum associated with the portal vein and hepatic artery; it then passes behind the first portion of the duodenum, and soon comes into relation with the pancreas, which it either grooves deeply, or passes through or behind, before it pierces the wall of the second part of the duodenum, where it empties into the diverticulum of Vater along with the duct of Wirsung. It may be divided into four portions: (a) The supraduodenal portion, (6) the retroduodenal portion; (c) the pancreatic portion; (d) the intraparietal portion. The supraduodenal and the retroduodenal sections of the duct are unimportant for our present purpose, but the relations of the pancreatic and the intraparietal por- tion of the duct require careful consideration. The fourth, or intraparietal, segment of the common duct comprises all that portion of the canal contained in the thickness of the wall of the duodenum. It passes obliquely through the muscular coat of the intestine, and then dilates into a little reservoir underneath the mucous membrane into which the main pancreatic duct also opens, known as the ampulla of Vater. The ampulla opens into the duodenum by a little round or elliptical orifice, which is the narrowest part of the bile-channel. The mode of formation of the ampulla of Vater and the termination of the common and pancreatic ducts are liable to at least six variations. These have already been Chronic Pancreatitis 419 considered in connection with the anatomy and anatomi- cal abnormaUties of the gland, but as they have such an important bearing on the subject of chronic pancreatitis it will not be out of place if they are again summarised here in this connection. The first type is the classical one, which is described above. In the second type the pancreatic duct joins the common duct some little distance from the duodenum, the ampulla of Vater is absent, and the duct opens into the duodenum by a small, flat, oval orifice. In the third type the two ducts open into a small fossa in the wall of the duodenum, while the caruncle and the ampulla of Vater are absent. In the fourth type the caruncle is well developed, but the ampulla is absent, the two ducts opening side by side at the apex of the caruncle. In the fifth type the common bile-duct opens along with the duct of Santorini, and Wirsung's duct enters the duode- num separately. In the sixth type the pancreas has three separate ducts opening into the duodenum, one only accompanying the common bile-duct. It will be seen that, while the normal termination and the second variety of termination of the ducts will favour the onset of pancreatitis in case of common-duct chole- lithiasis, the variations 3 and 4, in which the two ducts are separate, will possibly save the patient from the serious secondary pancreatic troubles, and in variation 5 and 6 a small portion only of the gland will become infected. But the pancreatic ducts themselves are also subject to great variations that may influence the course of events. The result of observations by Opie on 100 cadavers (Figs. 30, 61) in which the ducts were injected and photo- graphed was as follows : In ninety-nine specimens the two ducts were united; in ten, two wholly independent ducts entered the intes- tine. 420 The Pancreas: Its Surgery and Pathology 1 . Of the ducts in anastomosis : (i) Duct of Wirsung was the larger in eighty-four. (a) Duct of Santorini patent in sixty-three. (6) Duct of Santorini not patent in twenty-one . (2) Duct of Santorini larger in six. (a) Duct of Wirsung patent in six. (6) Duct of Wirsung not patent in any. 2. Ducts not in anastomosis in ten. (a) Duct of Wirsung larger in five. (6) Duct of Santorini larger in five. In 89 per cent, the duct of Wirsung was larger than the duct of Santorini, while in 2 1 per cent, the duct of Santo- rini was apparently obliterated near its termination. In six cases the duct of Santorini was larger than the duct of Wirsung. In all cases where the duct of Santorini was patent it diminished in size towards the duodenum. Thus the duct of Santorini cannot be relied on in many cases to supplement the duct of W-irsung, if it be ob- structed ; moreover, the duct of Santorini, even if patent and communicating with the duodenum, may itself be compressed by a moderate-sized gall-stone passing down the pancreatic portion of the common duct. It might be argued that if the two ducts communicate, why should not the duct of Santorini act as a safety-valve to the duct of Wirsung when it is compressed, and thus free the pancreas from the retained secretion, which is in danger of becoming septic? It will be seen that in only half or less than half of all cases will the duct of Santorini act as a safety-valve if the duct of Wirsung is obstructed, for although in 63 per cent, of cases the duct opens at the same time into the main channel and into the intestines, yet in probably less than half of these is the anastomosis efficient as a through channel. The reasons why gall-stones in the common bile-duct do not always produce pancreatic inflammation are: (a) Some gall-stones are so large that they never Chronic Pancreatitis -421 reach the pancreatic portion of the duct, but remain in the supraduodenal portions of the common duct, pro- ducing jaundice but no pancreatitis. (b) In some cases the bile-ducts and pancreatic ducts open by separate orifices, and any gall-stone passing down the common duct will not then necessarily compress or occlude the pancreatic duct, (c) In exceptional cases the duct of Santorini is the principal outlet for the pancreatic fluid, it being of such a size as to affor^"! a safe outlet to the secretion even when the duct of Wirsung is obstructed. The course of the third or pancreatic portion of the common duct is also of great interest, for if it passes through the gland, any congestion or swelling of the head of the pancreas will, by the pressure it exerts on the common duct, tend to induce jaundice and its various sequelae; whereas if it passes behind, and not through, the head of the gland, it will escape from pressure when the pancreas is inflamed. The passage of the common duct through the substance of the pancreas in a certain proportion of individuals probably explains many of the cases of so-called catarrhal jaundice, which may come on as an extension from gastro- duodenal catarrh, or in the course of various ailments, and which it is not unlikely are often dependent on catar- rhal inflammation and swelling of the pancreas, leading to pressure on the bile-ducts. Such are many of the cases of acute jaundice, especially the form coming on in young subjects, and which usually clear up under medical treat- ment. They are truly pancreatic and not biliary in origin, and some of these cases pass on from the simple 'congestive or catarrhal form to true interstitial pancreat- itis. The so-called chronic catarrh of the bile-ducts lead- ing to persistent jaundice is nearly always due to chronic pancreatitis, the obstruction to the flow being outside and not inside the common duct. 422 The Pancreas: Its Surgery and Pathology As the duct is completely embraced by the pancreas in 62 per cent, of all cases, we may conclude that in about that proportion of cases a swelling of the head of the pan- creas will produce jaundice, and, as supporting this view, this percentage corresponds with our clinical and path- ological investigations of the urine of pancreatic cases, when associated with gall-stones in the common bile- duct. A gall-stone passing down a duct thus embraced by the pancreas is almost certain to exert pressure on the gland, and the resulting inflammatory changes may in their turn give rise to compression of the duct, which will result in jaundice that may persist long after the gall-stone itself has passed. Occasionally the ducts are obstructed by other causes than gall-stones. A growth occurring in the ampulla of Vater, or in the papilla, will interfere with the free flow of the pancreatic secretion, and may give rise to catarrh of the ducts and chronic pancreatitis. An impacted pan- creatic calculus, or stenosis of the duodenal opening of the duct following ulceration, will also produce a similar result ; and recently a case has been reported in which a portion of hydatid membrane was the oostructing agent. How far the pancreatic lesion in these obstruction cases is to be attributed to the irritating action of the retained secretion, and how far to the associated bacterial infection, is difficult to say, but it is probable that in all cases the latter plays an important part. Even when the blocking of the ducts is complete, and no direct com- munication between the micro-organisms in the duodenum and the stagnant secretion appears to be possible, the inflamed walls of the duct present a ready path for the passage of infection. This has been proved in the bile- passages by aseptic ligature of the common duct. Abso- lutely complete blocking of the duct is, however, very uncommon, except in cancer cases, for bile-pigment can Chronic Pancreatitis 423 be found chemically in the faeces in nearly all other cases, even when the stools are free from colour to the naked eye. Chronic pancreatitis may result from a direct exten- sion of a duodenal catarrh to the pancreatic ducts, and this association of chronic pancreatitis with duodenal catarrh is not at all uncommon. It has been shown experimentally that by injecting fsecal material or bacil- lus coli into the pancreatic ducts, or by providing a chan- nel, such as an absorbent thread, by which organisms may enter from the bowel, pancreatitis is produced; and that after some time the gland, which is constantly being infected by a permanent channel, eventually be- comes sclerosed. We have had the opportunity of inves- tigating a considerable number of cases in which pancrea- titis has followed chronic gastric or gastro-intestinal catarrh and duodenal ulcer, and in many of them the condition has been relieved by operation. The pancreatitis which is occasionally met with as a sequel of typhoid fever is probably due to a specific infection occurring in a similar way, though infection by way of the blood and changes due to toxaemia cannot be excluded. In support of this hypothesis is the fact that typhoid bacilli have been recovered from the bile in the common duct and gall-bladder in some cases where operation has been undertaken. Influenza and some other zymotic diseases are also occasionally followed by inflammation of the pancreas. In these cases infection may take place by way of the duo- denum, or possibly through the blood. The chronic infections, tubercle and syphilis, may also give rise to pancreatitis. In the former chronic inflam- matory changes may be found in the absence of definite tuberculous deposits, and the experiments of Carnot with tubercle bacilli and tuberculin in dogs suggest that the changes are due to toxic substances circulating in the blood, rather than to the direct effect of the bacillus in 424 The Pancreas: Its Surgery and Pathology the gland. It is probable that syphilitic pancreatitis is a similar toxic manifestation. The influence of alcoholism in the production of cirrho- sis of the liver is still a debatable point, and similarly its relation to chronic pancreatitis has not been settled. In some cases a history of alcoholic excess can be obtained, but in many this is not so. It is probable that alcohol is not of itself a direct determining cause, but that indirectly, by the influence it exerts on the circulation and by the production of a catarrh of the duodenum, it may give rise to pancreatitis. Chronic pancreatitis and cirrhosis of the liver are not infrequently associated. Ac- cording to the observations of Lefas and Opie, chronic pancreatitis may accompany either atrophic or hypertro- phic cirrhosis, but while in the former the gland is en- larged and the newly formed fibrous tissue interacinar in distribution, in the latter no marked increase of size takes place, and it is the interlob- ular tissue that is increased in amount and density. It is not uncommon to find both diseases present in long-continued obstruction of the common duct by gall-stones. In that peculiar condition hcemochromatosis, the pan- creas is affected, and chronic interacinar pancreatitis folows the deposit of pigment and associated atrophy of the gland cells. As in the kidney and other organs, an increase of fibrous tissue occurs in the pancreas in general arteriosclerosis Fig. 134.— Chronic ulcer of the posterior wall of the stom- ach eroding the pancreas (R. C. S. Museum 2399). Chronic Pancreatitis 425 rlJ-^ ?^^*«^ Fig. 135. — Section of the pancreas in the neighbourhood of p,n adherent gastric ulcer, showing the secondary interstitial pancreatitis (X 40). and endarteritis, and it is possible that the moderate increase found micro- scopically in a certain number (10 per cent. of Bosanquet's cases) of patients over forty years of age may be attributed to this cause, but that it is not a common cause is shown by Opie's investigations. Occasionally one meets with chronic pancreatitis due to direct extension of the inflammatory process from a neigh- bouring organ, such as a chronic gastric ulcer eroding the gland. A malignant growth of the py- lorus may set up perigastritis and cause the stomach to become adher- ent to the pancreas, producing well- marked interstitial pancreatitis in the head of the gland, as in a case oper- ated on by one of us recently. Acute or sub- acute inflammation of the pancreas, if not ending fatally. Fig. 136. -Chronic suppurative pancreati- tis (X ca 35). 426 The Pancreas: Its Surgery and Pathology may resolve and be succeeded by chronic inflammatory changes that may ultimately lead to cirrhosis and to a fatal termination at a later date, as was demonstrated in one of our cases of abscess of the pancreas that died three months after operation, in which an opportunity occurred of examining the gland microscopically. In cystic disease of the pancreas, chronic pancreatitis is nearly always present ; in fact, it is probably the com- pression of the smaller duct by the contracting newly formed fibrous tissue that in many cases gives rise to the cysts. All the causes of chronic pancreatitis mentioned, except atrophic cirrhosis and ha^mochromatosis, are asso- ciated with interlobular changes in the fibrous tissue, and from a surgical point of view this is the most important form of chronic inflammxation to which the gland is liable, for it is capable of being distinctly benefited by operative interference, if recognized at a sufficiently early stage. The etiology of the interacinar variety is at . present for the most part obscure, although, judging from the early stages at which the islands of Langerhans are affected and the centrifugal character of the new fibrous tissue formation, it is probable that the disease owes its origin to an abnormal state of the blood. It is therefore not likely to prove directly amenable to surgical treatment. Symptomatology. — The onset of chronic pancreatitis varies with the cause. If it is due to obstruction of the common duct by a gall-stone, there will be a history of painful attacks in the right hypochondrium and in the epigastrium, associated with jaundice and possibly accom- panied by fever of an intermittent type. Tenderness at the epigastrium, with some fulness above the umbilicus, will usually be noticed ; loss of flesh soon becomes marked, and if the pancreatic symptoms predominate, the pain will pass from the epigastrium round the left side even to the renal and scapular regions. Chronic Pancreatitis 427 If, however, the condition arises by direct infection from a duodenal catarrh or from one of the other causes mentioned, and not connected with choleHthiasis, there may be merely an aching in the epigastrium or slight pain not at all pronounced, or the symptoms may come on painlessly, associated with dyspepsia and with slight jaundice, soon becoming more marked; in such cases the gall-bladder may dilate and give rise to a suspicion of cancer of the pancreas which the rapid loss of flesh will tend to confirm. In either case a train of symptoms of a very definite character is set up, and it is difficult to un- derstand how the idea has gained currency that chronic pancreatitis is, as a rule, undiagnosable during life. Physical examination of the patient will reveal in some few cases a swelling of the pancreas due to tumefac- tion of the head of the gland ; but as the recti are often rigid from the pain and tenderness in the epigastrium, it may be discoverable only when the patient is anaesthe- tised. . Pain and tenderness, though usually present, may be little marked, but in some cases the pain is paroxysmal and severe, and epigastric tenderness is well pronounced. By distending the stomach with gas, either by means of carbonate of soda and tartaric acid given in separate doses, or by pumping in air through the stomach-tube, the relation of the stomach to the swelling can be readily made out. Resonance on percussion, owing to the posi- tion of the stomach, unless the stomach is empty, com- municated non-expansile pulsation, and very slight move- ment on deep inspiration are characteristic. In the more chronic stages, especially when the disease has reached the cirrhotic stage, if the cause be not gall- stones, a tumour of the gall-bladder is found similar to that met with in cancer of the pancreas. The distension is due to mucus, the bile which first filled it having been absorbed and the backward pressure having prevented 428 The Pancreas: Its Surgery and Pathology fresh bile from entering the duct. The distension may occur so gradually as to be painless, and the gall-bladder is then free from tenderness, which is less frequently the case in distension due to gall-stones. In jaundice due to a stone in the common duct the gall-bladder is nearly always contracted and not capable of being felt. Jaundice is not necessarily present at first, although it is usually met with at some stage of the disease, and is often well marked. It may vary from a slight icteric tinge, most marked in the sclerotics, to an intense mahog- any hue. In chronic pancreatitis, due to obstruction of the common bile-duct by a gall-stone, the jaundice is frequently very marked, as it is also in those cases where the duct passing through, or grooving, the head of the pancreas is compressed by the swollen gland ; when, how- ever, the common bile-duct passes behind the gland, as it does in 38 per cent, of bodies, the patency of the passage may not be seriously interfered with and little or no jaun- dice be produced. An increase of temperature is, as a rule, associated with acute and subacute pancreatitis, but only rarely in any of the more chronic forms of inflam- mation, except in those cases where there is associated infective cholangitis, as in obstruction of the common duct by a biliary calculus and infection of the retained secretion, when there may be fever of an intermittent type and ague-like paroxysms. Dyspeptic disturbances are constantly complained of; they take the form of anorexia with discomfort from flat- ulency, sometimes offensive eructations, heartburn, nausea, distaste for fats and for meat. Frequent, bulky motions, pale in colour, offensive, and obviously greasy, are usually present in advanced conditions, though in the earlier stages there maybe constipation associated with flatulency. Marked and often excessive wasting is often a prominent symptom. The urine will give a more or less well-marked pan- Chronic Pancreatitis 429 creatic reaction according to the extent and intensity of the lesion, and a quantitative chemical analysis of the fasces will show an excess of unabsorbed fat, of which the greater part is unsaponified "neutral fat," particu- larly if there are advanced interstitial changes in the gland. Although in any single case we may not have all the symptoms and signs, yet in no case ought we to fail to find evidence of digestive, metabolic, or physical signs if chronic inflammation of the pancreas be present. No single symptom can alone be relied upon as diagnos- tic of chronic pancreatitis, but on considering all the available evidence there is not usually much difficulty in forming an opinion. Special stress can be laid upon the progressive wasting, the usual presence of jaundice, the dyspeptic disturbances, the pancreatic reaction in the urine, and the results of the chemical examination of the faeces. Each case, however, has to be considered on its merits, and in making the diagnosis one has to bear in mind the difference in the symptoms produced by the various causes as well as by the variations in the anatomy of the ducts. In the differential diagnosis of chronic pancreatitis the most important conditions to consider are cancer of the head of the pancreas, cancer of the common bile-duct, cancer of the liver, gall-stones in the common duct, and chronic catarrh of the bile-ducts. In cancer of the head of the pancreas the onset is usually gradual and painless, and the disease usually occurs later in life, generally after forty years of age. It is preceded by general failure of health, and when jaun- dice supervenes it becomes absolute and unvarying. The gall-bladder is nearly always distended, and may attain a large size. It is not tender on manipulation. The liver enlarges from biliary stasis, but there are no nodules to be felt. In some rare cases a hard nodular tumour may 430 The Pancreas: Its Surgery and Pathology be felt on the inner side of the distended gall-bladder. The fasces are usually acid in reaction, and contain a large amount of undigested fat, only a comparatively small pro- portion of which consists of fatty acids. The pancreatic reaction in the urine is negative by the improved method in about 7 5 per cent, of cases, but in the remaining 2 5 per cent, a more or less marked reaction, probably due to the asso- ciated inflammatory changes, is obtained. Preparations made by the original A-reaction show coarse crystals, soluble in 33 per cent, sulphuric acid in three to five min- utes, in many cases, but a typical reaction is not easily obtained and it may be necessary to make several prepara- tions from more than one specimen of urine before they are secured. The extremely rapid loss of weight and strength with increasing ansemia, but without ague -like seizures, is very characteristic, and it is common for there to be an absence of fever, or indeed a subnormal tempera- ture, with a slow feeble pulse, and later ascites with oedema of the lower limbs. The great importance of an accurate diagnosis between cancer of the head of the pancreas and chronic pancreatitis lies in the fact that while the latter is eminently a curable disease when submitted to early operation, the former is not benefited by surgical treat- ment, which, moreover, is attended by no little danger from various complications. Cancer of the comimon duct is rare and is usually asso- ciated with gall-stones. If the disease involves the pap- illa, the symptoms are indistinguishable from those of cancer of the head of the pancreas, except that the urinary pancreatic reaction is more likely to be of the inflamma- torv type from the associated changes in the gland due to the damming back of its secretion. If the growth is situated above the opening of the pancreatic duct, it will not interfere with the functions of the pancreas ; the loss of flesh will not be so rapid, the typical pancreatic reac- tion in the urine will be absent, and although there may be Chronic Pancreatitis 431 an excess of fat in the fseces, this will consist chiefly of combined fatty acids. Cancer of the liver is distinguished by the irregular enlargement and nodular feel of the organ, the rapid de- terioration of health, the less intense jaundice, and the absence of fever and paroxysmal pain. The pancreatic reaction is negative. A diagnosis of gall-stones may be made by the sequence of a long antecedent history of spasms without jaundice, then a severe attack of pain followed by jaundice, and after a time recurrent pains with increase of icterus asso- ciated with ague-like seizures. The absence of tumour is more common in gall-stones than in chronic pancreatitis, though in the latter the gall-bladder may be found con- tracted at times. The paroxysmal attacks in chronic pancreatitis may be equally as severe as those in gall-stone seizures, but there is usually less pain. The tenderness, however, with gall-stones will be over the gall-bladder, and in pancreatitis at the middle line where the swollen gland can sometimes be felt, especially if the patient is thin or under the influence of an anaesthetic ; moreover, the radiating pain in gall-stones is towards the right infra- scapular region, and in pancreatitis towards the left or to the mid-scapular region. When the gall-stones are situated in the first or second part of the common duct, the pancreatic reaction is negative and the faeces alkaline in reaction. The motions, although often containing a con- siderable excess of fat, do not show the high proportion of neutral fat usually found in pancreatic cases, but are, as a rule, rich in combined fatty acids. When, however, a stone is impacted in the third part of the duct, there is a proba- bility that, in the majority of cases, the pancreas will be inflamed; and when the calculus lies in the fourth part, the pancreas is almost certain to be affected. The diag- nosis of chronic pancreatitis from gall-stones is, however, not one of any great practical importance, since the two 432 The Pancreas: Its Surgery and Pathology conditions are often associated, and the treatment is, at least up to a certain point, the same. Chronic catarrh of the bile-ducts is characterised by jaundice and loss of flesh, coming on for the most part painlessly, but since it is usually of pancreatic origin, it is not necessary to spend time in discussing it further. Although the diseases mentioned are the most likely to cause confusion in diagnosis, they are not by any means the only ones for which chronic pancreatitis may be mis- taken. More than one patient has been sent to us who, from his colour and recurrent rigors, had been believed to be suffering from ague. The absence of malaria organ- isms in the blood, the presence of the pancreatic reaction in the urine, a chemical examination of the faeces, and a careful consideration of all the physical signs and symp- toms, have quickly revealed the true condition of things, which has been confirmed by operation and the subsequent course of the case. The blood changes met with in chronic pancreatitis suggesting pernicious anaemia have, in some instances, led to an incorrect view of the case being taken, until the possibility of their being of pancreatic origin was pointed out. Analysis of the urine and faeces by one of us in several cases diagnosed by various authorities as "hill diarrhoea, psilosis or sprue" has given results that have pointed to the pancreas being involved in the disease, and in one case that was operated on and a cholecystenterostomy performed the condition of the patient was much inproved . The faeces in this case before the operation contained 62 per cent, of total fat, of which 41 per cent, was neutral fat and 21 per cent, combined fatty acid. The urine gave a well-marked and characteristic "pancreatic reaction." After the operation the patient put on flesh, his appetite improved, and at the end of a month the fasces were found to contain 45 per cent, of total fat, 29 per cent, of neutral Chronic Pancreatitis 433 fat, and the same amount of combined fatty acid as before, namely, 21 per cent. There was thus not only a dimin- ished amount of fat in the stools, but there was also a much lower percentage of this in an undigested form. His general health had continued good nine months after the operation, and he then stated that, save for an occa- sional relapse which generally results from some error in diet, he had been free from pain and the faeces had been more nearly normal than for several years previously. A specimen of faeces examined during one of the relapses showed 60 per cent, of fat, of which 42 per cent, was neu- tral fat and 18 per cent, combined fatty acid. The pan- creatic element in at least some cases having the symp- toms of sprue has not, we believe, been previously insisted upon, but it is, we think, a point that should be borne in mind in the diagnosis and treatment of the disease. The prognosis of chronic interstitial pancreatitis surgi- cally treated is very favourable, but the longer the disease is left untreated, the more serious the outlook becomes. In some cases it may slowly progress for months or even years, but ultimately the well-marked cases die, either from asthenia or more rarely from haemorrhage or dia- betes. A marked haemorrhagic tendency usually shows the near approach of a fatal termination, and when dia- betes has supervened, the disease is, as a rule, so far ad- vanced that surgical interference is not likely to do more than possibly delay its progress, though a moderate degree of glycosuria need not be a bar to operation, as this may arrest the progress of the disease. Treatment. — Before considering either the medical or surgical treatment of pancreatitis, the importance of pre- ventive treatment by attention to the causes, some of which, such as gall-stones, are removable by operation in the very early stages with a very small risk, certainly not more than i per cent, in skilful. hands, must be in- sisted upon. Duodenal catarrh as a cause of pancreatic 28 434 The Pancreas: Its Surgery and Pathology catarrh and of interstitial pancreatitis is remediable by medical treatment; and duodenal ulcer, another cause, if not remedied by careful and thorough general treat- ment, can be cured by gastro-enterostomy with a very small risk. If, after a fair trial of general treatment, care in diet, wet packs to the epigastrium, rest, and mild mercurial purges, not too long continued, the symptoms persist, and the signs of failure in pancreatic digestion and metab- olism are manifesting themselves, the question of sur- gical treatment should be seriously considered, especially when the disease is associated with jaundice, for the condition is one that, if not relieved early, will certainly lead to serious degeneration of both the liver and pan- creas, and become dangerous to life in several ways. Rational treatment should aim at the cause, whether that be gall-stones, pancreatic calculi, duodenal catarrh, duodenal or gastric ulcer, alcoholism, or syphilis. In operating for chronic pancreatitis when medical treatment has failed to relieve, the surgeon must be pre- pared to do a thorough operation, so as to expose the whole length of the common bile-duct as well as the head of the pancreas. He will then be able to remove the cause should it prove to be a gall-stone, or a pancreatic calculus, or any other removable condition. In the absence of some obvious removable cause, it is advisable to secure efficient drainage of the infected bile-duct and pancreatic duct, either by cholecystotomy or cholecystenterostomy, preferably the latter. Where the pancreatic disease is dependent on duodenal catarrh associated with ulcer of the duodenum, it may be advisable, at the same time that the bile-passages are drained, to perform also a gastro- enterostomy in order to cure the original cause of the disease. Experience has taught that if the cause can be removed at an early stage, an absolute cure is possible ; and though complete restoration of the damaged gland Chronic Pancreatitis 435 in more advanced cases cannot always be promised, yet an arrest of the morbid process may be looked for, and the remaining portion of the pancreas will be able to carry on the metabolic, and, even if incompletely, the digestive functions of the gland. Surgical Treatment. — In several of our earlier cases of chronic pancreatitis the abdomen was opened and the biliary ducts and swollen head of the pancreas were exposed and manipulated without finding gall-stones. Whether it was that the manipulation of the parts dis- placed and pushed on a stone from the common duct into the duodenum, or that the breaking down of adhesions relieved tension, the fact remains that in a number of such cases the patients completely recovered and re- mained well. This was the course of events in the follow- ing case : The patient, a woman, aged forty-four years, had for some time been suffering from deep jaundice with con- siderable pain, some irregular fever, digestive disturbance, and emaciation. At the operation on June 22, 1890, after separating a number of adhesions, a tumour of the pancreas embracing the lower end of the common duct was found, which at the time was thought to be malignant. The tumour was freely manipulated in order to ascertain if any gall-stones were present in the common duct, but none could be felt, and as the gall-bladder was contracted and there was some tendency to heemorrhage from numer- ous small points, the investigation was not carried further and the abdomen was closed. Whether any concretion was pressed onward into the duodenum it is impossible to say, but the patient made a good recovery from the operation and within a few months she had regained her health. A letter received from her medical man states that she is now, fourteen years afterwards, in very good health. But in other cases of this kind, although the patients have recovered from operation and apparently become restored to health, an examination of the urine years 436 The Pancreas: Its Surgery and Pathology later has shewn the presence of the pancreatic reaction, and in some cases there has been glycosuria due to a seriously damaged pancreas. The following is a case in point : Mr. D , aged forty-two, had an attack of pain in the right hypochondrium ten years ago, but no jaundice. He had been free from attacks up to six weeks before seeing one of us, when he had a severe attack of pain in. the right hypochondrium, radiating to the back and shoulders, accompanied by rigors and vomiting and fol- lowed, by jaundice. The jaundice had persisted up to the time of his being seen, but then no swelling could be felt. An exploratory operation was performed on Octo- ber 27, 1898, when a mass, thought to be growth in the head of the pancreas, was discovered. The patient made a good recovery with gradual relief to the jaundice. The enlargement of the head of the pancreas was doubtless chronic pancreatitis, as it was too soft for scirrhus. It was very freely manipulated in order to ascertain if there was a gall-stone in the termination of the common bile- duct, and this may possibly, though if so, unconsciously, have dislodged an obstruction, leading to relief of the jaundice. A specimen of his urine was obtained in 1904, and although he was reported to be quite well, this was found to give crystals by the "A" reaction, which dis- solved in 33 per cent, sulphuric acid solution in half a minute, and to contain sugar in fair quantity. This, along with other cases, shews that it is unwise not to thoroughly drain the bile-ducts in all such doubtful cases, and if cholecystenterostomy is not performed, drainage ought to be continued until the bile becomes free from organisms and its normal route is free from obstruction. In certain cases, doubtless, recovery occurs without operation, and we have notes of one case where a gentle- man of advanced age had deep jaundice associated with glycosuria and with well-marked pancreatic reaction in the urine, pointing to the case being one of pancreatic Chronic Pancreatitis 437 diabetes. Under general treatment, combined with massage, he regained his health, and is now said to be quite well. In this case it is quite possible that the mas- sage may have dislodged a concretion which was blocking the common bile-duct and the pancreatic duct, but as no search was made in the faeces, this cannot be proved. As the patient lives abroad, we have not been able to test the urine, which will probably still contain glucose. This case raises the question whether operations ought to be declined because of the presence of a small amount of sugar in the urine. In future, should the patient's condition be fair, one would feel inclined to recommend operation in order to remove the obstruction, and, by drainage, to arrest the pathological process going on in the pancreas. It was only after the complete and perfect recovery of a case of interstitial pancreatitis in 1891, after the performance of cholecystotomy by one of us, that the indication for drainage of the bile-passages in inflamma- tory swelling of the head of the pancreas was made mani- fest. The following is a report of the case : The patient, a man, aged fifty, was deeply jaundiced and supposed to be suffering from a gall-stone in the com- mon duct, but on exploration on February 17, 1891, no biliary concretion could be felt, though a swelling of the head of the pancreas was found. The patient recovered after a simple cholecystotomy and regained his health, but we have not been able to trace his subsequent history. In 1892 came the first opportunity of actually proving the true pathological condition by a microscopic exami- nation of the head of the pancreas. The following is a description of the case : The patient was a man, aged thirty-two, w^ho was seen in April, 1892. He was extremely ill and emaciated at the time, and sufl:ering from deep jaundice and great pros- tration with dilatation of the gall-bladder. Operation 438 The Pancreas: Its Surgery and Pathology was undertaken too late and death resulted from shock and exhaustion on the second day. A necropsy revealed a cirrhotic condition of the head of the pancreas compress- ing the common bile-duct, there being no evidence of malignant disease (Fig. 137). Up to this time all the cases were examples of chronic pancreatitis, either independent of cholelithiasis, or in which no gall-stones were present at the time of operation, though one suspected that gall-stones had been the origi- nal cause of the trouble in both the first and second cases. It was not until 1895 that one of us actually found the associated condition of gall-stones in the common duct with chronic interstitial pancreatitis , and after removing the calculi from the common duct, short - circuited the gall-bladder into the duodenum, by the operation of cholecystenterostomy, after which the patient recovered and was in excellent health three years later. The following is a brief descrip- tion of the case : The patient, a woman aged fifty-one, had chronic jaundice and irregular fever, associated with spasmodic pains and great loss of flesh. Operation was performed on July 15, 1895, when gall-stones were removed from the common duct and a hard swelling of the pancreas was felt which was thought to be cancer. The gall-bladder Fig. 137. — Advanced chronic inter- stitial pancreatitis of the interlobular type (X ca 35). Chronic Pancreatitis ^ 439 was connected to the duodenum to establish permanent drainage. The patient recovered and was in excellent health three years later. This was followed shortly ?jy other cases of interstitial pancreatitis, some associated with gall-stones, while in others there was no evidence of cholelithiasis, but all of which were treated by cholecystenterostomy, which has since been repeated on numerous occasions ; at times the short-circuiting operation alone being done, at others the operation being associated with choledochotomy or duodenocholedochotomy. The following cases will serve as examples : A man aged thirty-four was seen in 1897. He had had painful attacks resembling cholelithiasis since June of 1896, and deep jaundice since December. The gall- bladder was distended and easily palpated. The patient was extremely ill and emaciated. At the operation on February 25, 1897, the gall-bladder was found to be dilated and surrounded by numerous adhesions, but no gall- stones could be felt. Cholecystenterostomy was per- formed. The patient made a good recovery and rapidly lost his jaundice. His medical attendant was good enough to write on January 24, 1904, to say that the patient was in good health and had never had a day's illness since his return home, the operation having taken place seven years previously. A woman was seen who had had gall-stones removed from the gall-bladder three years before, in Canada. She had never been free from jaundice since the opera- tion, and was subject to frequent vomiting. She was much emaciated, the stomach w^as dilated, and there was marked tenderness over the gall-bladder region. At the^pperation, on October 18, 1897, ver}^ extensive adhe- sions were found and the pancreas was much enlarged. Cholecystenterostomy was performed, after which she gradually regained her health. Her medical man was kind enough to send word on January 24, 1904, that the operation had been a complete success and. that the patient was leading an active life and was well six years after operation. 440 The Pancreas: Its Surgery and Pathology A female patient was seen on October 20, 1899, who had been suffering for three years from attacks resembling those of gall-stones, each attack being followed by jaun- dice. During the past fourteen weeks the seizures had been more frequent and severe, and jaundice had never quite cleared away before another attack came on. She had lost flesh and strength considerably and had vomited from time to time between the attacks. Her digestion was much impaired and there was a want of appetite. She had had no rigors and had only slight fever at the time of each seizure. The urine contained abundant lithates and a slight trace of albumin, but no sugar. An examination of the abdomen showed no manifest enlarge- ment of the liver or gall-bladder, but some tenderness over the gall-bladder and at the epigastrium, where there was an indefinite sense of fulness. An operation was per- formed on the 23d, when, after detaching numerous adhesions, fifteen gall-stones were removed from the cystic and common ducts, but, as a large nodular mass was occupying the head of the pancreas and partly ob- structing the common duct, it was deemed advisable to perform cholecystenterostomy so as to make a perma- nent opening between the fundus of the gall-bladder and the duodenum. The tumour gave the impression that it was malignant . Recovery was , however , uninterrupted , the button was passed on the tenth day, the wound healed by first intention, and the patient immediately began to put on flesh. She returned home within the month and has since been perfectly well in every respect. A man aged twenty-five was seen on January 11, 1905, on account of deep jaundice with serious deterioration of health, accompanied by loss of weight and strength. He gave a history that he had been out big game shooting in Uganda and had had an attack of fever, from which he had made a good recovery, returning home in Sep- tember, 1904, in fairly good health. In October he had what he took to be a return of the fever, the attack being ushered in by a rigor, and followed, within a few days, by jaundice, which gradually deepened, but he had abso- lutely no pain in the abdomen or elsewhere. He lost his appetite forthwith, and speedily began to lose flesh, so that when he was seen his weight was less by a stone than Chronic Pancreatitis 441 on his return home. There had been neither sickness nor vomiting since November, and he had had no more rigors. His pulse had been very slow (from 40 to 50) and the temperature subnormal. These symptoms continued up to the time he was seen, when he was found to be suffering from deep, almost black, jaundice, and from anaemia. The liver reached well below the costal margin, almost to the umbilicus, but the spleen could not be felt. A little tenderness was elicited an inch above the umbili- cus and half an inch to the right of the middle line, where it was thought a slight fulness could be felt, but this was indefinite. No dila- tation of the stom- ach could be made out, and beyond the jaundice, with the pale motions and dark urine, no other physical signs could be elicited. The motions were bulky, but not frequent , and there was no tend- ency to diarrhoea. The tongue was somewhat coated. Chronic pancreatitis was suspected and the urine and faeces were examined, the following being the report: Urine — Reaction, acid. Specific gravity, 1.014. Albumin, nil; some nucleo-proteid. Dextrose, nil. Pen- tose, nil. Maltose, nil. Glycuronic acid reaction, nega- tive. Indican, marked reaction. Ferric chloride reac- tion, negative. Bile-pigment, much. Microscopically, a few bile-stained epithelial cells. " Pancreatic reaction" : "A," many fine crystals soluble in 33 per cent, sulphuric acid in one-half to three-quarters of a minute (Fig. 138) ; "B," some fine crystals soluble in 33 per cent, sulphuric Fig. 138. — Crystals from the urine ob- tained by the "A-reaction" in a case of chronic pancreatitis (X 190) described in the text. 442 The Pancreas: Its Surgery and Pathology acid in one minute. Fseces^ — Reaction, alkaline. Sterco- bilin, traces. Microscopically, crowds of fat globules, many fatty acid crystals, much vegetable tissue, some partly digested muscle fibre, epithelial cells, and granular debris. Total, fat, 56.8 per cent.; neutral fat, 29.5 per cent.; fatty acids, 27.3 per cent. These results indicated a pancreatic lesion of an inflam- matory nature. The character of the crystals obtained from the urine and the relations of the "A" and "B" reactions suggested that, while the condition was probably of some standing, there was at the time some active inflam- mation of the gland. The large amount of bile-pigment in the urine, and its almost complete absence from the f^ces, together with the high percentage of fat and the presence of muscle fibre in the latter, pointed to an obstruction of the common duct as the probable cause of the condition. The considerable reaction for indican given by the urine suggested that there was some catarrh of the upper part of the intestine. As medical treatment with rest in bed and care in diet had been thoroughly tried without any benefit, an opera- tion was advised.. This was performed at a nursing home on January 16, 1905. On opening the abdomen by a vertical incision through the centre of the right rectus a little ascitic fluid, deeply bile-stained, escaped. The liver was found to be enlarged nearly to the umbilicus. It was dark and mottled, and showed evidences of cirrho- sis, apparently due to biliary stagnation. The gall- bladder was thickened, but not greatly distended, though it had evidently been inflamed, as adhesions were found between it and the cystic and common ducts and the neighbouring viscera, stomach, duodenum, and colon. The foramen of WinsloAV was obliterated by adhesions. The head of the pancreas was much enlarged and widened in area, so that it extended some distance up by the side of the common duct, which it enveloped. A hard nodule could be felt in the head of the pancreas, which hardness could be traced into the wall of the duodenum, and which faded off into the body of the pancreas, the body and tail of the organ being apparently of almost normal consis- tency. Adjoining the portion of the pancreas, which was stony hard, could be seen a number of lobules of the Chronic Pancreatitis 443 pancreas, which were firmer than normal and very defi- nitely outlined, a condition seen in a number of cases of chronic interstitial pancreatitis previously operated on. The localised hardness raised the question as to whether there might be a growth of the papilla extending into the duodenum, or whether there might possibly be a pancre- atic calculus impacted in the duct. It was therefore felt desirable to thoroughly explore the pancreas, and to this end the visceral peritoneum was incised over the duodenum and stripped from the pancreas. An incision was made into the indurated area and a portion of the hardened mass was removed. Although very hard, it Fig. 139.^ — Section of a nodule of pancreatic tissue removed at operation, showing chronic interstitial pancreatitis: a. Low power (X ca 37); b, high power (X ca 190). did not cut like cartilage and had not the appearance of a malignant growth. Subsequent microscopical exami- nation showed well-marked interstitial pancreatitis (Fig. 139). No calculus could be found. The incision in the pancreas was then closed by several catgut sutures, and the peritoneum replaced. The duodenum was then opened in the centre of the descending portion, w^hen a hard lobule was felt on the inner side of the papilla, and this proved to be continuous with the stony hard lump in the head of the pancreas. It was suspected that this was an accessory pancreas in the wall of the duodenum. It was not ulcerated, and did not give the impression of being malignant. The common bile-duct was clearly 441 The Pancreas: Its Surgery and Pathology compressed by the hardened head of the pancreas, and in order to estabHsh drainage of the infected bile the only desirable course seemed to be that of performing a chole- cystenterostomy. The opening made for exploration into the duodenum was therefore united to one made in the fundus of the gall-bladder by means of a Murphy button, and the abdomen was closed without drainage. The operation was unaccompanied by haemorrhage, as chloride of calcium had been given in 20-grain doses thrice daily for three days before operation, and the drug was continued subsequently to operation in 30-grain doses in the nutrient injections for a few days. The after-progress was very satisfactory. The wound healed entirely by first intention, and the jaundice began to diminish visibly within two days of operation. His bowels were moved on the fourth day after a dose of calomel, the motions showing the presence of bile. His appetite rapidly returned, and, after the button had passed on the tenth day, he was allowed to take food freely, his appetite being very keen. At first the motions were bulky, fre- quent, and offensive, evidently due to the passage of undigested milk. Pankreon tablets were therefore given after each meal, with the result that the food was better digested, and the motions were diminished in number. An examination of the blood was made on January 26th with the following result: Red corpuscles, 3,472,000 per cubic millimetre. Haemoglobin, 58 per cent. Haemoglo- bin index, 0.58. White corpuscles, 9,965 per cubic milli- metre. Proportion of red to white corpuscles, 348 to i. Differential leucocyte count: Polymorphonuclear white cells, 71 per cent.; small lymphocytes, 22 per cent.; large lymphocytes, 4 per cent. ; eosinophile leucocytes, 3 per .cent. ; mast cells, i per cent. At the end of the third week the patient was able to sit up, and when he was weighed at the month-end he had gained a stone in weight, and expressed himself as feeling well. His blood, urine, and fasces were examined again on February 20th, with the following result: Blood. — Red corpuscles, 4,634,000 per cubic millimetre. Haemoglobin, 92 per cent. Haemoglobin index, i.o. White corpuscles, 5,855 per cubic millimetre. Propor- tion of red to white corpuscles, 791 to i. Differential Chronic Pancreatitis 445 leucocyte count: Polymorphonuclear white cells, 70 per cent.; small lymphocytes, 22 per cent.; large lympho- cytes, 3 percent. ; eosinophile leucocytes, 4 per cent. ; mast cells, 0.5 per cent. Urine. — Reaction, acid. Specific gravity, 1.022. Al- bumin, nil. Dextrose, nil. Pentose, nil. Maltose, nil. Glycuronic acid reaction, negative. Indican, trace. Fer- ric chloride reaction, negative. Bile-pigment, faint traces. Microscopically, urates. Pancreatic reaction : "A," a few fine crystals soluble in 33 per cent, sulphuric acid in three-quarters to one minute; "B," a few fine crystals soluble in 33 per cent, sulphuric acid in one minute. This specimen of urine shows a marked improvement on that examined on January 12th. The pancreatic reaction was only slight, and the amount obtained by the "A" and "B" methods equal, indicating probably the fibrotic changes following the pancreatitis previously found. There was still a faint trace of bile-pigment in the urine, but it was exceedingly slight compared with the large amount present on the last examination. The indican reaction was very much diminished, and was not much more than is at times found in health. FtFC^.s-.— Reaction, alkaline. Stercobilin, a considera- ble amount. Microscopically, vegetable tissue, granular debris, no fatty globules or fatty acid crystals, no muscle fibre. Total fat, 15 per cent.; fatty acids, 11 per cent.; neutral fat, 4 per cent. On February 20th, when the patient left the nursing home, his weight had increased to 10 stones 9 pounds, it having been 9 stones on admission. It will be seen that whereas before operation the total fat in the fasces was 56.8 per cent., on February 20th it had diminished to 15 per cent., the fatty acids, which were 27.3 per cent. January 12th, had diminished to 11 per cent, on Feb- ruary 20th; and the neutral fat, which was 29.5 per cent, on January 12th, had diminished to 4 per cent, on Feb- ruary 20th. The blood had also very materially improved, as will be seen on comparing the reports, for as on Jan- uary 26th the red corpuscles were 3,427,000 per cubic millimetre, on February 20th they were 4,634,000 per cubic millimetre; the h^emiOglobin, which on January 26th was 58 per cent., on February 20th was 92 per cent. ; 446 The Pancreas: Its Surgery and Pathology the hsemoglobin index on January 26th was 0.85, on February 20-th it was i.o ; the white corpuscles, which on January 26th numbered 9965 per cubic millimetre, on February 26th had diminished to 5855 ; and the propor- tion of red to white corpuscles, which on January 26th was as 348 to i, on February 20th wa's in the proportion of 791 to I. A simple drainage of the gall-bladder by cholecystotomy is frequently unsatisfactory, and cannot be relied on in well-marked cases of obstruction, as the drainage of the bile-passages is not sufficiently long continued. This applies especially to the cases in which the interstitial pancreatitis has persisted for some length of time, in which cases, although a cholecystotomy may lead to a disappearance of the jaundice and the digestive symp- toms may be alleviated, the metabolic signs found in the urine many months or even years subsequently show that recovery has only been partial. The following are examples: Mr: D , aged forty-five, had had painful epigastric attacks for twelve months, with vomiting, but no jaun- dice. There had been deep jaundice since January last, with ague-like attacks, and the patient had lost 2^ stones in weight. Cholecystotomy was performed on March 29, 1898. Thickened duct felt, together with a swelling of the pancreas, thought to be cancer of the head of the pancreas and the common bile-duct. Drainage of the gall-bladder for ten days. The patient made a complete recovery, and in August was apparently quite well, having gained a stone in weight. He was in good health in 1 90 1. Though apparently well in January, 1904, an examination of the urine gave the pancreatic reaction and showed that the original damage to the pancreas had not been completely repaired. Mrs. D , aged forty-six, had had spasms for years. Acute seizure in July and three times since. Since July, pain and sickness every two weeks. No tumour felt at any time ; jaundiced occasionally after an attack of pain ; Chronic Pancreatitis 447 lost one stone in weight. She had never vomited blood and never had melaina. There was tenderness over the gall-bladder, but no tumour. Slight enlargement of the head of the pancreas. Cholecystotomy was perfonned on December 11, 1899. Empyema of the gall-bladder. Many stones removed from the gall-bladder and cystic duct. Adhesions broken down. Xodular condition of the head of the pancreas found. The patient made a good recovery and was well in 1904, though an examina- tion of the urine showed the A and B jjancreatic reaction and proved that the metabolic functions of the pancreas were still not normal. In some cases where operation has been delayed, or drainage of the bile-ducts not performed or not long enough continued, the original interstitial pancreatitis may advance so that the islands of Langerhans become in- volved, and glycosuria ensues, as in the two following cases : Mrs. C , aged fifty-one, who was suffering from persistent jaundice, with periodical pains, and ague-like seizures that had extended over a long period, was oper- ated on in July, 1895, when several gall-stones were removed and others crushed in the common duct. A tumour of the pancreas was felt, which it was thought at the time might be malignant. The gall-bladder was therefore drained into the duodenum by a cholecysten- terostomy. The patient completely recovered, and has remained well since the operation, over nine years ago, but an examination of the urine showed there to be an abundance of dextrose, but no acetone or diacetic acid. Pancreatic crystals were obtained by the '"A" reaction, which dissolved in three-quarters to one minute, but none could be isolated by the "B" method. This showed that although the patient has been relieved by the opera- tion, and had apparently enjoyed good health, yet that she was living with a damaged pancreas and consequent glycosuria. A man, aged forty-five, was seen on the 25th of October, 1898. The patient was very deeply jaundiced, and said 448 The Pancreas: Its Surgery and Pathology that he had lost a stone in. weight since the onset of his illness, five weeks before. He gave the history of having had attacks of pain, referred to the region of the gall- bladder, nine years previously, but they were unaccom- panied by jaundice and passed off after prolonged treat- ment. From that time onwards he had been free from attacks of pain up to the onset of the present illness, five weeks before, when he was suddenly seized with severe pain at the pit of the stomach and became jaundiced. The pain had recurred daily and had been so severe as to necessitate his taking morphia. His medical attendant noticed a swelling in the region of the gall-bladder a fort- night after the onset of his illness, and there was all along well-marked tenderness at the epigastrium, with gradually increasing enlargement of the liver. The patient's gen- eral health rapidly failed, and the loss of flesh was well marked. When seen he looked pinched and ill, he was very deeply jaundiced, and the urine was loaded with lithates, but contained neither albumin nor sugar. There were well-marked tenderness at the epigastrium and a smooth tumour, which was not very tender, in the gall-bladder region; the liver was enlarged and the edge was smooth and could easily be felt an inch below the costal margin. A diagnosis of gall-stones in the com- mon duct was made and the patient was admitted into hospital. The operation was performed on September 27, 1898. On opening the abdomen numerous adhesions between the gall-bladder and liver, and the pylorus, colon, omentum, and duodenum were found. The gall-bladder was slightly distended, but no gall-stones were felt either in it or in the cystic or common duct. There was, how- ever, a hard nodular swelling of the head of the pancreas, which, at the time, was thought to be malignant. In order to give relief, the adhesions were detached and the gall-bladder was drained by cholecystotomy. On Octo- ber 28th a letter was sent to his medical man, telling him that we feared the disease of the pancreas might be malig- nant, but that there was a possibility of its being a chronic pancreatitis. On November 5th another letter was writ- ten by one of us, to this effect: " I am pleased to be able to tell you that your patient has improved very much and the jaundice has nearly disappeared. I hope, there- Chronic Pancreatitis 449 fore, that the tumour of the head of the pancreas may have been inflammatory, and not malignant. At the time of operation it occurred to me that it was not quite hard enough for a malignant tumour, but under the cir- cumstances I felt it my duty to give you my suspicions." From that time onwards recovery, was uninterrupted, and the patient left the hospital with the wound closed, within the month. In December, 1899, the patient called to report himself. He looked perfectly healthy and had gained over a stone in weight since his return home. He had neither pain nor tenderness and he said he felt as well as if he had never ailed anything. The scar was firm, the liver was normal, and there was not the slightest ten- derness in the epigastrium or in the gall-bladder region. Five years later the patient was apparently well so far as the local symptoms were concerned, but an examination of the urine showed the presence of glucose and the pan- creatic reaction was present, thus pointing to the persis- tence of chronic interstitial pancreatitis, which had evidently extended and invaded the islands of Langerhans. The urine contained no bile and no albumin, but an abun- dance of oxalates. Had the gall-bladder been longer drained in this case or cholecystenterostomy performed, it seems highly prob- able, arguing from the results of operation in other cases, that the sequelae above mentioned might have been pre- vented. Occasionally, however, the simple operation of chole- cystotomy may be sufficient to bring about a cure, as in the following cases : The patient, a man aged forty-five, was seen on March 19, 1898, the history being that he had been well up to twelve months before, when he began to have painful attacks at the pit of the stomach, ending in vomiting, but not followed by jaundice until an attack on January i, 1898, since which time he had been deeply and continu- ously jaundiced. He had also from that time onwards had ague-like attacks, and two days before he was seen he had had within twenty-four hours three of these 29 450 The Pancreas: Its Surgery and Pathology seizures, each accompanied by pain. Within a twelve- month he had lost 2 stones 8 pounds in weight. On examining him there was some swelling in the gall-bladder region but no tenderness. The liver was a little enlarged but the margins felt smooth. There was decided tender- ness in the middle line just above the umbilicus, and on deep pressure the pain was considerable and an indefinite fulness could be felt. The diagnosis of gall-stones in the common duct was made, and an operation was advised. The patient was operated on on March 30th, when the gall-bladder was found to be slightly distended and sur- rounded by adhesions to the pylorus, duodenum, colon, and omentum. No gall-stones could be discovered, but there was a well-marked swelling of the head and the first two inches of the pancreas, which, though nodular and irregular, was not very hard. This extended further to the right than normal, so as to cover in the lower end of the common bile-duct. Cholecystotomy was per- formed. Within twenty-four hours of the operation nearly four pints of very offensive bile were discharged through the tube. A specimen was examined by the Clinical Research Association and their report was as follows: "The bile contains both staphylococci and streptococci, but no bacillus coli communis could be found either under the microscope or in the culture." Fearing that the disease might be malignant, and the patient being so extremely weak and ill, a poor prognosis was given, but in a few days the following report was given: "The patient is progressing satisfactorily, though he is still pro- foundly weak. Bile has appeared in the motions so that the obstruction is evidently overcome. The bowels have been moved naturally and the patient is less deeply jaundiced and looking better generally." On April 5th he was taking his food well and bile was passing freely in the motions. He had had no recurrence of the shiver- ing attacks. Drainage was continued for fourteen days and the patient returned home on the 20th. The urine was then free from bile and the motions were assuming a natural colour; he was taking food well, gaining flesh, and looking better generally. A guarded prognosis was still given, however, as it was thought that the tumour would prove to be inflammatory and not malignant. Chronic Pancreatitis 451 From that time onwards his progress to recovery was extremely rapid, and he was said to be perfectly well in every respect a few months later, and had fully regained his lost weight. Two years later he was still in perfectly good health. A woman, aged thirty-five, was seen on September 11, 1899, with the history of having been subject to attacks of spasms in the upper abdominal region for twelve years, the intervals between the seizures having varied from a few days to several months, but of late they had become much more frequent, and during the week before she was seen she had had four attacks, all severe ones. The seizures began with pain in the epigastrium accom- panied by cold sweats and faintness; the pain passed through the midscapular and to the right subscapular region, and lasted from two to six hours, having to be relieved at times by morphia. Jaundice followed the seizures, and if the attacks recurred frequently it was in- tensified with each, but if there was a long interval only an icteric tinge remained. Palpation revealed a point of tenderness in the mid-line, one-and-a-half inches above the umbilicus, where there was a sense of resistance with an abnormal fulness, but there was no tenderness over the gall-bladder, nor could any swelling of the gall-blad- der or liver be discovered. On September 21st a vertical incision through the right rectus exposed adherent viscera, and, on the separation of the adhesions, a thickened gall-bladder was exposed, but there were no gall-stones in it or in the ducts. The lower part of the common duct was surrounded and overlaid by a well-marked swelling of the pancreas, which was harder than' usual, but not sufficiently hard to be mistaken for cancer, though it was somewhat nodular. Cholecystotomy was performed and drainage was carried out for a fortnight. Recovery was uninterrupted and the patient returned home within the month, and she has remained well since. If the gall-stone causing obstruction be removed by operation from the common duct, and drainage of the infected bile-ducts be effected before the catarrhal has passed into the interstitial form of pancreatitis, a com- 452 The Pancreas: Its Surgery and Pathology plete cure may be expected even after simple drainage of the bile-ducts, as in the following cases: A lady, aged thirty-four, had had symptoms of gall- stones for four years, and had been under treatment for ulcer of the stomach, but there had been no hsematemesis. Four months previously jaundice had come on after an attack of pain, since which time the attacks had been frequent and were always followed by an increase of the jaundice, and by rigors and fever. On one occasion the gall-bladder was distended ; when seen there was a slight tinge of jaundice. She had lost 3 stones in weight. There was an absence of enlargement of the liver or gall-bladder, but marked tenderness over the gall-bladder was elicited . Pancreatic crystals were found in the urine and digestive symptoms were present. At the operation, on April 23, 1903, one large calculus was removed from the cystic duct, and some smaller ones from the common duct, by separate incisions in the two ducts. The common duct was sutured and the cystic duct drained. The pancreas was slightly swollen. The patient made a good recovery and remains well. The explanation of the pancreatitis in these two cases was manifestly the obstruction of the pancreatic duct and infection of the secretion ; but the complete recovery after operation showed that the inflammation was prob- ably only catarrhal and not advanced interstitial trouble. If the gall-stone obstructs the common duct for long, what was at first a simple catarrhal pancreatitis may assume a truly interstitial form, and unless drainage of the bile-ducts is continued for some time or permanent drainage in the shape of cholecystenterostomy is estab- lished, relapse will speedily occur. The following case is an example : Mrs. W — — , aged fifty-seven, had had two operations previously in Scotland. On the occasion of the first operation, in September, 1902, a number of gall-stones were removed from the gall-bladder, which was drained Chronic Pancreatitis 453 for a few days, but after the wound had healed the attacks had been repeated as before. A second operation was undertaken by the same surgeon without finding any- thing definite. After the wound had healed, and the temporary drainage had ceased, the attacks again re- turned, and the subsequent history up to the time of our seeing her was that she had almost daily attacks of pain, followed by slight jaundice, and on five or six occasions, usually at intervals of a month, she had had violent seiz- ures necessitating the use of morphia. About five weeks before being seen by us the pain was so violent as to cause her to faint, and just before coming to London another violent seizure, accompanied by collapse, occurred. A rigor with high temperature, 104° or 105°, had followed each attack, the temperature between the seizures ris- ing nightly to 101° or io2°F. She was rapidly losing flesh and strength. An examination of the urine showed no albumin or sugar, but well-marked pancreatic crystals by the A reaction, which dissolved in from one to one-and- a-half minutes, and a smaller number of similar crystals by the B method, rendering, along with other signs, the diagnosis of chronic pancreatitis certain. At the opera- tion on November 20, 1903, the adhesions w^ere found to be most extensive. There was well-marked enlargement and hardness of the pancreas along its whole length, but it was not nodular. The common duct was carefully examined, but found to be free from concretions, and on opening the gall-bladder a probe was passed through it and the cystic and common ducts into the duodenum. While the probe was in position, the pancreas was manip- ulated and found to compress the duct, thus accounting for the obstruction. Cholecystenterostomy was there- fore performed, the union being effected by means of a decalcified bone bobbin. At the time of operation the gall-bladder was separated from its fissure in the liver in order to make it reach the bowel without tension. For a few days after operation bile was discharged from the torn liver surface in free quantities, but there was no leakage from the newly joined viscera. As the bile obtained a free passage into the bowel, it gradually ceased being discharged from the liver, and the tube was able to be left out at the end of ten days. The wound healed 454 The Pancreas: Its Surgery and Pathology by first intention and the patient was up at the end of three weeks. She was then able to take and digest her food, and has since been quite free from her old attacks. After cholecystotomy, the patient may become impa- tient of the continued drainage and demand too speedy relief. This was well shown in the following case : A military officer, aged sixty, was seen on the 8th of July, 1904, He was in good health up to May 2d of that year, when painless jaundice developed. He had a feel- ing of discomfort after food, the jaundice deepened, and he rapidly lost flesh . When seen he was deeply j aundiced , the liver was enlarged, nearly to the umbilicus, and the gall-bladder was distended. He said he had no pain and there was no evidence of ascites or oedema of the legs. As he did not improve at all under general treatment, and an examination of the urine showed many oxalate crystals, and a well-marked pancreatic reaction, and the faeces were acid in reaction, and contained 58.7 per cent, of the dry weight as fat, of which 31.4 per cent, was neu- tral fat, and 27 per cent, fatty acid, a diagnosis of intersti- tial pancreatitis was made, and operation was performed on July 20, 1904, when the pancreas was found to be much enlarged and compressing the common bile-duct. There was no positive evidence of malignant disease, as although the glands were enlarged, they were discrete and not nodular. As there was a decided haemorrhagic tendency, and the patient was too ill to bear a prolonged operation, the enlarged gall-bladder was simply drained by cholecys- totomy. The patient made a good recovery and improved considerably in his general health, and the jaundice en- tirely disappeared. In consequence of our temporary absence on a holiday the patient got uneasy about the persistent discharge of bile and was advised to consult another surgeon, although he had been counselled to bear with the cholecystotomy for at least two months before having anything further done. Despite this advice, and without our knowledge, the gall-bladder was short-cir- cuited into the colon. After the operation he was very much distressed by diarrhoea, and after a time he began to suffer from symptoms of septicaemia with rigors, which Chronic Pancreatitis 455 ended in death from pycemia. At the autopsy the infec- tion of the bile-passages was found to have occurred through the communication with the colon, and the liver was riddled with abscesses. An examination of. the pancreas showed a simple interstitial pancreatitis. Drainage of the common or hepatic duct may have to be performed for jaundice due to interstitial pancreatitis where there is absence or contraction of the gall-bladder, either owing to the gall-bladder having been removed at a previous operation or to its having contracted as the result of gall-stone irritation. This is a much less satis- factory operation than cholecystenterostomy, as drainage of the common duct has to be continued for some length of time and the biliary fistula is a source of great distress to the patient. We have had to drain the common duct in several such cases where it was impossible to relieve the patient in any other way ; and in one case it necessitated the biliary fistula being continued for a considerable time. This brings into prominence the imdesirability of removing the gall-bladder as a routine practice in operat- ing for gall-stones, for unless it is seriously damaged or ulcerated, or is the seat of malignant disease, or unless there is ulceration or stricture of the cystic duct, removal is quite unnecessary, and we think it better practice to drain it simply and not to perform cholecystectomy, since on some future occasion, should trouble develop in the deeper ducts or in the pancreas and the gall-bladder be absent, it will be impossible, with few exceptions, to short- circuit the obstruction. Moreover, after cholecystotomy gall-stones have no greater tendency to re-form than they have after cholecystectomy, and should cholelithiasis again develop, it will be in the common duct, a much more serious position than if in the gall-bladder. The following case affords a good example of the advan- tage of sparing the gall-bladder in operating for gall- stones : 456 The Pancreas: Its Surgery and Pathology Mr. T , aged forty-five, was seen by one of us on July 27, 1905, suffering from jaundice and a biliary fistula. He gave the history that he had been operated on in October, 1904, by a hospital surgeon for suppurating gall-bladder, but that the wound had never healed and a biliary sinus had persisted. He consulted a well-known Continental surgeon, who advised operation. When seen by us he had a temperature of 100° F., and looked ill. He said that he was subject to shivering attacks. His tongue was coated and there was slight jaundice. A biliary fistula was present, which was discharging a small quantity of bile and pus. He had lost weight con- siderably. The urine showed a well-marked pancreatic reaction and the fasces contained a quantity of fat and muscle fibre. On July 28, 1905, the abdomen was opened and the gall-bladder was separated from the fistula, which was excised. The head of the pancreas was hard, and evi- dently the seat of interstitial pancreatitis, which com- pressed the common bile-duct. No gall-stone could be felt in the common or hepatic ducts. The gall-bladder was therefore connected to the duodenum so as to short- circuit the obstruction. The patient made a good recovery and forthwith began to regain his lost flesh, the jaundice disappeared, and his skin soon assumed a healthy colour. He called to report himself in January, 1907, and said that he was in perfect health. An opportunity was taken of examining his urine some time after the operation, when it was found to be normal and to show no traces of the pancreatic reaction. If the common duct is greatly dilated it may be possi- ble to make an anastomosis between it and the duodenum so as to short-circuit the obstruction in the head of the pancreas. The following case affords an example of choledoch- enterostomy in such a condition : Miss F , aged twenty-eight, seen with Dr. G- in June, 1903. She gave the history that four years' previously she had had typhoid fever, since which time she had never been well. A year previously she had an Chronic Pancreatitis 457 attack of pain followed by jaundice with some enlarge- ment of the gall-bladder. In January, 1902, she was operated on by Dr. G . No gall-stones were found, but the head of the pancreas was much enlarged. Chole- cystotomy was done and the wound healed within the month. She made a good recovery from the operation, and was apparently well until March, 1903, when she had a recurrence of the jaundice with sickness and pain. She became very ill and rapidly lost flesh. When we saw her together there was some enlargement of the gall- bladder and a distinct cystic swelling over the pancreas, and the urine gave the characteristic pancreatic reaction. On June 4, 1903, an operation was performed by one of us, when a large cyst was found on the inner side of the gall-bladder, containing bile and pus, which was evidently a dilated common bile-duct. No gall-stones were found, but there was some swelling of the head of the pancreas. The gall-bladder was also distended and inflamed, and it was drained by a separate tube. The patient made a good recovery from the operation, and returned home wearing both tubes. She was seen again in October, 1903. Since the former operation there had continued to drain away through the tube in the dilated common duct from 20 to 30 ounces of bile, and from the tube leading into the gall-bladder from 4 to 6 ounces of clear mucus. The patient was thin and feeble, had no appetite for food, and was unable to digest anything beyond a little milk. An examination of the urine revealed the characteristic pan- creatic reaction, and the faeces contained muscle fibre and much fat. On the 8th of October a further operation was undertaken, when the head of the pancreas was again found to be much enlarged, but no concretions could be felt in it or in the common bile-duct. The gall-bladder was completely excised and the cystic duct ligatured. The dilated common bile-duct was then connected to the duodenum by means of a decalcified bone bobbin and the wound was closed. The same evening the patient ex- pressed herself as feeling hungry for the first time since her illness began. She straightway began to absorb whatever nourishment was taken, had her bowels moved on the second day, gained strength, resumed her natural colour, and made such a rapid convalescence that she 458 The Pancreas: Its Surgery and Pathology returned home within the month, having gained 7 pounds in weight. In 1906 a report was received to say that the patient was in perfect health. Details of the Operation for Exploring the Head of the Pancreas and the Common Bile-duct. — Certain modifications of the operation for exploring the head of the pancreas and the common bile-duct have converted what was formerly a most difficult procedure, involving prolonged manipula- tion, special appliances, and at least two assistants, into a comparatively simple operation, in the greater number of cases, requiring the help of only one assist- ant and not calling for the use of any special appa- ratus. By this method, suggested and put in practice by one of us, the time involved in the operation is reduced considerably, and where adhesions do not give unusual trouble, it is easy to complete the work in from thirty to forty minutes, which not only means a saving of time and fatigue to the operator, but a considerable saving of shock to the patient. A firm sand-bag should be" placed under the back opposite to the liver, which not only pushes the spine, and with it the pancreas and common duct, forwards, but acts like the Trendelenburg position in pelvic sur- gery by letting the viscera fall away from the field of operation, or the same advantage may be obtained more readily and conveniently by employing an operating table specially designed so as to be able to effect the pro- jection of the liver region forwards. The one we regularly employ is the Guyose-Greville table. A vertical incision is then made over the middle of the right rectus, the fibres of which are separated by the finger, which is the most expeditious and the most effective method of exposing the gall-bladder and bile-ducts ; but when it is necessary to open either the common duct or the deeper part of the cystic duct, instead of prolonging the incision downwards, Chronic Pancreatitis 459 as was formerly done, it is better to carry it upwards in the interval between the ensiform cartilage and the right costal margin as high as possible, thus exposing the upper portion of the liver very freely. It will now be found that by lifting the lower border of the liver in bulk, so as to rotate it if needful, first drawing the organ downwards "from under cover of the ribs, the whole of the gall-bladder and the cystic and common ducts are brought close to the surface, and, as the gall-bladder is usually strong enough to bear traction, the assistant can take hold of it by fingers or forceps, and by gentle traction can keep the parts well exposed, at the same time that, by means of his left hand with a flat gauze sponge under it, he retracts the left side of the wound and the viscera, which would other- wise fall over the common duct and impede the view. It will now be observed that, instead of the gall-bladder and cystic duct making a considerable angle with the common duct, an almost straight passage is found from the opening in the gall-bladder to the entrance of the bile-duct into the duodenum, and if adhesions have been thoroughly separated, as they should always be, the sur- geon has immediately under his eye the whole length of the ducts with the head of the pancreas and the duode- num. So complete is the exposure that, if needful, the peritoneum can be incised, and the common duct can be separated from the structures in the free border of the lesser omentum, but this is not necessary except where a growth has to be excised. By incising the peritoneum passing from the duodenum to the pancreas, the duodenum can be lifted up and the posterior surface of the pancreas and the common bile-duct can be fully exposed. The surgeon, whose hands are both free, can with his left finger and thumb so manipulate the common duct as to render prominent any concretions, which can be cut down on directly, the edges of the opening in the duct being caught by pressure forceps. The assistant can now take hold 460 The Pancreas: Its Surgery and Pathology of the forceps with his left hand, as that instrument with the sponge will form a sufficient retractor, since the duct is so near the surface. When the duct is incised there is usually a free flow of bile, which, it must be remembered, is. infected, but a gauze swab in the kidney pouch and the rapid mopping up of bile as it flows, by means of sterilized gauze pads, avoid any soiling of the surrounding parts, and, if thought necessary, the bulk of the infected bile can be drawn off by the aspirator, either from the gall-bladder or from the common duct above the obstruction, before the incision into the duct is made. After removing all obvious concretions the fingers are passed behind the duodenum and along the course of the hepatic ducts to feel if other gall-stones are hidden there, and a gall-stone scoop, the only special instrument necessary, is passed up into the primary division of the hepatic duct in the liver and quite down to the duodenal orifice of the com- mon bile-duct, and to ensure the opening into the duode- num being patent, a long probe is passed into the bowel. The incision into the bile-duct is now closed by an ordi- nary curved round needle held in the fingers without any needle-holder, a continuous catgut suture being used for the margins of the duct proper, and a continuous fine green catgut, or spun celiuloid, thread being employed to close the peritoneal edges of the gut. In such cases, where the pancreas is indurated and swollen from chronic pancreatitis, and is likely to exert pressure on the com- mon duct for a time, a drainage-tube is inserted directly into the duct, and the opening closed around it by a purse - string suture, the tube being fixed into the opening by a catgut stitch, which will hold for about a week ; but where this is not done, a drainage-tube may be fixed into the fundus of the gall-bladder in the same way, as this drains away all infected bile and avoids pressure on the newly sutured opening in the duct; or, better still, the gall- Chronic Pancreatitis 461 bladder may be short-circuited into the duodenum by the operation known as cholecystenterostomy. vSo easy is it to remove impacted stones after this method of exposure that a long time need not be spent in manipulating stones impacted either in the cystic or common duct, but the duct can be incised at once, the concretions removed, and the opening closed without damaging the duct by prolonged manipulation. Although there is seldom any fear of leakage or of infection, yet owing to the separation of extensive adhesions, there is usually some tendency to pouring out of fluid in the first twenty-four hours. It is therefore generally advisable to insert a gauze drain through a split drainage-tube, bringing it out by the side of the gall-bladder drain, or, better still, both tubes may be brought out of a separate opening external and posterior to the operation wound, which can then be permanently closed. The wound is closed in the usual way by continuous catgut sutures, first to the peritoneum and deep rectus, next to the anterior rectus sheath, and lastly to the skin. Even in acute or subacute, as well as in chronic pancreatitis, this method is advantageous, as, at the same time that the pancreas is exposed, the bile-ducts can be explored, and if the cause be gall-stones, they can be removed. Should it be necessary to expose the under surface of the pancreas, an extension of the incision downwards gives enough room to raise the transverse colon and to get directly at the body of the pancreas through the trans- verse mesocolon. To those having little experience in this operation the modifications described may seem trivial, but to those who have experienced the difficulties of the ordinary operation, a method which enables the pancreas and the whole of the bile-passages to be dealt with close to the surface will be sufficiently appreciated. But the tech- nique of the operation is not the only important part of the 462 The Pancreas: Its Surgery and Pathology treatment of these serious cases, which require care and thought, not only before and at the time of, but subse- quently to operation. A careful study of the causes of mortality in operations on the common duct, associated with jaundice and pan- creatitis, shows that haemorrhage, either immediate, con- secutive, or secondary, cannot be ignored as a danger, and that shock, apart from haemorrhage, has next to claim our attention. Sepsis is no longer the bugbear that it used to be, thanks to a rigid all-round asepsis, the employment of gauze drainage, and the careful avoidance of soiling the wound by infected bile. Although there is a greater tendency to bleeding in chronic jaundice from pancreatic disease than when jaundice is due to gall-stone obstruction, there can be no doubt that in all cholasmic conditions the blood becomes so altered that the coagulability is seriously diminished, and that these features demand serious attention before any operation is undertaken in cases of common-duct cholelithiasis. By administering chloride of calcium in the case of jaundiced patients, both before operation, in 30-grain doses by the mouth, and afterwards in 60 -grain doses by the rectum daily for several days, the haemorrhagic tendency can be successfully combated. It is important to ligature all bleeding points and not to trust simply to forci pressure ; and, while in non-jaun- diced patients adhesions may be simply separated, in these cases it is preferable to divide adhesions between ligatures where practicable. Where there is persistent oozing of blood from innumerable points, a tampon of sterilized gauze forms a useful means of haemostasis, and this may be made more efficient by employing at the same time a solution of suprarenal extract to the bleeding surface. The best treatment of shock is preventive, and to that end it is desirable to lose as little blood as possible, though Chronic Pancreatitis 463 shock in operation is not always dependent on loss of blood. The patient is enveloped in a roughly made suit of gamgee tissue, and where he is very feeble, or the operation is likely to be prolonged, it is performed on a heated table. A large enema of normal saline solution with or without stimulant, given from fifteen to twenty minutes before, and the administration of five minims of solution of strychnia, subcutaneously just after the operation, are useful. Expedition in operating is an im- portant factor in lessening shock, especially in abdominal surgery, for it stands to reason that prolonged manipu- lation and exposure of the viscera in patients so ill as those composing the class of cases which we are now con- sidering must generally be, will be badly borne, for it is not only the work of the surgeon but the deep anaes- thesia that adds to the shock, since for the operation to be well and expeditiously performed the muscles must be thoroughly relaxed. After the operation a pint of saline fluid, with one ounce of brandy, is given by enema, and five minims of strychnia are given subcutaneously in two hours and repeated if desirable. The rectal injection is repeated in two hours, and afterwards every four hours with an ounce of liquid peptonoids added. Subcutaneous in- jections of saline fluid or intravenous infusion are only rarely required. Cholecystenterostomy. — The operation of cholecysten- terostomy consists in establishing an artificial opening between the gall-bladder and duodenum, jejunum, or colon, preferably the duodenum, at the part lying nor- mally close to the gall-bladder. Although the conception of the operation occurred independently to Harley, Gaston, and Nussbatim, the first operation was actually performed by Winniwarter, of Liege, in 1880, and a case operated by one of us in 464 The Pancreas: Its Surgery and Pathology 1889 was the first cholecystenterostomy performed in Great Britain. Since 1889 we have performed the operation forty-eight times, and for the following conditions : 1. Interstitial pancreatitis compressing the common bile-duct. 2. Biliary fistula, due to stricture of the common bile- duct, or to compression of it by a swollen and inflamed pancreas. 3. Cancer of the head of the pancreas, where relief of the urgent symptoms appeared to be desirable. A recent statement, to the efl:ect that cholecystenter- ostomy is a very serious operation with a heavy mortality, is clearly incorrect when performed for non-malignant conditions, if the operation is properly carried out and with all necessary precautions. The following statistics shew the results in our practice : The operation has been performed in thirty-nine cases for chronic interstitial pancreatitis, with two deaths, and in ten for cancer of the pancreas, with seven deaths. The cause of death in the two fatal cases of chronic pan- creatitis was in no way connected with the operation, for in one death occurred from acute nephritis, with suppression of urine and ursemic convulsions, without apparent cause, after the patient was apparently well and the wound had been soundly healed for a week, and in the other there was, in addition to the pancreatitis, suppurative cholangitis and abscess of the liver. The very high mortality in the cancer cases clearly proves that operative interference is highly undesirable and that every means should be taken to diagnose the condition from simple inflammation, in which the results following operation are nearly always most satisfactory. It is therefore an extremely useful operation in suitable cases, such as obstruction of the common bile-duct from interstitial pancreatitis, and in biliary fistula dependent Chronic Pancreatitis 465 on stricture of the common bile-duct, but only rarely is it justifiable to perform the operation in cancer of the head of the pancreas, as, at the best, life in such cases is not considerably prolonged by any operation. It can rarely be justifiable or wise to perform it in gall-stone obstruc- tion, as the modern operation of choledochotomy, which removes the obstruction, can be performed in as short a time and is curative, whereas any short-circuiting opera- tion performed for gall-stones leaves the irritating foreign bodies, which may lead to other complications. The anastomosis ought unquestionably to be made between the duodenimi and gall-bladder, as in that way the secretions mix with the food in the normal position and, the duodenum being part of the intestinal canal less frequented by organisms than the bowel. lower down, there is practically little or no danger of infection of the bile-passages. As a matter of fact, we have never seen infection to occur in any case of duodeno-gall-bladder anastomosis, but we know of cases in which an anasto- mosis between the gall-bladder and colon has been fol- lowed after a time by multiple abscesses in the liver and death from pysemia. If it should be thought desirable to perform cholecysten- terostomy in cancer of the head of the pancreas, it will be desirable to carry out the operation with great expedi- tion, and in such a case it may be justifiable to make the anastomosis to the colon, or, better, to a loop of the jejunum which can be brought up on the right side of the great omentum. The late von Mikulicz suggested an entero-anastomosis of the jejunal loop, as shown in the diagram (Fig. 140), but this has never been found necessary in our experience, and as it prolongs the operation by a few minutes, we do not think it should be carried out in any patient very seriously ill. If the adhesions around the duodenum are not too 30 466 The Pancreas: Its Surgery and Pathology extensive so that much time would be occupied in detach- ing them, the operation of cholecystenterostomy may be facilitated by Kocher's method of mobilizing the duode- Fig. 140. — -I, Anastomosis of gall-bladder with duodenum; 2, anastomosis of gall-bladder with jejunum; 3, anastomosis of gall- bladder with small intestine and lateral anastomosis of intestine; 4, anastomosis of gall-bladder with colon. num. This is effected by incising the parietal peritoneum of the posterior abdominal wall vertically about an inch to the outer side of the duodenum ; by inserting the finger Chronic Pancreatitis 467 into this sHt, the loose cellular tissue in front of the kid- ney is easily stripped, and the duodenum can be displaced inwards and, without difficulty, brought forward and clamped before the anastomosis is made, but, as a rule, mobilization of the duodenum is unnecessary if the curved clamps (first invented by Sir Thomas Smith for intestinal work and shown in the diagram. Fig. 141) are applied to the duodenum when the patient's hepatic region is made to bulge forwards, either by the special table we use, or by the sand-bag under the back. Having grasped the fundus of the gall-bladder and the nearest part of the duodenum separately in clamps, the Fig. 141. — Intestinal clamp employed in the operation of cholecyst- enterostomy. two viscera are approximated and the serous surfaces united by a suture of fine Pagenstecher's thread for at least an inch, or better li inches, in length; in front of this the two viscera are incised to the extent of an inch and the margins of the incision in the two viscera are united all around by a fine continuous chromic catgut suture, after which the Pagenstecher suture is continued around the front of the circle outside the marginal suture until it reaches the starting-point, where the two ends are knotted and cut short. The anastomotic opening is thus secured by a marginal continuous catgut suture, which unites all the coats of the two viscera, and an external continuous suture of 468 The Pancreas: Its Surgery and Pathology Pagenstecher's thread, which unites the serous and sub- serous coats external to the marginal sutures. The junction may be made by a Murphy button, or by a Mayo Robson's decalcified bobbin, but we prefer the method of union by suture, which is both expeditious and effectual. The advantage of cholecystenterostomy in interstitial pancreatitis is that it provides for permanent drainage of the bile-passages, so that should the inflammatory tissue in the pancreas further contract and cause increased pressure on the bile-duct, no jaundice will occur, and should the inflammatory process in the pancreas take some weeks or months to completely pass away, the patient will not be distressed by the presence of a biliary fistula or inconvenienced by the absence during that time of the bile from the intestines; moreover, there will not be the anxiety of any further operation to face, as there would be after a cholecystotomy. But, besides the relief of jaundice, the operation acts on the pancreas by reliev- ing tension and thus enabling the gland to discharge its contents, which when infected and imprisoned in the ducts, tends to keep up infiammation and to lead after a time to atrophy of the gland substance proper and the formation of fibrous tissue. It will be gathered from the foregoing arguments that we believe the operation of cholecystenterostomy to be the operation of choice for the treatment of interstitial pancreatitis, and only in case of absence or contraction of the gall-bladder, or in case of unusual difficulties from adhesions or from the serious condition of the patient, would we counsel cholecystotomy being done. Results. — In considering the after-results of the surgi- cal treatment of the class of cases under consideration it is necessary to give both the immediate risks of opera- tion and the ultimate issue of those cases that recovered. To this end letters were addressed to the friends or medi- Chronic Pancreatitis 469 cal attendants of all patients who had not recently been heard of. Of one hundred and two operations undertaken in patients where chronic pancreatic trouble constituted the chief disease, or where it formed a serious complication of other diseases, 96.1 per cent, of cases were followed by recovery, giving a mortality of 3.9 per cent., but since compiling the foregoing figures, in 1904, our experience has very largely increased, and the mortality has dimin- ished to a little over 2 per cent. Of the four cases that died, one was a cholecystotomy undertaken in a patient very deeply jaundiced and reduced to the last stage of exhaustion before a surgical opinion was sought, and where at autopsy a cirrhotic condition of the head of the pancreas was found. The second was a cholecystenterostomy undertaken in a deeply jaundiced patient in . the presence of extensive adhesions, which, on account of the feeble condition of the subject, seemed too formidable to deal with. In this case a necropsy revealed a calculus in the pancreatic por- tion of the common bile-duct, occluding the opening of the pancreatic duct, which would have been discovered had the patient's condition permitted a thorough ex- ploration. A third case was in a very feeble patient operated on away from home, extremely jaundiced, and suffering from repeated rigors. Drainage was imperfectly carried out, and she died of choleemia two weeks later. And a fourth case was a choledochotomy in an aged, feeble man, who died of heart failure, accelerated by intestinal hemorrhage, in the third week after operation, when the wound had healed. In the fifty-five cases of catarrhal interstitial pancrea- titis where gall-stones were found obstructing the pan- creatic portion of the common duct, choledochotomy was performed in forty-two, cholecystotomy in nine, and cholecystenterostomy in four. 470 The Pancreas: Its Surgery and Pathology Of the fifty-two patients that recovered, forty-eight were Hving and well when last heard of ; one is apparently well nine and a half years subsequent to operation, though sugar has recently been found in his urine ; one died from cirrhosis of liver and ascites a year after, it being present and far advanced at the time of operation. Another has since died of acute bronchitis, and another from some other non-specified ailment. In one case where the cause was pancreatic lithiasis, and where calculi were removed both from Wirsung's and Santorini's ducts, the patient is now in very good health. In forty-six cases of interstitial pancreatitis without gall-stones or other removable cause, the bile-ducts, and thus indirectly the pancreatic ducts, were drained in nineteen cases by simple cholecystotomy, in seventeen by cholecystenterostomy, and in five by separation of adhesions and thoroughly freeing the ducts. Of the forty-five patients that recovered, no reply to letters was received from six, who were well some time after operation. The rest were in good health when last heard of, with the exception of one (not drained) who has developed glyco- suria some years after operation, but is otherwise well; one who shows signs of permanent damage to the pan- creas by the urinary test; and one who has anaemia suggestive of the pernicious type. Besides the nineteen cholecystotomies, were five where the pancreatitis was associated with duodenal ulcer, and in these cases a posterior gastro-enterostomy was per- formed at the same time, with good results in every case. It will thus be seen that in a very large percentage of cases the removal of the cause, together with drainage of the bile-ducts, or, in the absence of a removable cause, the simple drainage of the bile-ducts alone, is an opera- tion that may be safely recommended in suitable cases that have failed to yield to general treatment. Chronic Pancreatitis 471 Literature Adams: Lancet, Dec. 9, 1905. Anders: North. Med. Soc. of Phila., Jan. 5, 1904. Ansperger: Rev. de Chirurg., Oct. 10, 1905. Barling: Brit. Med. Journ., Feb. 24, 1905; Ibid., April 25, 1903. Barnard: Clin. Journ., June 14, 1905. Boeckmann: St. Paul's Med. Journ., Jan., 1904. Bosanquet: Lancet, i, 1904. Bottersby: In Fauconneau-Dufresne, 1856. Carnot: Thesis, Paris, 1898. Chalmers: Journ. of Ceylon Tr. Brit. Med. Assoc, i, 2, 1904. Cotter: Brit. Med. Journ., May 5, 1906. Dalziel: Brit. Med. Journ., 1902. Davian: Thesis, Paris, 1906. Deaver: North. Med. Soc. of Phila., Jan. 5, 1904; Amer. Journ. of Med. Sci., Feb., 1903. Deaver and Miiller: Amer. Med., March 19, 1904. Delbet: Cong, franc, de Chir., 1902. Desjardins: Thesis, Paris, 1905. Ehler: Wiener klin. Wochen., Dec. 17, 1903. Estes: Journ. Amer. Med. Assoc, Oct., 1902. Evans: Journ. Amer. Med. Assoc, 1901. Fitz: AUbutt's "System of Medicine," iv, 264. Gamges: Midland Med. Soc, Feb. 17, 1904. Giordano: Thesis, Perrin, Lyons, 1902. Gosset: Thesis, Desjardins, 1905. Hardin: Med. Bull., Quebec, 1904. Jaboulay: Lyon medical, 1898. Kehr: "Technic des Gallensteinoperationen," 1905; Munchener medic. Woch., April 5, 1904; XXXIII. Kong. d. D. gesellsch. f. Chir., 1904. Klippel and Lefas: Arch. gen. de Med., July, 1899; Ref. de Med., Jan. 10, 1903. Korte: Chir. des voies billiares, 1905; Deutsche Chir., 1898; Centralbl. f. Chir., 27, 1904. Lejars: Cong, de Chir., Paris, 1905. Lowe: Brit. Med. Journ., Feb. 27, 1904, p. 493. Mahomed: Brit. Med. Journ., July 2, 1904. Marcy: Journ. Amer. Med. Assoc, Oct., 1902. Mayo: Med. News, June 11, 1904. Megnin and Nocard: Friedberger and Frohner's "Vet. Path.," tr. Hayes, 1905. Moore: Lancet, July 2, 1904. Mosetig-Moorhof : Wiener med. Presse, Jan., 1902. Moynihan: Lancet, June 6, 1903; Brit. Med. Journ., Dec. 31, 1904; Edinb. Med. Journ., May, 1906. Myles: Brit. Med. Journ., 1902. Opie: "Diseases of the Pancreas," 1903. Oser: "Nothnagel's Encyclop. of Pract. Medicine." Owen: Brit. Med. Journ., 1902, p. 1310. Qu^nu: Bull. Soc. Chir., March 7, 1905. Quenu and Duval: Rev. de Chir., Oct. 10, 1905. Raven: Brit. Med. Journ., Oct. 15, 1904. Raynds: Quoted Villar, Cong, of Surgery, Paris, 1905 Riedel: Berl. klin. Woch., i, 1896. 472 The Pancreas: Its Surgery and Pathology Robson, Mayo: Lancet, July 28, 1900; Philadelphia Med. Journ., June I, 1901; Lancet, March 19, 26, April 2, 1904. Montreal Med. Journ., Nov., 1904; Lancet, Dec. 23, 1905; Edin. Med. Jour., Dec, 1905; Surgery, Gynecology, and Obstetrics, Jan., 1906. Rodocanachi : Lancet, July 7, 1906. Schmieden: Miinch. med. Woch., 1906. Segond: Soc. de Chir., Feb. 14, 1906. Sendler: Deut. Zeit. f. Chir., 1896, 1897. Senn: Trans, of the Amer. Surg. Assoc, Phila., 1886. Terrier: Soc. de Chir., Feb. 7, 1906; quoted Desjardins, Thesis, 1905. Tuffier: Bull, et Rev. de la soc de Chir. de Paris, xxxvi, 1905. Walther: Quoted by Desjardins, Thesis, 1905. White, Hale: Brit. Med. Journ., July iS, 1903; Guy's Hosp. Rep., liv, 17. Wiener: Quoted by Quenu and Duval. Zeller: Berliner klin Woch.. 1902. CHAPTER XVII PANCREOLITHIC CATARRH AND PANCREATIC CALCULI Pancreatic calculi are exceedingly rare. Two cases were observed by Panarol and Galea in 1667, one by Morgagni in 1765, and Cowley, in 1788, referred to an instance observed by himself. ]\Iatthew Baillie, physi- %% Fig. 142. — Pancreatic lithiasis (Baillie). cian to St. George's Hospital, in a work on "Morbid Anatomy," published in 1799, figures a case of pancreatic calculi most carefully dissected and showing the relation of the bile-duct and pancreatic duct (Fig. 142). 473 474 The Pancreas: Its Surgery and Pathology In 1883 Johnston collected the notes of thirty-five recorded cases. The fullest account was given in 1896 by Guidiceandrea and was based upon forty-eight recorded cases and two observed by the author. Others have since been reported, but they are so uncommon that Oser said there were in 1903 only seventy recorded cases. We have been able to collect others described since that date, and not included in that series, but doubtless there are many un- recorded. The subject of pancre- atic concretions can never assume the importance that attaches to cholelithiasis, but as pancreatic calculi are associated with serious and usually progressive disease of the gland, and as they can be re- moved by operation, their recog- nition and treatment are matters that demand some consideration. Calculi are never found in a healthy pancreas, and it seems highly probable that, like gall- stones, pancreatic concretions are the result of catarrh of the ducts with stagnation of secretion, which generally, if not always, results from infection. Instead of calculi being formed, the ducts may actu- ally be lined with calcareous mate- rial that may accumulate so much as to close completely, or almost completely, the lumen. An example of this, taken from a specimen in the University College Hospital Museum (Fig. 143), illustrates this condition. It will be noticed that the duct of Wirsung is widely dilated and that it also contains calculi. An interesting case in which the head of the pancreas Fig. 143. — Calcifica- tion of the duct of Wir- sung with atrophy of the pancreas (Univ. Coll. Hosp. Museum, 3197). Pancreolithic Catarrh and Pancreatic Calculi 475 was infiltrated with calcareous material has been reported by Delageniere. The patient was a man of fifty-nine, who complained of violent epigastric pains followed by increasing jaundice. At operation the gall-bladder was foiind to be distended and to contain a biliary calculus. A resistance was felt in the region of the common duct which was thought to be a large calculus, but, on careful examination, it was found to be a hardening of the head of the pancreas around the common duct. On making an incision into the affected area it was found to be densely infiltrated with small, hard, calcareous granules, 15 grams of which were removed with a ciurette. The cavity thus formed was drained and the patient eventually made a good recovery, being relieved of his symptoms and able to return to work. Microscopical examination of the material removed at operation showed that it consisted of pancreatic tissue infiltrated with calcium salts which formed hard, yellowish-white granules, mostly of the same size, but in some instances agglomerated to form larger masses. The composition of pancreatic calculi is important from the diagnostic point of view, for they contain lime, either in the form of carbonate or phosphate, or, as in one case reported by Mr. Shattock, of oxalate, which latter was found in a cyst. Johnston gives two analyses of pancreatic calculi : I Phosphorus salts 72.30 Carbon salts 18.90 Organic matter 8.80 II Carbon salts 91-65 Magnesium carbonate 4.15 Organic matter 3.00 J. A. Milroy reported as follows on a pancreatic calculus removed by Moynihan: "The stone contains nearly 50 per cent, of calcium carbonate, A small portion of the 476 The Pancreas: Its Surgery and Pathology solution in which the magnesium was estimated was unfortunately used for qualitative testing, so that I can- not state the exact quantity of magnesium present. These were the only inorganic substances found. I was somewhat surprised to find phosphates absent. The organic substances consisted almost entirely of proteid. Traces of organic substance soluble in alcohol and ether were present. In the residue from the alcoholic and ethereal solutions cholesterin and fat were the only bodies identified. Purin bases and uric acid were absent. The quantity of the original pow- Jtk -^H^ der was rather ^BP >^i1.'"' Wr too small to al- low of an accurate estimation of the constituents." In examining the urine of cases suffering from chronic pancrea- titis, oxalate of lime crystals have been found in over 40 per cent, of our cases. In jaundiced cases in which the bile acids take up the lime salts they have been found only in 6 per cent, of cases. It would be inter- esting to know if this fact has any bearing on the compo- sition of pancreatic concretions, for it is a well-known fact that the normal pancreatic secretion contains no cal- cium carbonate. The subject is worthy of further investi- gation. The consequence of their chemical composition is that pancreatic calculi are opaque to the ^-rays, and in this Fig. 144. — Gall-stones and pancreatic calculi. Pancreolithic Catarrh and Pancreatic Calculi 477 way we have a means of diagnosing their presence and of differentiating them from gall-stones, which are not seen in a skiagram. An ;\;-ray photograph of concretions taken from a case where at the same time gall- stones and pancreatic calculi were removed by one of us is shown in the figure (Fig. 145), and for com- parison an ordinary photograph of the two classes of con- cretions together is also shown (Fig. 144). in the skiagram is readily seen. The next photograph, 'm Fig. 145. — .T-Ray photograph of the pancreatic calcuU and gall-stones shown in Fig. 144, showing that the former are opaque to the rays. The difference in their appearance ,.^^ Fig. 146. — Skiagram of pancreas containing calculi; vessels injected. taken by Dr. J. Mackenzie Davidson, shows the calculi in situ, and as the vessels have been injected they also 478 The Pancreas: Its Surgery and Pathology show distinctly. So far as we are aware, this method of diagnosis was suggested for the first time by one of us in the Hunterian Lectures delivered at the Royal Col- lege of Surgeons in March, 1904. By means of this and the urinary pancreatic reaction it will probably be possible to confirm the diagnosis by demonstrating the associ- ated chronic pancreatitis. Fig. 147. — A pancreas, with cal- culi of various sizes in its ducts, which are dilated (Museum R. C. S., specimen No. 2833). Fig. 148. — Some of the larger calculi from the specimen shown in Fig. 147 (Royal Coll. of Surg. Museum, 2834). The stones are rounded, ovoid, or elongated like a date-stone. They are found in all parts of the ducts of Pancreolithic Catarrh and Pancreatic Calculi 479 the pancreas, though much more frequently in the head ; in the tail of the gland they are rarely seen. The calculi may be branched like coral, the trunk of the stone lying in the main ducts and its offshoots in the secondary ducts, but they are usually smooth. Shupmann has recorded a calculus measuring 2^ inches by ^ inch and weighing 200 grains, and Matani has reported one weigh- ing 2 ounces. In colour they are pale and they may be white, but if they pass into the common bile-duct they receive a covering of cholesterin and may be stained by the bile so as to look like gall-stones. Concretions, consisting chiefly of carbonate of lime, are sometimes found in the pancreatic ducts of cattle. They vary in size from a millet-seed to a hazelnut and are white, cylindrical, angular, or facetted. The duct and its chief tributaries are generally dilated like a string of beads and its walls are thickened. Symptoms. — The symptoms depend on the associated condition, whether that be cyst, abscess, chronic inflam- mation, or other pathological state; doubtless in some cases symptoms are vague, or even wanting, and in some cases pancreatic calculi have only been discovered post- mortem. Pains at the epigastrium radiating towards the inferior angle of the left scapula, often agonising in charac- ter and associated with vomiting, may be present, and the attacks may be brought on by exertion, or they may be irregular, coming on at any hour, day or night. The pain frequently comes in sharp colicky attacks, similar to, but less severe than, those due to gall-stones. A sense of soreness or stiffness is noticed for a day or two after the attack. When the pain is at its height, vomiting, hiccough, rigors, cold sweats, or collapse may be noticed. After the attack some fragments of stone may be found in the motions (Minnich, Leichtenstern, Kinnicutt). That pan- creatic colic is associated with the passage of stones down the ducts seems clearly to be proved by the cases observed 480 The Pancreas: Its Surgery and Pathology at different times by Minnich and Holzmann, for it was only after each attack that fragments of stone were found in the motions. Dyspepsia and flatulence are usually present. Liporrhoea and azotorrhoea, as well as bulky pale motions, are present where there is well-marked in- terstitial pan- , creatitis, and in some cases where the inter- stitial changes have advanced to atrophy or fatty degenera- tion of the whole gland, glyco- suria is found. Glycosuria was recorded by Lancereaux in twelve out of forty cases, but his statement that in each at- tack of colic there is a tem- porary glycosuria has not been borne out by the experi- ence of subsequent observers. The presence of sugar may be observed at intervals. Holzmann found it in- Fig. 149. — Pancreatic lithiasis (Leeds Museum). Pancreolithic Catarrh and Pancreatic Calculi 481 termittently in his case, but while the same patient was under Minnich's observation at an eariier date, no sugar was found, though regularly sought. Caparelli records the case of a woman who developed, after many attacks of acute epigastric colic, an abscess above the umbilicus. The abscess burst and discharged some pus and gritty material. Through the fistula, which persisted for six years, many small stones, over one hundred in all, were expelled. After the spontaneous closure of the fistula, diabetes developed and the patient died. The pancreatic re- action in the urine was well marked in a case operated on by one of us, to be re- ferred to subsequently, and it will probably be generally found. If a calculus descends into the ampulla of Vater, jaundice will ensue and the case will prob- ably be diagnosed as one of gall-stones in the common duct, but the pancreatic reac- tion and the use of the x-rsiys should enable a differential diagnosis to be made. In a case recorded by Korte a patient sufi:ered from biliary colic for which the abdomen w^as opened. A large calculus was removed from the gall-bladder, which was drained. While the bile was still discharging the patient experienced an attack of pain similar to that for which operation had been undertaken. After death a tumour in the head of the pancreas, which during life had been 31 Fig. 150. — Section of the fibrosed pancreas in a case of pancreatic lithiasis (X ca 40). 482 The Pancreas: Its Surgery and Pathology diagnosed as malignant, was found to be an abscess with large concretions in it. Kummell records an almost exactly similar case in which cholecystotomy had been performed ; pain recurred, and at the necropsy a large soft calculus was found in the canal of Wirsung. Kinnicutt reports an interesting case in a woman, aged forty-two, who had had three attacks of sudden severe pain, beginning in the back and running around the right side along the lower intercostal spaces, with nausea and vomiting. After an interval of eight months, another extremely severe attack occurred. The pain began, as before, in the back between the scapulae, but on this occasion it ran through — not around — into the epigastrium, and became localised to the right of the middle line. On the sixth day after the commencement of the attack six small stones, the size of a pea, were passed per rectum. Four of these were analysed and found to be composed of carbonate and phosphate of lime with no trace of cholesterin or bile-pigment, thus indicating their origin in the pancreatic ducts. Similar stones, or detritus of similar composition, were recovered from the stools during more than one subsequent attack of colic. Some of the patient's later attacks were associated with jaundice and on one occasion two typical gall-stones were recovered from the stools. This case, the author points out, shows the difficulty there is in distinguishing between the pres- ence of biliary and pancreatic calculi. There is nothing distinctive in the nausea, vomiting, diarrhoea, character of pain, or jaundice. The points which are helpful are an rjc-ray photograph, the finding of the calculus and its analysis, glycosuria, and a deficient splitting of ingested fats into fatty acids and soaps. In the case described fat-absorption was normal, but of the fat recovered from the faeces 42.6 per cent, was in the form of neutral fat. Treatment. — ^According to the results of the experi- mental work of Kiihne and Lea, the subcutaneous in- Pancreolithic Catarrh and Pancreatic Calculi 483 jection of pilocarpine incites the flow of the pancreatic juice but the treatment of pancreatic calculi by sialagogues is probably useless and mere waste of time, although in a case reported from Eichhorst's clinic of "undoubted pancreatic lithiasis," in which subcutaneous injections of pilocarpine were tried, "the attacks of colic disappeared completely and the patient was better than for many months before." Relief to pain may be given by sedatives, and other treatment must be adopted as occasion arises, but as soon as pancreatic stones can be diagnosed, they should be removed, as destruction of the pancreas is otherwise certain, and it is quite clear that medical treatment can do no real good in these cases. Surgical treatment has until quite recently been merely palliative, but fortunately surgery can now offer a rea- sonable hope of cure. The pancreatic duct can be readily explored by an incision in the second part of the duodenum, and by then laying open the biliary papilla the opening of Wirsung's duct can be seen. From this a probe can be passed two inches along the duct to explore it, and by this method we have removed a pancreatic calculus from the duodenal end of the duct. A very exhaustive search through English and foreign literature has only resulted in the discovery of five opera- tions for pancreatic calculi. ]\Ir. A. Pearce Gould's case, operated on March 3, 1896, died on the twelfth day from exhaustion. In Dr. Dalziel's case a stone of the size of a very large pea was removed from the pancreatic duct through an incision in the duodenum, the opening in the duct being stitched to the wound in the posterior wall of the duodenum. As the bile-duct was clear there was no jaundice. A good recovery followed. In Mr. B. G. A. Moynihan's case a pancreatic stone was removed from the ampulla of Vater through an incision in the duodenum and the patient recovered. In Dr. L. W. Allen's case two 484 The Pancreas: Its Surgery and Pathology calculi were removed from a cyst between the lesser curvature of the stomach and the liver. The patient died on the fifth day after operation. In a case which came under the care of one of us on February 13, 1903, four calculi were removed from a woman aged fifty-seven, one from the duct of Santorini, or one of its branches, by direct incision into the pancreas close to the common duct, afterwards closing the opening by deep and by peritoneal sutures ; the second and third stones were reached through an incision in the duodenum by laying open the papilla, when by means of fine for- ceps a calculus was removed out of Wir- sung's duct, along which a probe was afterwards passed for two inches, and a fourth concretion was rem.oved by di- rect pancreatotomy from the middle of the duct of Wirsung, the stone being reached by incising the gastro-hepatic omentum, drawing the stomach downwards, incising the pancreas freely, and opening the duct directly on to the stone, which was of the size of a small bean. The duct was then closed with catgut, the wound in the body of the pancreas being sutured so as to leave no dead space and the peritoneal wounds being closed without direct drain- age. The right kidney pouch was then drained, as some infected bile had escaped. Recovery was ultimately com- plete. In this case pain and vomiting were marked features Fig. 151. — Pancreas showing calculi in the duct of Wirsung (St. George's Hosp. Museum, 203). Pancreolithic Catarrh and Pancreatic Calculi 485 and the pancreatic reaction was of the utmost importance from the point of view of diagnosis. This is, apparently, the first case in which either the duct of Wirsimg or the duct of Santorini has been deliberately opened, and, after the removal of a calculus, closed by a suture. The Operation of Pancreo-lithotomy. — For the purpose of removing calculi from the pancreas an incision 3 or 3^ inches to the right of the middle line will be found the most convenient, as the fibres of the right rectus can be split and the incision lengthened upwards and downwards without unnecessarily weakening the abdominal wall. A sand-bag under the lumbar spine will bring the gland several inches nearer the surface. If the opening of the duct of Wirsung has to be explored, the second part of the duodenum may be incised and the papilla common to the bile-duct and pancreatic duct laid open, when the edges of the opened diverticulum of Vater can be seized with small catch forceps and drawn to the surface; a probe or fine forceps can then be readily passed into Wir- sung's duct and any concretion removed. If the calculi are more deeply placed in the ducts, the pancreas may be exposed either through the gastro-hepatic omentum by drawing the stomach downwards, or by lifting the stom- ach it may be reached through a slit in the omentum or by raising the colon, by a slit in the transverse meso- colon ; or by peeling the duodenum from the parietes the back of the pancreas may be readily reached. The calculi may be then cut down on and extracted by a scoop or forceps. Any bleeding must be arrested by ligatures. The duct can be sutured and the incision in the gland must be brought together by buried sutures, the periton- eal covering being coapted by a continuous suture. If leakage is feared a gauze drain may be applied, but the position may be difficult for this, and if it has to be done the gauze must be surrounded by a rubber drainage-tube and brought through it to the surface. In the case of 486 The Pancreas: Its Surgery and Pathology pancreo-lithotomy above referred to the closure of the gland was so secure as not to require gauze packing, and the result justified its not being used. When the duode- num is opened it must be closed in the usual way by a muco-muscular and serous suture, the latter being of fine celluloid thread. The incised papilla need not be sutured. If a calculus be felt in the head of the gland, but not in the duct of Wirsung, it may be reached by incising the peritoneum over the duodenum and separating it gently from the head of the pancreas, or if more deeply placed near the back of the gland the reflection of peritoneum from the duodenum to the abdominal wall may be incised and the duodenum may then be displaced inwards, when the back of the pancreas will be exposed, and, if thought advisable, it may be incised and treated as in the incision from the front. Literature Allen: Annals of Surgery, 1903, p. 741. Ancelet: "Etudes sur les mal. des. Pancreas," Paris, 1864. Baillie: "Morbid Anatomy," 1799. Caparelli: Arch. ital. de Biol., 1894, xxi, 398. Cowley: Lond. Med. Journ., 1788. Dalziel: Privately communicated. Delageniere: Arch. Provinc. de Chirurg., 4, 5, 1906. Eichhorst: Eulenberg's Realencyklop., ii. Gaeia: Graaf, "De succo pancreat.," 1667. Guidiceandra : II Policlinico, 1896. Gould: Anat. Museum of Boston, 1847, P- ^47- Gould, Pearce: Trans. Clin. Soc. of Lond., 1896. Graaf: "Opera omnia," 1667. Holzmann: Miinchener med. Wochenschr., 1894, Nr. 20. Johnston: Amer. Journ. of Med. Sci., 1883. Kinnicutt: Amer. Journ. of Med. Sci., 1902, cxxiv, 948. Korte: Berliner Klinik, Dec, 1896. Lancereaux: Journ. de med. enterne, Feb., 1889. Leichtenstern : "Handb. d. spec. Therap. v. Penzoldt-Stintzung," 1896, iv, 203. Matani: Giorn. di med. Venezic, iv, 174. Minnich: Berliner klin. Wochenschr., 1894, S. 187. Morgagni: "Opera omnia," iii, 68, i. Moynihan: Lancet, Aug. 9, 1902, p. 355. Oser: Nothnagel's "Encyclop. of Pract. Med." Panarol: Jatzologismorium, Roma, 1652, p. 51. Park, Roswell: Amer. Med., 1903, v, 949. Phillipps: Brit. Med. Journ., Feb. 20, 1904; Clin. Soc. Trans., 1904. Robson, Mayo: Hunterian Lectures, Lancet, March 19, 26, April 2, 1904. Shattock: Trans. Path. Soc. of London, 1896, xxi, 4; Brit. Med. Journ., 1896, i, 1034. CHAPTER XVIII PANCREATIC CYSTS Although cysts of the pancreas cannot be said to be of frequent occurrence, they have to be taken into account in the diagnosis of any cystic tumour in the abdomen ; for, as will be seen later, they may appear in various regions and may simulate many other diseases. A search through the literature reveals the fact that, excluding thirteen cases in our own experience, one hun- dred and sixty cases of operation for pancreatic cysts have been recorded. Although larger numbers have been reported in various works, the above figure is probably as nearly correct as possible; for on verifying the records, the same case had sometimes been reported twice, and, in many, the details were so meagre that the nature of the operation was not even given. Dr. Hale White has recorded the fact that in nearly six thousand post-mor- tem examinations at Guy's Hospital from 1883 to 1894, pancreatic cysts were only found in four cases, and one of these was a hydatid cyst. Cysts of the pancreas may be divided into false and true. The false or pseudo-cysts may be due to a distension of the lesser peritoneal sac, or to a localised collection of fluid in the neighbourhood of the pancreas. Seeing that simple drainage is usually sufficient to bring about relief or cure of the disease, surgery offers a poor opportunity for pathological intervention, since experience has shown that the patient's interests are best considered by a limitation of the incision to a size suffi- cient to empty and drain the cyst, and not sufficiently large to satisfy pathological investigation; hence it is 487 488 The Pancreas: Its Surgery and Pathology highly probable that many reported cases of operation for pancreatic cysts have been for cysts of other organs, and it is an undoubted fact that quite a number of the cysts supposed to originate from the pancreas are pseudo-cysts. True cysts may be due to retention from various causes, to parasitic disease, — e.g., hydatids, — to new-growths, as in proliferation cysts, and to haemorrhage. A few cases of congenital cystic disease have been recorded. The greater number of chronic cases that come un- der the care of the surgeon are due to retention of the gland secretion, the outflow of which is hindered in some way. Senn found that ligature of the pancreatic duct did not result in the formation of a cyst, though chronic or inter- mittent obstruction might re- sult in cyst-formation ; just as ligature of a ureter, or acute obstruction, leads to atrophy of the kidney, though chronic obstruction or an obstruction of an intermittent character tends to the development of hydronephrosis. The outflow of secretion in the pancreas may be hin- dered in different ways by obstruction of the excretory duct, or by a combination of compression from without and obstruction from within. The most frequent cause is probably chronic interstitial pancreatitis, in which compression and constriction of the ducts result from the development and contraction of connective tissue, thus leading to stagnation of the secretion. Wirsung's duct Fig. 152. — Calcification of the orifice of the duct of Wir- sung and dilatation of the duct (Univ. Coll. Museum, 3196). Pancreatic Cysts 489 may be closed by gradual compression, as, for instance, in the development of a tumour along its course, or by the gradual development of a duodenal tumour, or a stricture due to ulceration, which compresses the orifice of the duct. Pressure by swollen lymphatic glands, or by adhesions near the head of the pancreas, or even by a gall-stone or Wirsung's duct, may lead to stagnation of secretion, and thus to cystic development. Occasionally a cyst of the pancreas may result from chronic pancreatitis due to ulceration extending into the pancreas from a chronic ulcer of the posterior wall of the stomach, as in a case which was treated successfully by gastro-enterostomy, and at the same time drainage of the cyst, by one of us. Large cysts may also be caused by obstruction within the duct, as, for instance, by a pancreatic calculus or by a gall-stone in the ampulla of Vater. Doubtless some cysts are altogether independent of obstruction and cannot be accounted for by any of these explanations. A particularly interesting case has been reported by McPhedran, in which a pseudo-cyst and a true cyst were observed in the same individual : " G. A. B., male, aged fifty-three, in 1891 had an attack of biliary colic, with well-marked jaundice and pale motions. Had two or three similar attacks every year. Condition became gradually worse and there was almost constantly some epigastric discomfort, indigestion, and flatulence. One severe attack of pain lasted three days. The epigastrium was tender and pain radiated in several directions. Was losing flesh. On examination there was a thickening to be felt on deep pressure in the epigastrium. Three days later a large, smooth, cyst-like tumour was found in the epigastrium, extending from the right para- sternal line to the left mammary line and down to the umbilicus. The upper boundary was ill defined. The stomach resonance was above and to the left. A C5^stic 490 The Pancreas: Its Surgery and Pathology collection in the bursa omentalis was diagnosed, and the abdomen opened. The cyst was emptied. At the bot- tom lay the pancreas, irregularly enlarged and firm, but somewhat elastic. The peritoneum over it was smooth and healthy looking. There was no sign of hemorrhage anywhere. Five months later a tumour was again found in the epigastrium. It extended down to the level of the anterior superior spinous process, and laterally to the mammary line on the right, and the anterior axillary line on the left. It forced the diaphragm upwards, so that the cardiac impulse was in the fourth intercostal space. The abdomen was again opened and a cyst exposed lying behind the stomach. The cyst wall was about 2 mm. thick. The fluid was opaque, whitish, and contained many flocculi and masses of fibrin. It was alkaline in reaction, and contained albumin, but no digestive ferment. After drainage the cavity contracted rapidly, but a fistula persisted, and the discharge from this irritated the skin. On examination it was found to possess marked action on albuminoids, fats, and starches, leaving no doubt as to the presence of pancreatic secretion. The condition causing the repeated attacks of colic lay in the pancreas, and may have been a calculus or a localised inflammatory deposit causing mechanical obstruction. In the most acute attacks the symptoms were those of acute pancreatitis." The symptoms produced by a pancreatic cyst vary according to the cause, as well as from the size and the seat of the tumour. They are at first dependent on the disease which leads to the cystic formation, though later the pressure exercised by the tumour itself on the neigh- bouring viscera has to be taken into account. Seeing that cystic disease is generally associated with some pan- creatitis, either local or general, we may expect to find digestive disturbance with loss of fiesh and pain at the pit of the stomach quite early in the disease, preceding by some time the recognition of the cyst at the surface. If the cause be dependent on some obstruction in the duct, we may expect to find paroxysmal pains accompanied by vomiting and followed by jaundice and wasting. Pancreatic Cysts 491 If the interstitial pancreatitis is at all extensive, there will be marked loss of flesh associated with fatty stools, azotorrhoea, and bulky, pale motions, and rarely the presence of glucose in the urine. In all the cases of pan- creatic cyst that we have recently observed there has been a well-marked pancreatic reaction in the urine, indicating catarrh of the pancreatic ducts, or interstitial inflammation. The Rontgen rays may also form a useful help in diag- Fig. 153. — Diagram to show the relations of the peritoneal reflections of the pancreas (Testut) . nosis in certain cases, as they may establish the presence or absence of pancreatic calculi. It is to be borne in mind that there have been cases of pancreatic cyst, presenting very few symptoms except the presence of a tumour, which have been under observa- tion for a long time and have needed no active treatment, but these cases are exceptional. On the other hand, the tumour may be associated with severe pain and distress and with marked digestive and metabolic symptoms. 492 The Pancreas: Its Surgery and Pathology t The physical signs of cysts of the pancreas are by no means constant. A consideration of the peritoneal reflec- tions from the pancreas on to the viscera, and how they influence the ultimate position and relations of pancreatic cysts, will render the reason for this clear (Fig, 153). For instance, a tumour may spring from ,the anterior surface of the head or body of the pancreas above the stomach, between it and the liver, or, between it and the transverse mesocolon. On the state of "^'x"" N\^' 1 distension of the stomach will depend the extent of contact of the tumour with the ab- dominal wall. By distending the stomach with air through a tube, or by giving doses of soda and tartaric acid in sep- arate draughts, the relation of the stomach to the cyst can be readily shown. If a cystic tumour arise from the pan- creas to the right of the omental bursal reflection, it may make its way forwards to the right hypochondrium and simulate a gall-bladder or right renal or suprarenal cyst. Should a cyst arise from the posterior part of the head or tail of the gland, it may project either into the right or left lumbar region and resemble a cyst of the kidney. If a tumour springs from the head of the pancreas below the reflection of the transverse mesocolon, but to the right of the mesenteric vessels, it will reach the surface below the hepatic flexure of the colon on the right side, and may simulate a right renal tumour, or a tumour of the caecum, or ascending Fig- 154- (Figs. 154 to 158 are a series of diagrams to show the direc- tions in which pancreatic cysts may develop.) Fig. i: Fig. 15 7- Fig. 156. 493 Fig. is8. 494 The Pancreas: Its Surgery and Pathology colon, as the mesentery will prevent it passing to the left of the spine ; but should it arise from the small por- tion of the processus uncinatus on the left of the mesenteric vessels, but below the attachment of the transverse meso- colon, it may burrow between the layers of the mesentery and simulate a mesenteric cyst, or it may bulge on the left of the mesentery and reach the surface below the transverse colon on the left of the spine, when it may resemble a left renal or ovarian cyst, or a tumour of the descending colon, or small intestine. A tumour arising from the body or tail of the pancreas above the reflection of the transverse mesocolon will pass upwards beneath the left costal margin, and resemble a cyst of the spleen, or of the left lobe of the liver. A pancreatic cyst in this region may be opened and drained under the idea that it is a cyst of the spleen, and a chronic abscess of the spleen may, on the other hand, be opened and drained under the idea that it is a cyst of the pancreas. Tumours springing from the pancreas on the left of the duodeno-jejunal junction, where the lower surface of the gland lies on the transverse mesocolon, have a tendency to press the great omentum forward and to project above the transverse colon, but they may grow downwards towards the central region of the abdomen and arch the transverse colon, or even project below it, so that the colon lies above the tumour. The relationship of the colon to the cyst may be ascertained by distending the colon with air introduced per anum. In an interesting case recorded by Dr. S. P. Phillips a thin-walled pancreatic cyst spring- ing from the head of the pancreas completely filled the abdomen and presented the physical signs of ascites. The explanation of these variations, which may, and often do, lead to difficulties in diagnosis, is an anatomical one, and depends on the site of origin of the cyst, which in making its way to the surface proceeds in the line of least resistance, and is thus influenced by the reflections Pancreatic Cysts 495 Fig. 159. — a, Traumatic pancreatic effusion into the lesser peritoneal sac, in a boy set. two years, knocked down by a cab ; b, cyst of pancreas treated by incision and drainage, man set. thirty-five years; well seven years later. Fig. 160. — a, Cyst of tail of pancreas treated by incision and drain- age ; cure, b, Cyst of pancreas from man set. thirty-seven years ; drain- age, recovery. 496 The Pancreas: Its Surgery and Pathology Fig. 161. — a, Pseudo-cyst of pancreas formed around necrosed pan- creas in a man ast. fifty-eight years. Patient in good health two years later, b, Cyst of pancreas treated by drainage; man set. fifty-three years; short fistula remains, otherwise well. Fig. 162. — a, Cyst of body of pancreas; drainage; recovery, b, Pan- creatic cyst resembling ovarian tumour. Pancreatic Cysts 497 of the peritoneum and the arrangement of the viscera overlying the gland- Diagnosis. — A cyst of the pancreas may thus simulate a dilated and tense gall-bladder, a cyst of the liver, spleen, or kidney, an omental or mesenteric cyst, an ovarian or uterine cyst, a cystic dilatation of the bile-duct, a supra- renal cyst, a tubercular peritonitis, or even an ascites. It is evident, therefore, that the presence of a cystic tumour alone, even in a characteristic position, will not justify the diagnosis of cyst of the pancreas, though, as a rule, the combination of symptoms together with the physical signs should leave little doubt in the majority of cases as to the na- ture of a tumour, even before an exploration of the abdomen is done. It used to be a favourite diagnostic method to explore by a hollow needle any cystic tu- mour ; but it can be only under very exceptional circumstances that this aid to diagnosis would be justifiable, as it is by no means devoid of danger from perforation of an overlying vis- cus {e. g., stomach, colon, etc.), or perforation of a large vessel or extravasation of the cyst contents. Not only so, but the examination of the contents will not always make the diagnosis certain. If, however, such an exploration be decided on, it is better to employ a small aspirator needle, and at the same time to completely empty the cyst, which, if tense, would otherwise be liable to leak into the peritoneal cavity, and produce disastrous consequences. While it is easy to say what will be the physical signs on 32 Fig. 163. — Cyst of tail of pancreas, from a woman set. thirty-eight years ; drain- age; recovery. Recurrence; excision of cyst; recovery. 498 The Pancreas: Its Surgery and Pathology percussion and palpation of a cyst appearing above, be- hind, or below the stomach, or above, behind, or below the transverse colon, it will be seen that no one descrip- tion can in any way guide the sur- geon as to the regular signs to be found in a pancreatic cyst reaching the surface. The shape of a cyst varies accord- ing to the way in which it originates from Wirsung's duct, or from the smaller canals within the gland. Thus there may be a rosary-like dilatation of the whole duct, as in a photograph taken from a specimen in the College of Surgeons' Museum, shown in the illustration (Fig. 164). Virchow termed this "ranula pan- creatica," from its analogy to the well-known cystic tumour in the mouth. If . several small ducts are con- stricted, the resulting cysts may be small and multiple, especially if as- sociated with diffuse chronic pan- creatitis (Fig. 165). In case of par- tial cystic-dilatation of Wirsung's duct, large cysts may form which may be oval or rounded and may vary from the size of a fist to enor- mous sacs containing as much as 20 to 30 pints of fluid, though the ordinary size of pancreatic cysts is something between that of an orange and a child's head. The thickness of the cyst ' wall will vary according to the amount of pancreatic tissue entering into its struc- ture, but in some cases it may be quite thin. It should Fig. 164. — Dilata- tion of the duct of Wir- sung (Royal Coll. of Surg. Museum, 2832 A). Pancreatic Cysts 499 not be forgotten that large blood-vessels may be encoun- Fig. 165. — Pancreas showing small retention cysts (Leeds Medical School Museum, E E 203). tered in the walls of the cyst. The lining of the cyst is gen- erally smooth, but in some cases it may be roughened and show ridges and septa the remains of several orig- inal cysts ; or there may be found, adherent to the inner surface of the cyst, clotted remains of profuse hasmor- rhages. The contents of a cyst may resemble water, and may give the appear- ance of a hydronephrosis having been tapped, or the fluid may be thick and slimy. More frequently, however, the contents of the cyst are light brown, or coffee-ground, in colour. The fluid may also be syrup-like and gelatinous, or colloid or purulent. In some cases it may be yellowish- green, as if mixed with bile. Fig. 166. — Retention cyst of the tail of the pancreas (St. George's Hosp. Museum, 202 A). 500 The Pancreas: Its Surgery and Pathology It will thus be seen that the naked-eye appearances of the contents of the cyst do not always form a guide as to its true nature. A chemical analysis of the fluid often affords positive assistance. The fluid from a pancreatic cyst is alkaline in reaction, and generally of low specific gravity, i.oio to 1.020, although Gussenbauer in one case found a specific gravity of 1.160. The amount of solid matter is not high ; thus, Herter analysed the contents of two pancreatic cysts and gives the following results of three examinations : I II III Total solids 24.1% 24.1% 23.8% Organic matter i7-9% i4-9% 18.5% . Ash 6.2% 9.2% 8.7% Albumin is always present, and, occasionally, mucin also. Traces of urea have been noted in some instances, and cho- lesterin is frequently found. A very complete analysis of the contents of a pancreatic cyst has been recorded by Alay and Rispal. In their case the following results were obtained : Reaction feebly alkaline , albumin 8 . 9 grams per mille (serum albumin 5.2 grams per mille, globulin 0.6 gram per mille, albumose 3.0 grams per mille, peptone nil), urea 0.14 gram per mille, uric acid traces, fat and cholesterin o. 1 6 gram per mille, sugar nil, acetone about 0.05 gram per mille, chlorides 5.8 grams per mille, phosphates 0.16 gram per mille, sulphate traces, calcium and magnesium 0.05 gram per mille. The ash contained iron and traces of copper. Microscopically blood cells are usually found. There may be epithelial cells, fat globules, and cholesterin crys- tals. The most important characteristic, however, is the possession by the fluid of digestive powers resembling those of pancreatic juice. When it is found to readily digest albumin, starch, and fat, there can be no doubt as to the nature of the cyst from which it is derived, but the exact value of the discovery of one or other of the fer- ments alone is still a matter of dispute. Korte considers Pancreatic Cysts 501 that a powerful starch-converting ferment is of great diagnostic value, and it is no doubt very suggestive, but it must be remembered that von Jaksch has shown that the fluid from various abdominal cysts, and even from cases of ascites, has distinct starch-converting powers, so that it is now generally acknowledged that the presence of a diastatic ferment alone is of little value in diagnosis. According to Boas, the presence of an albumin-digesting ferment is most characteristic, but in many cases of un- doubted pancreatic cyst this has been absent, especially in old encapsuled cysts. Its absence in more recently formed cysts is explicable by the fluid not having been activated by the enterokinase of the succus entericus, while its presence in other cases may be due to the action of soluble calcium salts, etc., which, according to Delez- enne and Wohlgemuth, have a similar effect to enteroki- nase. Since the fat-splitting power of pancreatic juice is its most characteristic property, it might be expected that the possession of this power by a fluid will at once decide its origin, and, according to our own observ^ations and those of Zeehuisen, this is, in fact, the case. Unfor- tunately this question has not been gone into in many- cases, but as the methods of examination now available are comparatively simple, it is to be hoped that a larger body of evidence will shortly be available. With regard to the results of an examination for fer- ments, it may be said that the presence of all the enzymes in considerable amounts points to a cyst arising from, or directly connected with, the pancreas, while their ab- sence is of no value one way or the other. The presence of a starch-converting ferment alone is of little value in diagnosis, but the proteolytic and fat-splitting ferments are important, and afford presiimptive evidence of a cyst of pancreatic origin. The presence of the ferments does not necessarily mean that the fluid is contained in a true pancreatic cyst, however, for they may be found in pseudo- 502 The Pancreas: Its Surgery and Pathology cysts arising from injury or laceration of the pancreas. In such cases the ferments are furnished by the pancreatic secretion finding its way through the injured or torn peri- toneum into the lesser sac, where they mingle with the inflammatory exudate forming the bulk of the fluid. The digestive powers of a fluid of this description are not so well marked as those seen in a typical pancreatic cyst, but they may be equally as active as those met with in true cysts of some standing, in which the surrounding pancreatic substance has been replaced by fibrous tissue from the associated inflammatory changes. The termination of pancreatic cysts, in the absence of treatment, varies in different cases. There is usually a steady progress of the disease that has caused the cystic condition — as, for instance, in the case of interstitial pancreatitis towards atrophy, and its consequence, diabe- tes; but pressure symptoms may produce danger before this slower termination, or the cyst may rupture into the peritoneal cavity and cause death by shock or by perito- nitis. Rupture into the stomach or intestine has also been known to occur. In some cases pancreatic cysts have existed for many years without producing any serious symptoms, though this is exceptional. Treatment. — It is quite clear that medical treatment can be of no avail in the case of pancreatic cysts, and that surgical treatment alone is available for relief or cure. Aspiration and other forms of tapping are inadequate and ineffectual methods, which are attended with more danger than is the operation of incision and drainage. They are, therefore, not to be recommended, even for diagnostic purposes. Occasionally complete extirpation of the cyst. may be performed, as in a case that came under the care of one of us, where the tumour returned a few months after it had been apparently successfully treated by drainage; but the greater difficulty in performing Pancreatic Cysts 503 excision, its impracticability in certain cases and the greater mortality attending it, as compared with the operation of incision and drainage, make it quite clear that drainage should always have a fair trial unless the circumstances prove to be very exceptional, as, for instance, in the case of a cyst of the tail of the pancreas, or in the case of a pedunculated cyst. As to the situation for drainage, that will depend on circumstances. The tumour will usually be attacked most readily from the front at a point where it very nearly reaches the surface. Occasionally, however, it may be drained from the loin. Fistula does not, as a rule, follow the drainage of pan- creatic cysts, but in some cases a small fistula may persist and may go on for years without hurt to the patient and with very little discomfort. The following is a description of the operation usually performed: An incision is made through the parietes opposite the most prominent part of the cyst. When the peritoneum is opened, the finger can be employed to ascertain the relations of the cyst and its attachments. If the stomach is in front of the cyst, it will be better to displace that viscus upwards and to make a slit through the great omentum in order to expose the cyst wall; if the colon is in front, it may be displaced downwards. But no rule can be formulated, as the cyst must be reached in the most convenient way, and that can be ascertained only when the abdomen is open. By means of an aspi- rator the fluid is then drawn off, and an opening made in the cyst sufficiently large to allow of a drainage-tube being inserted. The tube may then be fixed to the mar- gin of the incision in the cyst by a single catgut suture, and if the opening into the cyst is surrounded by a purse- string suture, which can be tightened around the tube, all fear of leakage from the cyst into the peritoneal cavity is avoided. Any vessels coursing over the cyst must be 504 The Pancreas: Its Surgery and Pathology avoided, but should an artery or vein be pricked, it must be caught between pressure forceps and ligatured. The edge of the cyst may then be fixed to the aponeuro- sis by three or four sutures, but it is better not to attach it to the skin. The abdomen is then closed, and if the tube is sufficiently long it will readily drain into a bottle containing some antiseptic fluid. If, on exploration, the cyst is found to have a narrow attachment to the pancreas and the adhesions are not too extensive, it may possibly be shelled out, or the pedicle may be ligatured, but this is rarely feasible. Some surgeons have suggested the desirability of fixing the cyst to the surface and only opening it after a few days, when adhesions have formed, but this operation h deux temps seems to be quite unnecessary. Statistics.— In the cases that have come under our personal observation, one cyst was enucleated, recovery following; drainage was carried out in ten cases of true cyst, recovery following in nine ; whereas of two pseudo- cysts, one due to traumatic hsemorrhagic pancreatitis and the other to necrotic pancreatitis, one recovered. Of the one hundred and sixty cases of operation re- corded by others, there were one hundred and forty recoveries ; in four cases the ultimate issue was doubtful ; in eight out of the one hundred and forty reported re- coveries after operation the patients died subsequently — one from diabetes four months later, one from hsemorrhage one-and-a-half years later, one from concomitant peri- tonitis seven weeks later, one from a zymotic fever a few weeks later, and three from causes not stated, a few weeks later. Death is recorded as the result of operation in twenty cases. In five of these the cause of death and the time after operation are not given. One patient died in collapse, one died before operation could be completed (the next day), one died from "ileus," one died eighteen days after operation (cause not stated), two died from Pancreatic Cysts 505 shock, one died from gangrene of the pancreas, and eight died from peritonitis ; one died at an interval not stated, one after ninety-six hours, one after six days, one after an exploratory incision, two after two days, one on the eighth day, and one on the second day. In one hundred and thirty-eight cases incision and drainage were per- formed, with sixteen deaths, equal to a mortality of 11.6 per cent. In fifteen excision was performed, with three deaths, equal to a mortality of 20 per cent. In seven partial excision was done, with one death, equal to a mor- tality of 14.3 per cent. Although larger numbers have been reported by others, the above figures are as nearly correct as possible, for on verifying the records sometimes the same case had been reported twice, in others wrong dates had been given, and in a few the details were so meagre that the nature of the operation was not given. The evidence is clearly in favour of drainage, but the mortality should certainly be reduced by one-half. PROLIFERATION CYSTS Many of these tumours are on the border-line between cystic carcinoma and proliferating cystomata, and it is only the subsequent course of events that indicates to which class they belong. The cystic epitheliomata, or carcinomata, are cystic formations, generally multilocular, with cancerous depos- its in their walls. There are usually secondary deposits of growth in the liver or adjoining structures. Only very few cases have been reported, and the presence of metas- tatic growths generally renders them inoperable. The cystadenomata or cystic simple tumovirs, although more common than the malignant variety, are yet uncom- mon. Cumston, writing in the "Annals of Surgery" for February, 1903, was only able to find reports of fifteen cases. They are almost always multilocular, lined with 5o6 The Pancreas: Its Surgery and Pathology cyHndrical epithelium, and form crypts, or polypoid masses, projecting into the cavity of the cyst. They are more common in the tail than the head of the gland. Since their contents are frequently blood-stained, some observers have considered that they are of hasmorrhagic origin and have referred to them as "apoplectic cysts." Others believe that they are formed by the fusion of min- ute cysts resulting from obstruction of the small ex- cretory ducts fol- lowing interstitial inflammation. In support of this view it has been pointed out that no evidence of a proteolytic fer- ment could be found, and that it was possibly used up in digesting the walls separating adjacent cysts. The proteolytic ferment is not, however, always absent, and it is sometimes the starch or fat-splitting ferment that has been lacking. A case in which a multilocular cystic tumour of the pancreas was removed with good results, has recently- been reported by Mr. J. D. Malcolm, and is the second in which this surgeon has operated. Fig. 167. — Thin- walled cyst of the pancreas (Royal Coll. of Surg. Museum, 2834 A). Pancreatic Cysts 507 The patient, a female, aged fifty, felt something move in her abdomen and was very sick about six months before she sought medical advice. This disturbance soon passed off and she "forgot" about the "lump," but she began to lose flesh and strength about that time. A month before she fainted, and was very pale for the following week. The faeces were " like ink" for three days and then resumed their natural colour. Apparently there had been a hasmorrhage into the upper part of the alimentary Fig. 168. — Multilocular cystic tumour of the pancreas (Royal College of Surgeons Museum, A 2835). tract. The patient again became conscious of something abnormal in her abdomen. Her general condition was good, the urine was normal, and the bowels acted well without medicine. There was an oval, hard tumour in the left side of the upper abdomen measuring about 4 inches in its lateral diameter and rather more from above downwards. It had considerable mobility, exactly resem- bling that of a large loose kidney, the greatest fixity being towards the spine. The percussion note over the most 5o8 The Pancreas: Its Surgery and Pathology prominent parts was dull before and behind, the area of dulness varying with the position of the tumour. The right kidney was somewhat mobile. A malignant growth in a loose left kidney, the capsule of which had not yet ruptured, was diagnosed. At the Samaritan Free Hospital, on April 26, 1905, the abdominal cavity was opened by an incision outside the left rectus muscle. The transverse colon was seen to be displaced downwards and pressed forwards by a growth between it and the stomach. There was only one layer of serous membrane over the tumour, which appar- ently had developed to the left of the lesser sac of the peritoneum. The kidney was in normal position behind the neoplasm and the tail of the pancreas lay across the upper inch, or rather more, of its anterior surface, inti- mately attached to it . Many vessels over the tumour were ligatured as they were divided, and the mass was gradu- ally freed without much loss of blood until its only attach- ment was to the pancreas. This gland was cut into so as to get the whole growth away. The pancreatic tissue bled freely, but it showed no friability, and liga- tures applied to it held well. A large vein, quite a third of an inch in diameter, was cut across and there was a profuse hcemorrhage for a moment, but both ends were secured with forceps and safely tied. When the pancreas was released it immediately turned round so that its anterior aspect presented unharmed. Evidently the new-growth had arisen from the posterior surface of the gland, and rather from the upper part of it. The fact that the pancreas had been incised gave rise to no sub- sequent trouble. In enucleating the tumour both layers of the transverse mesocolon were divided and the edges of the inferior layer were brought together by sutures. A drainage-tube was inserted through an incision below the twelfth rib, the anterior abdominal wound was closed, and healing gave no trouble. From the first there was considerable distress caused by vomiting and retention of flatus. The bowels acted when enemata were given, but every attempt to give the patient a sufficient quantity of food was followed by symptoms of a partial obstruction of the bowel, and there was an increasing tendency to intestinal distension. A Pancreatic Cysts 509 great deal of peristaltic movement became visible in the epigastric and umbilical regions and down almost to the pubes. It was thought that the small intestine was adher- ent somewhere, probably to the wound of the transverse mesocolon, and as the patient was losing ground the abdo- men was reopened on the sixteenth day after the operation. It was then obvious that the large intestine was at fault and that the difficulty was caused by an unusually acute angle at the splenic flexure, the mobility of the transverse colon being interfered with by adhesions between it and the anterior abdominal wall. The ascending and trans- verse colons were elongated, distended, and tortuous, so as to fill the area where peristalsis had been visible. The adhesions were released, a lateral anastomosis was made between the transverse and descending colons, and flatus escaped freely from the rectum a few hours after the operation. There was for a time a tendency for fasces to collect in the colon, but there were no further urgent symptoms, and the patient, who had become very emaciated, gradually put on flesh and left the hospital on July 3. She was seen again on March 13, 1906, when her general health was good and the bowels were acting regularly without medicine. She complained of a drag- ging pain in the right side, and the left kidney was much more mobile than it had been, but a belt and pad gave relief. The tumour was a multilocular cystoma measuring about 4 inches by about 3I- inches, with a small amount of solid tissue here and there between the cysts. It is preserved in the museum of the Royal College of Sur- geons of England. Congenital cystic disease of the pancreas is exceedingly rare. In a work on the pancreas by Mayo Robson and Moynihan reference is made to three cases, and we have been unable to meet with a record of any other instance subsequent to these. The condition resembles that met with in other organs, with which it has been associated in at least one case. Hosmorrhagic Cysts. — Many pancreatic cysts contain a considerable quantity of blood, and it has been sought 5IO The Pancreas: Its Surgery and Pathology to distinguish between "hasmatomata," or retention cysts into which blood has escaped, and "apoplectic cysts," resulting from haemorrhage into softened gland substance. Although it is not impossible that a blood-containing cyst may arise in either way, there is no certain criterion by which they can be distinguished, and even if it were possi- ble, the distinction would merely be of theoretical inter- est and of no practical value. The arguments bearing on the question have already been discussed in the chap- ter on pathology. Hydatid cysts of the pancreas are extremely rare, and present no surgical peculiarities calling for special men- tion. They must be treated by drainage and evacuation of the daughter cysts, and on no account must incision of the cyst be attempted, though in some cases it may be possible to completely evacuate the endo-cyst and thus to expedite recovery. Literature Becourt: " Recherches sur le Pancreas," Strasb., 1850. Boas: "Magenkrankheiten," i, 295. Bozenaum and Ganignes: Med. Record, 1882. Brackel: Zeit. f. Chir., xlix, 293. Bull: New York Med. Journ., 1887. Cartledge: Journ. of Amer. Gynecol., Jan., 1903, p. 16. Churton: Brit. Med. Journ., 1894, i, 1190; Lancet, 1894, i, 1374. Glutton: St. Thomas's Hosp. Rep., xxi. Cruveilhier: "Traite d'anftt.," 1856, iii, 366. Cumston: Annals of Surg., Feb., 1903, p. 230; Rev. de Chir., June, 1903. Delezenne: Brit. Med. Journ., Dec. 22, 1906, p. 1785. Dieckhoff: "Beitrage z. path. Anat. des Pank.," Leipzig, 1896. Dixon: Med. Record, March, 1884. Dunning: Amer. Journ. of Obstetr., 1905, p. loi. Duponchel: Med. Rep., xxii, 162. Durante: Cong, d'ital Chir., 1893; Ref. Med., 1893, i'^> 359- Engel: Med. Jahrbuch d. ostern Staates, 1840, S. 411; Ibid., 1841, S. 193. Fisher: Guy's Hosp. Rep., xxxiv. Fitz: Amer. Journ. of Med. Sci., cxx, 184. Gilbert: "Etudes sur les Malades du Foie." Gourand: Gaz. de Hopit. Civils et Milit., April 2, 1904. Graham: "Hydatid Dis. in its Clinical Aspects," 1891. Gussenbauer: Langenbeck's Arch. f. Chir., 1883, p. 355. Hagen: Arch. f. klin. Chir., Ixii. Hagenbech: Deutsche Zeitschr. f. Chir., 1887, xxvii, no. Pancreatic Cysts 511 Hartmann: Cong. Franc, de Chir., 1891; Rev. de Gynecol, et de Chir. Abdom., Sept., Oct., 1905. Haynes: Annals of Surg., 1905, xli, 950. Heaton: Brit. Med. Journ., 1901, ii. Herter: "Lectures on Chemical Pathology." Hollander: Med. Press., Aug. 2, 1905, p. 113. Hoppe: Virchow's Arch., 1857, xi, 96. Horrocks and Morton: Lancet, 1897, ^• V. Jaksch: Prager med. Wochenschr., 1880, S. 193 Kellock: Brit. Med. Journ., Dec. 16, 1905. Klob: Oestern Zeitschr. f. Heilk., i860, 529. Korte: Berlin Klinik, Dec, 1896. Kiihnast: Inaug. Dissert., Breslau, 1887. Lloyd, Jordan: Brit, Med. Journ., Nov., 1892. Ludolph: Inaug. Dissert., Bonn, 1890. Lynn: Lancet, 1894. Malcolm: Lancet, June 16, 1906; Tr. Med. Soc. of London, xxi, 97. Marseron: Thesis, Paris, 1881. Martin: Virchow's Arch., 1890. McPhedron: Brit. Med. Journ., 1897, i- 1400. McReynolds: Annals of Surg., 1905, xlii, 130. Manprofit: Gaz. m6d de Paris, March 12, 1904. Murray: Amer. Med., July, 26, 1902, 133. Narath: Arch. f. Chir., 1, 4. Neumann: Centralbl. f. Chir., xxxii, 1904. Ochsner: Annals of Surg., June, 1906, 949. Oser: Nothnagel's "Encyclop. of Pract. Med." Park, Roswell: Amer. Med., 1903, v, 949. Pepper: Centralbl. f. med. Wissen., 187 1, S. 156. Perry and Shaw: Guy's Hosp. Rep., 1893. Peters: Canad. Pract. and Review, Feb., 1901. Phulpin: Bull. Soc. Anat., 1892, p. 9. Poucet and Cubert: Gaz. des Hopit., 1896. Przewoski and Pawlik: Arch. f. Chir., liii, S. 571. Pye-Smith: Trans. Path. Soc. of Lond., 1885, p. 17. Ranschoff: Amer. Med., July 27, 1901. Richardson: Boston Med. and Surg. Journ., 1891, 1892, 1895. Riedel: Arch. f. Min. Chir., 1885. Robson and Moynihan: "Diseases of the Pancreas," 1903. Roux: Thesis, Paris, 189 1. Salzer and Paltauf : Zeitschr. f. Heilk., 1886. Senn: Amer. Journ. of Med. Sci., July, 1885; Tr. Amer. Med. Assoc, 1886. Sheen: Clin. Journ., March. 29, 1905. Stieda:_Cent. f. Path. Anat., 1893, ^o- 12. Thiroloix and du Pasquier: Bull, de la. Soc. Anat., 1892. Tilger: Virchow's Arch., cxxxvii, 348. Villar: Arch. Provinc de Chir., 1904. Virchow: Die krankhaften Geschwulste, 1863, i, 276. Watkins-Pitchford : Trans. Path. Soc. of Lond., liv, 354. White, Hale: Guy's Hosp. Rep., liv. Wohlgemuth: Biochem. heit., 1906, 2, S. 264. Wyss: Virchow's Arch., xxxvi, 455. Zukowski: Wien. med. Presse, 188 1. CHAPTER XIX NEOPLASMS Tumours of the pancreas are far from common and are usually of a malignant nature. Among the lower ani- mals, according to Nocard and Friedberger, new-growths of the gland are especially common in dogs and are usually of a carcinomatous or adenomatous type. Carcinoma. — Carcinoma is no doubt the most common of the neoplasms found in the pancreas, but until recently it has not figured prominently in the text-books, the reason probably being, as Dr. Sidney Phillips has pointed out, that the secondary nodules which form in the liver during the course of most cases of cancer of the pancreas are so much more readily recognisable both during life and at post-mortem examination than a nodule of cancer in the pancreas, which requires to be searched for, that many cases of cancer of the pancreas have been considered as examples of cancer of the livpr. This is borne out by the fact that since more attention has been directed to diseases of the pancreas, the deaths certified from cancer of the organ have risen, while the deaths certified as from cancer of the liver have fallen in number. Thus the deaths certified to the Registrar-General from cancer of the pancreas increased year by year from 281 in 1899 to 436 in 1904, the most marked rise being from 340 in 1903 to 436 in 1904, while in the same twelve months the deaths registered from cancer of the liver and bile- ducts fell from 3901 to 3736. No doubt many cases of cancer of the pancreas still escape registration as such. Out of 53,000 necropsies gathered from various sources, where post-mortem examinations were presumably care- 512 Neoplasms 513 fully made, there were 226 cases of primary malignant disease of the pancreas (Roswell Park), but as these in- clude Remo Segre's cases from the Ospedale Maggiore, Milan, in which the primary and secondary growths are not separated, the proportion of primary growths is probably not so large. Secondary growths are much more common; for instance, in Eppinger's statistics, of 13 14 necropsies there were 308 cancers in various organs, of which 19 were in the pancreas, but of these only 2 were primary. It seems adidsable to remark here that all past post-mortem records with regard to , .■'->:.'* 'V 7'' '^a cancer of the pancreas -,- ■^^v^'Hllv''^ must be fallacious, as " -* -- ■ ..^w^ until the appearance of -'■'■■ '■'■''■ "'"'^ our paper on chronic pancreatitis in July, .h^, . 1900, cases of chronic "■^-' interstitial pancreatitis causing occlusion of the - ' common bile-duct were not differentiated. ,, The tumour in cancer mav takp the- fnrm nf ^^^- 169.— Primary columnar-celled may laKe me lorm Ot carcinoma of the pancreas (X 50). scirrhus or encephaloid cancer, columnar celled carcinoma, or colloid cancer, and these are given in their order of frequency. Secondary deposits may be found in any situation, but are most common in the liver. In Fig. 170 is shown a metastatic deposit in the liver secondary to cancer of the pancreas, showing how the secondary disease conforms to the type of the primary. Cancer usually occurs after forty years of age, though we have met with a case of cancer of the head of the pan- creas at thirty-two years of age, and rarely it has been known to occur in childhood. 33 514 The Pancreas: Its Surgery and Pathology Symptoms. — The first symptoms are loss of weight and strength with indigestion and general malaise, pain being usually absent or unimportant. When the head of the pancreas is involved jaundice rapidly ensues, the skin assumes a dark almost black colour, unlike the yellow colour of jaundice from gall-stones, the liver swells, and the gall-bladder dilates. The patient then seems almost to dissolve away, the loss of flesh being so rapid. The motions become pale and contain fat and muscle fibre, if meat be taken. Chemical analysis of Fig. 170. — a, Primary spheroidal-celled carcinoma of the pancreas; b, edge of a metastatic deposit in the liver, from the same case, growth to the right, liver to the left ( X 42). the stools shows that there is not only a large excess of unabsorbed fat, but that the "neutral fat" is much in- creased relative to the "combined fatty acid," owing to the absence of the fat-splitting ferment of the pancreatic juice from the intestine. In cancer of the head of the pancreas obstruction of the common bile-duct is, as a rule, absolute at an early stage, and consequently the fasces contain no stercobilin. The urine gives no "pan- creatic" reaction by the improved or " C-method" in most cases, but in some the attendant inflammatory changes cause a more or less marked deposit of crystals. By the Neoplasms 515 original "A-reaction" large, slowly soluble crystals may be secured in most instances, but in some no reaction is obtained, and in others it may be necessary to examine several specimens of urine, and to make more than one preparation from each, before they can be obtained. In interpreting the results of the "pancreatic" reaction in the urine it is important that these points should be borne in mind, and particularly that by the C-reaction a deposit of crystals due to a pancreatitis, associated with the spread of the growth, may be present in some 25 per cent, of cases (see table, page 225), or otherwise they may lead to an error in diagnosis. Sugar is found in the urine only when the whole gland is involved in the growth, or when by secondary interstitial pancreatitis both the secreting parenchyma and the islands of Langerhans have been destroyed. It is therefore not commonly met with. There is, as a rule, no difficulty in diagnosing cancer of the head of the pancreas from chronic interstitial pan- creatitis, for, apart from the usual difference in the re- sults of the pancreatic reaction, the general symptoms, especially the duration and mode of onset of the disease, as well as the information obtained by a complete and thorough chemical and microscopical examination of the faeces, will usually enable the diagnosis to be made. In exceptional cases, especially when the tumour is large and growing rapidly, pain may be severe and excruciating, and, if the stomach is involved, vomiting assumes a prom- inent place. In some instances an analysis of a test meal for free and physiologically active hydrochloric acid may afford useful confirmatory evidence in suspected cases of cancer, but although their absence may be taken as supporting a diagnosis based upon other grounds, their presence, even in considerable amounts, cannot be de- pended upon as indicating the simple nature of a tumour. The haemorrhagic tendency in connection with cancer of 5i6 The Pancreas: Its Surgery and Pathology the head of the pancreas and jaundice is well known, but even in cancer of the body or tail of the organ a hsemor- rhagic condition may ensue. Probably the excretion of lime salts from the blood may account for the hsemorrhagic condition and for the relief that can be given by the use of calcium chloride. Bleeding from the stomach or from the nose and mouth, and from the intestine, or haemor- rhages under the skin are apt to occur spontaneously, and to become serious, or even fatal, and in case of opera- tion, unless the blood be previously charged with lime salts, bleeding is likely to occur in the shape of persistent oozing, both at the time of operating and subsequently. When the tumour attains any size, it may be palpated from the front, but in ordinary scirrhus of the head of the pancreas no tumour can be felt, except enlargement of the gall-bladder, which is generally present. Occasion- ally, however, there may be no marked enlargement of the gall-bladder until late in the case, either from the cystic or hepatic duct being gripped by the growth, or no enlargement throughout, from absence of the viscus, as in the case reported by Stewart, or its almost complete obliteration from previous inflammatory changes. Bronz- ing of the skin may come on if the adrenals are involved. Ascites or dropsy of the lower limbs may follow from pressure on the portal vein or inferior vena cava, or from secondary involvement of the liver, but apart from pres- sure, slight oedema of the feet is often an early sign. Death occurs from exhaustion, as a rule, within a few months, and is never very long delayed; in fact, cancer affecting the head of the pancreas is more rapidly fatal than when occurring in any other organ. The typical clinical picture of malignant disease of the pancreas may thus be drawn: A patient suffers for a time from indef- inite symptoms of digestive disturbance, then jaundice appears, coming gradually, but persistently increasing; the gall-bladder is usually distended and the liver is normal Neoplasms 517 or slightly enlarged at first and greatly enlarged later. A tumour may be found in the neighbourhood of the pancreas. Cachexia rapidly develops, and, in some rare cases, pain disturbs the patient's rest. There is soon a feeling of intense prostration and weakness. The faeces are massive and contain fat, the normal relation between the; 'neutral fat" and "combined fatty acids" is disturbed, and an undue proportion of undigested muscle fibre is Fig. 171. — Carcinoma of the head of the pancreas causing dilata- tion of the pancreatic duct, common bile-duct, and gall-bladder (St. Thomas' Hosp. Museum, 1414). present in the stools. The urine contains albumin fre- quently, and sugar and fat rarely. The whole clinical course is run, as a rule, within twelve months, and after the appearance of jaundice within from six to eight months. Differential Diagnosis. — In malignant disease of the pancreas the symptoms are not constant. The cases may be divided into three chief types: (i) Where the t-umour extends to the right and compresses or occludes the common bile-duct and the pancreatic ducts. (2) 5i8 The Pancreas: Its Surgery and Pathology Where it takes an upward and forward direction and, besides compressing the bile-duct, leads to pyloric steno- sis. In this case, to the typical symptoms are added those of dilated stomach. (3) Where the extension is back- wards, causing compression of the vena cava, and of the portal veins, thus leading to an early onset of ascites and later to oedema of the lower extremities. When the body and tail of the pancreas are involved the symptoms are atypical, and the development of a tumour with steady loss of strength and increasing anasmia are such as might be due to any malignant tumour outside the pancreas. In the differential diagnosis of cancer of the head of the pancreas we must consider common-duct cholelithiasis, interstitial pancreatitis, cancer of the common bile-duct, cancer of the liver, cancer of the pylorus, and chronic catarrh of the bile-ducts. Whenever, in a patient at or past middle age, jaundice comes on painlessly and becomes absolute, at the same time that the gall-bladder gradually enlarges so as to form a perceptible tumour, and the pa- tient rapidly loses flesh and strength, a diagnosis of cancer of the head of the pancreas will probably be correct. The diagnosis will be made more certain if there is an absence of tenderness below the right costal margin, associated with a tumour opposite to, or above, the umbilicus, hav- ing communicated pulsation and not moving with respira- tion. On distending the stomach with air or carbonic acid gas it will be found that the tumour, at the best rather indefinite, becomes hidden behind the resonant stomach, and that the distended gall-bladder becomes pushed to the right. If cholelithiasis has preceded the onset of cancer, the gall-bladder will not be enlarged, but the rapid deterioration of health and the presence of anasarca and ascites will, as a rule, leave no doubt of the nature of the disease. In common-duct cholelithiasis there is always a pre- liminary history of gall-stone attacks, though it may Neoplasms 5ig have been years x^reviously. The jaundice will have come on after pain and is probably never absolute, for some bile nearly always escapes past gall-stones in the common duct. The bile soon becomes infected, and ague -like seizures follow, with an irregular temperature, at times almost resembling pyasmia. In place of a distended gall- bladder a rigid right rectus will be felt, which often makes it difficult to examine the parts beneath. A tender spot will usually be found an inch above, and to the right of, the umbilicus, and the pain will be found to pass back- wards to the midscapular region or to a spot beneath the right shoulder-blade. Whereas cancer of the head of the pancreas is only a question of months, in cholelithi- asis it may be one of years. Pancreatic catarrh or chronic pancreatitis frequently accompanies gall-stones in the common duct and clears up after their removal, but it may persist after the cause has passed away. Whenever a tumour of the head of the pancreas is felt during a gall-stone operation, especially if before middle life, hope may always be felt that the disease may be simple and may clear up by the drainage of the ducts. A long history is in favour of the simple disease, as are the presence of adhesions, the history of painful attacks, and the presence of tenderness above the umbilicus. In chronic pancreatitis it is not uncommon to find enlarged glands in the free border of the lesser omentum, but they are discrete when the disease is simple and generally confluent in cancerous affections. The jaundice may be absolute, but, as a rule, it is not complete. Infective cholangitis and infection of the pancreatic ducts are commonly present, as shown by the temperature and by ague-like seizures. Although the loss of flesh is marked in chronic pancreatitis, it is less evident than in cancer of the head of the pancreas, and, until the disease has existed for a longer time than cancer gives its victim, there is no sign of anasarca or ascites or of enlarged abdo- 520 The Pancreas: Its Surgery and Pathology )j'Wtr5un5 minal veins. The gall-bladder is seldom distended, though this is not an absolute rule, as in several cases we have had the opportunity of observing it much enlarged. Between gall-stones in the common duct and chronic pancreatitis (which frequently coexist) it is often difficult to determine, but this is of no moment from a practical point of view, as surgical treatment is, as a rule, demanded in both conditions. Anaemia is much more marked in cancer of the head of the pancreas than in chronic inter- stitial pancrea- titis. Cancer of the common duct is rare and is usu- ally associated with gall-stones ; if the disease in- volves the pa- pilla the symp- toms are indis- t inguishable from those of cancer of the head of the pancreas; but if it be situated above the opening of the pancreatic duct, it will not inter- fere with the functions of the pancreas, and therefore the loss of flesh will not be so rapid. In two cases of cancer of the common duct in which we have made an analysis of the fseces they were found to be soft, white, and friable, like chalk or white Castile soap. No trace of stercobilin could be found in either instance. A quantitative estima- tion of the fats showed 90 per cent, of total fat, of which 32 per cent, was neutral fat and 58 per cent, combined fatty acid, in the one case, and total fat 8^ per cent., Fig. 172. — Diagram showing the sites of origin of malignant disease in the ducts and ampulla of Vater (modified from RoUeston). Neoplasms 521 consisting of 30 per cent, of neutral fat and 53 per cent, of combined fatty acid, in the other. Suppurative cholan- gitis is occasionally present, but this is not a constant event. The accompanying diagram shows the positions that growths may occupy in the neighbourhood of the papilla (Fig. 172). Cancer of the liver is distinguished by the jaundice being absent or much less intense, and by the enlarge- ment of the liver, with irregular nodules on its surface and edges. In simple catarrhal jaundice the symptoms are almost negative, except for the jaundice and loss of appetite ; and the way in which it yields to treatment shows the slighter nature of the ailment. In cancer of the pylorus the predominance of gastric symptoms, the dilatation of the stomach, with absence of free hydrochloric acid and the presence of blood in the vomit, usually enable a diagnosis to be made, but it should not be forgotten that cancer of the pylorus and of the head of the pancreas frequently coexist. In all these cases the urinary test affords most valuable help in diag- nosis. Treatment. — Medical treatment must be purely sympto- matic : morphine if needed for the relief of pain ; calcium chloride for the prevention of hsemorrhage; pankreon tablets or liquor pancreaticus for the digestion of food; and other remedies for symptoms as they arise. Surgical treatment is not very hopeful and has usually been under- taken under the idea that the cause of the jaundice might be a removable one, or that drainage of the bile-ducts might afford relief to the jaundice, but if the disease has involved the head of the pancreas it is hopeless however treated. Treatment may be radical or palliative. Ruggi, of Bologna, removed through the loin a cancer of the pan- creas weighing 23 ounces. It was probably growing from 522 The Pancreas: its Surgery and Pathology the tail of the gland. Complete recovery followed and the patient was well for three months, after which secondary disease developed and the patient died at the end of six months. Professor Ruggi himself has kindly furnished these details. Cades' was the second successful case, in 1895, a tumour of the tail of the pancreas of the size of a child's head being removed. Terrier, in 1892, removed a tumour weighing five pounds, but lost his patient. Of sixteen operations for removal of solid tumours of the pancreas, eight recovered, which, considering the diffi- culty of the operation and the depth of the organ to be operated on, is better than one would have expected. Successful pancreatectomies, it will be seen, are excep- tional and are feasible only where the growth is not involving the head of the gland; they, however, clearly demonstrate that a tumour of the body or of the tail of the pancreas may be removed with equal chance of recovery, and should the disease be primary, and no secondary growths or glandular involvement have occurred, great prolongation of life is quite possible. The palliative operations, cholecystotomy and cholecyst- enterostomy for the relief of jaundice in cancer of the head of the pancreas, have been performed by one of us in twenty-eight cases, all the patients being extremely ill at the time of operation. Many of these cases occurred when there was difficulty in making a diagnosis between cancer of the head of the pancreas and gall-stones in the common duct, or between cancer and interstitial pancreatitis, difficulties which have now been overcome to a large extent. Of the fifteen cases in which the gall- bladder was drained, eight recovered from the operation, the longest survival being eight months, but the average survival being about four months ; of the six cholecysten- terostomies, two recovered and the duration of life was only a few weeks. Even a simple exploratory operation in these cases is attended with danger, for out of six cases, Neoplasms 523 four only recovered from operation. Dr. Murphy, of Chicago, was kind enough to furnish a report of his col- lected statistics of cholecystenterostomy up to 1897. Of sixty-seven non-malignant cases there had only been three deaths, but of his twelve malignant cases ten died, giving a mortality of 83.3 per cent. Thus it will be seen that any palliative operation for the relief of cancer of the head of the pancreas associated with jaundice is useless, as, even if recovery occurs, life is not prolonged to any great extent. Sarcoma. — Primary sar- coma of the pancreas is undoubtedly rare, though secondary ' disease, espe- cially of the melanotic type, seems to be less un- common. There are exam- ples of sarcoma in several of the museums, photographs of some of which are shown in the illustrations. The first is of melanotic sar- coma from the Hunterian Museum (Fig. 173). It was taken from a girl, aged twenty years, and was sec- ondary to a melanotic growth in the eye, which was removed three years before her death. The next specimen is a very large spindle- celled sarcoma from University College Hospital Museum, No. 3200 (Fig. 174). The growth has completely de- stroyed the gland and has left no trace of gland tissue. The next specimen (Fig. 175), No. 2836 A, in the Royal College of Surgeons Museum, was removed from the tail Fig. 173. — Melanotic sarcoma in the pancreas (Royal Coll. of Surg. Museum, 2836). 524 The Pancreas: Its Surgery and Pathology of the pancreas, but the child succumbed shortly after the operation. Operation for sarcoma of the pancreas is uncommon, Fig. 174. — Spindle-celled sarcoma of the pancreas (Univ. Coll. Mu- seum, 3200). Fig. 175. — Sarcoma of the tail of the pancreas (Roy. Coll. Surg. Mu- seum, 2836 A). though the few cases operated on prove that if the tumour be in the tail of the pancreas the case is amenable to siu:- Neoplasms 525 gical treatment. The abdomen was explored by one of us in a case of the kind, but the disease was found to be too extensive for removal. Kronlein, in 1894, removed a tumour of the size of the fist, but the patient died seven days later, A tumour which was successfully removed by Briggs proved to be sarcomatous degeneration of an echinococcus cyst. Adenoma. — Adenoma of the pancreas is extremely rare. Instances have been recorded by Thierf elder, Biondi, Cesaris-Demel, Neve, and Nicholls. Several other cases described as adenomata were probably exam- ples of malignant growths, and cannot therefore be in- cluded in the list. It is possible that some of the cases described as cystadenomata originated in simple adenoma, but, as this is uncertain, they are best for the present considered under the heading of cysts. Diagnosis and Treatment. — Adenomata of the pancreas present no characteristic symptoms by which they can be distinguished during life, but should such a tumoiu" be met with in the pancreas during the course of an explora- tory operation an attempt should be made to extirpate or enucleate it. Tuberculous Disease of the Pancreas. — Tuberculosis of the pancreas is usually considered to be rare, and as a pirmary lesion it undoubtedly is, but Kudrewetzki found that in a series of one hundred and twenty-eight cases of tuberculosis the pancreas was affected fifteen times, five times as part of an acute miliary tuberculosis, seven times as part of a chronic tuberculosis, and twice from tuberculo- sis in its neighbourhood. Children furnished the greater number, for he found 44.44 per cent, of pancreatic tuber- culosis in tuberculous children, but only 9 per cent, in adults. He emphasises the fact that tuberculosis of the pancreas occurs only as a secondary condition in connec- tion with tuberculosis of other organs. In Hale White's series of one hundred and forty-two post-mortems in which 526 The Pancreas: Its Surgery and Pathology the pancreas appeared to be diseased or injured there were four examples of tubercle. Three of the patients suffered from general tuberculosis and one from tubercu- lous peritonitis. He states that tubercle of the pancreas was found in considerably less than i per cent, of all cases of tuberculosis, and he therefore considers that it is a rare disease. Loheac would explain the relative infrequency of tuberculous disease by virtue of the peculiar pancreatic secretion, which he thinks to be protective against this form of infection. Tuberculosis of the pancreas may occur as numerous small, granular, infiltrating tubercles or as large caseous masses. The latter probably originate in most cases from lymphatic gland buried in the substance of the gland. Treatment. — A few cases of successful removal of tuber- culous masses from the pancreas have been recorded. Thus Sendler opened the abdomen of a thin woman, who had a movable tumour above the umbilicus, and found behind the stomach a hard mass the size of a walnut, which he extirpated. This proved to be a tuberculous lymph-nodule of the pancreas. The patient recovered, Kudrewetzki reports a case in which a caseous tubercu- lous mass burst into the stomach of a man of forty-two. The patient was operated on and a number of caseating glands found in the lesser omentum, but he became so collapsed on manipulating a tumour occupying the site of the head of the pancreas that nothing further was done. The wound healed and he was discharged. Symptoms and Diagnosis. — At present it is impossible to diagnose tuberculosis of the pancreas during life, and since it is practically always secondary to disease else- where, its recognition is of no practical importance unless there are pressure symptoms. Syphilis. — Syphilis, like tubercle, may occur in the pancreas in two forms, interstitial and gummatous, which may, however, coexist. Of the t\^o, the former is Neoplasms 527 the more common. Both may result from either acquired or congenital disease. In acquired syphilis disease of the pancreas is regarded as uncommon, but this may be due to there being no characteristic symptoms during life, in most cases, point- ing to the pancreas. There is no reason why the pancreas should not be involved with other organs in visceral syphilis, which, being recognised and suitably treated, recover, for a gumma of the pancreas is just as likely to yield to treatment as gumma of the liver. Post-mortem records with regard to syphilitic affections of the pancreas are but meagre. Peterson in eighty-eight cases that had suffered from tertiary syphilis found only one in which the pancreas was affected. Occasionally syphilitic disease of the pancreas may give rise to secondary symptoms ; thus, H. Betham Robin- son has given the details of a case in which there was obstructive jaundice due to a gummatous infiltration involving the head of the gland, in which cholecystoco- lostomy was successfully performed. In congenital syphilis the pancreas is less commonly affected than the spleen or liver, but syphilitic deposits have been noticed as early as the fifth month. As a rule, the disease occurs as an interstitial inflammation starting from the vessels, but occasionally it occurs in the form of large or small gummatous masses. The interstitial overgrowth spares the islands of Langerhans, so that dia- betes does not occur. Treatment. — The surgical treatment of syphilis of the pancreas is limited to such cases as that of Betham Robin- son above referred to, and reliance must be placed upon general medical means for dealing with this disease. Literature Baudael: Dissert., Freiburg, 1885. Biach: Wien. med. Wochen., 1883. Biondi: Ref. Med., 1896; Clin. Chir., 4, 1896. 528 The Pancreas: Its Surgery and Pathology Birch-Hirschfeld : Arch. d. Heilk., xvi. Briggs: St. Louis Med. and Chir. Journ., 1890, p. 154. Brunton, Lauder: Brit. Med. Journ., June 11, 1904, p. 1353. Cesaris-Demel : Arch, per le Soc. m6d., 1895, xix. Codivilla: Rendoconto statist d. Sezone Chih. dell' ospedale de mola, 1898. Dovan: Brit. Med. Journ., Oct. 22, 1904, p. 1073. Dunning: Amer. Journ. of Obstetr., Jan., 1905, p. 161. Eppinger: Prager Vierteljahresschr. f. Heilk., cxiv. Fawcett: Lancet, May 7, 1904. Franke: Arch. f. klin. Chir., xliv, 1901. Hancock: Journ. Araer. Med. Associat., Jan. 27, 1906. Healey: Journ. of Roy. Army. Med. Corps, iv, 3, 362. Kakels: Amer. Journ. of Med. Sci., 1902, cxxiii, 471. Kronlein: Weiner med. Wochen., 1895, S. 1318; Beitrage z. klin. Chir., 1895, _S. 663. Kudrewetzki: Prag. Zeit. f. Heilk., 1892. Lawrence: London Med. Gazette, 1845, xxxvi, 951. Loh6ac: Quoted by Roswell Park, Amer. Med., 1903, p. 949. Malcolm: Lancet, March, 1902. Malthe: Zeitschr. f. Prof. Herberg, Kristiania, 1895. Martin: Trans. Path. Soc. of Lond., 1900. Monprofit: Gaz. m^d. de Paris, March 12, 1904. Neve: Indian Med. Rec, 1892, p. 208; Lancet, 1901, p. 9. Nichelsohn: Dissert., Wurzburg, 1894. NichoUs: Journ. of Med. Research, viii, 2, 385. Nocard and Friedberger: "Vet. Path.," Friedberger and Frohner, tr. Hayes, 1905. Opie: "Diseases of the Pancreas," 1903. Oser: Nothnagel's "Encyclop. of Pract. Med." Park, Roswell: Amer. Med., v, 24, 949. Paulicki: Allgemein medicin Centralzeitung, 1868, No. 90. Peterson: Monatshft. f. prakt. Derm., 1891. Phillips: Lancet, Feb. 16, 1907, p. 418. Raven: Brit. Med. Journ., Oct. 15, 1904, p. loii. Rhode: Inaug. Dissert., Keil, 1890. Robinson, Betham: Brit. Med. Journ., 1900, p. 1004. Robson, Mayo: Clin. Soc. Trans., 1889, xxiii; Lancet, July 2, 1900; Lancet, March 19, 26, and April 2, 1904. Routier: Rev. de Chir., 1892. Segre: Ann. Univers. della. med. e chir., cclxxxiii. Sendler: Deut. Zeit. f. Chir., 1896, xliv. Schlagenhauf er : Arch. f. Derm. u. Syph., 1895. Schleisinger: Virch. Arch., cliv, p. 501. Soyka: Prag. med. Wochenschr. , Oct., 1876. Stewart: Brit. Med. Journ., April 16, 1904, p. 885. Terrier: Nimier: Rev. de Chir., 1893, 1894. Trendelenburg: Deut. Zeit. f. Chir., 1886. Triconi: Cent. f. Chir., 1898. White, Hale: Guy's Hosp. Rep., 1897, liv, 26. INDEX OF AUTHORS AbELMANN, 208, 209, 210, 221, 223, 267 Abia, 165 Abram, 306, 308, 309 Acard, 303, 308 Ackeron, 241, 267, 329, 344 Adami, 95 Adams, 471 Addison, 256, 286, 302 Alay, 500 Aldehoff, 271, 308 Alexander, 339 AUbtitt, 317 Allen, 359, 485, 486 Alt, 194 Ancelet, 164, 486 Anders, 471 Andrews, 471 Anschiiltz, 303, 308 Ansperger, 471 Arnozan, 164 Arris, 243, 268 Aschoff, 296 Auche, 370, 383 Babler, 410 Baillie, 140, 188, 361, 362, 383, 4.7^, 4.86 Baillie, 14c, ^„„, j„^, j„_, j„ 473. 486 Bainbridge, 98, 122, 124, 307, 308 Baines, 410 Baldi, 209, 237, 267 Baldwin, 240, 267 '" '' "" 190, 191, 192 Balser, 141, 188, ig 194, 195, 205, 362 Bancroft, 124 Bandel, 527. Barcroft, 98 Barling, 410, 471 Barnard, 471 Battersby, 332, 344, 471 Baudach, 185 Baudael, 188 Bayliss, 96, loi, 104, 105, 124 Becker, 355, 357, 359 Becourt, 47, 510 Beddard, 307, 308 34 Benda, 194 Bender, 194 Bernard, 96, 112, 118, 119, 124, 208, 225 Bernheim, 171 Bertram, 355, 359 Biach, 527 Bidder, 108, 124 Bierry, 122 Biondi, 187, 188, 234, 325, 344, 525, 527 Birch-Hirschfeld, 149, 161, 162, 188, 374, 383. 528 Birmingham, 28, 40 Bloch, 279, 308 Blum, 302, 308 Blume, 197, 205 Blumenthal, 242, 267, 275, 308 Boas, 265, 267, 501, 510 Boeckmann, 471 Bohm, 76, 184, 188 Boldt, 181, 188 Borchardt, 355, 357, 359 Borrell, 186, 188 Bosanquet, 127, 164, 188, 273, 279, 280, 281, 287, 288, 291, 297, 308, 413, 425, 471 Bouchardat, 269, 278, 308 Bousfield, 380 Bowditch, 264, 268 Bozeman, 176, 188 Bozenaum, 510 Brachet, 44 Brackel, 510 Bramann, 355, 357, 359 Bramwell, 131, 132, 188, 237, 268 Braune, 29, 31 Brennecke, 410 Briggs, 185, 188, 525, 528 Bright, 277, 308, 323, 344. Broadbent, 380 Brodie, 123, 124 Brown, 351, 359, 410 Bruce, 287 Briicke, 119, 124 Brunn, 44 Brunton, 528 529 53° Index of Authors Bryant, 410 Bull, 176, 188, 325, 344, 510 Bunger, 57, 63 Bunting, 393, 410 Cacchini, 55 Cades, 522 Cajal, R. y. 40 Caldwell, 357, 359 Cammidge, 243, 268, 334, 342, 344, 395 Capparelli, 271, 308, 325, 332, 344, 481, 486 Carnell, 355, 359 Carnot, 142, 151, 153, 163, 164, 188, 423, 471 Cart ledge, 510 Castle, 266, 268 Cathcart,_35i, 359 Cavazzani, 209, 221, 268 Cayley, 149, 188 Cesaris-Demel, 187, 525, 528 Cesaris-Deruch, 188 Chalmers, 378, 471 Chamberland, 103 Chari, 325, 344 Charles, 307, 308 Chauffard, 299, 302, 303, 308 Chiani, 188 Chiari, 128, 191, 194, 205 Chopart, 277, 308 Choronschizky, 44 Churton, 282, 317, 510 Claessen, 56 Clark, 264, 268 Classen, 410 Claus, 276 Clayton-Greene, 106, 124 Cleland, 259, 260, 261, 268 Glutton, 510 Codivilla, 528 Cohnheim, 116, 124, 269, 275, 276, 294, 307. 309 Coivisart, 112, 124 Cooke, 264, 268, 410 Coombs, 351, 359 Cooper, 347, 359 Cotter, 471 Courvoisier, 181, 188 Cowley, 277, 309, 325, 344, 473, 486 Crofton, 274, 307, 309 Cruveilhier, 510 Cubert, 511 Cumston, 505, 510 Cunningham, 28 Dale, 92, 93, 95 Dalziel, 471, 483, 486 Dargan, 357, 359 Davian, 471 Davidoff, 76 Davidson, 477 De Renzi, 221, 268 De Witt, 26, 80, 81, 82, 83, 84, 85, 87, 88, 90, 91, 93, 95, 293, 309 Deanesly, 410 Deaver, 134, 146, 410, 471 Delageniere, 475, 486 Delbet, 471 Delezenne, 113,124, 265, 353,359, 501. 510 Demme, 210, 268, 325, 344 Desjardins, 152, 156, 188, 471, 472 Dettmer, 196, 201, 205 Deucher, 211, 268 Diamare, 91, 95, 292, 294, 309 Dieckhoff, 149, 150, 186, 188, 195, 205, 282, 284, 286, 510 Dixon, 510 Dobrzycki, 56 Dogiel,_ 84, 95 Dominicus, 236, 237 Dovan, 528 Drasche, 410 Drozda, 163, 188 du Pasquier, 511 Dunning, 177, 188, 510, 528 Duponchel, 510 Durante, 510 Durno, 410 Dutil, 184, 188 Dutourier, 303, 309 Duval, 157, 189, 471 Earl, 410 Eberle, 118, 124 Ebner, 79, 8i, 83, 85, 95 Ecker, 47, 63 Edgecombe, 373, 383 Edie, 306, 309 Edsall, 233, 268 Ehler, 471 Ehrlich, 275 Eichhorst, 483, 486 Elliotson, 264, 268, 277, Embden, 276 Engel, 49, 63, 510 Eppmger, 513, 528 Ernst, 195 Estes, 471 Eulenberg, 486 309 Index of Authors 53^ Evans, 471 Ewald, 113, 339, 344 Fawcett, 204, 205, 528 Fehling, 267, 283 Felix, 44 P^enwick, 66, 67 Filger, 188 Fischer, 115 Fisher, 510 Fison, 190, 191, 394, 396, 410 Fitz, 126, 141, 149, 188, 195, 196, 205, 287, 324, 344, 359, 362, 383, 386, 399, 410, 471, 510 Fleig, 105 Fleiner, 164, 188, 291, 309 Fles, 222, 277, 309, 323, 328, 344 Fletcher, 410 Flexner, 141, 142, 145, 150, 151, 157, 158, 159, 188, 197, 201, 202, 205 Flint, 72, 84, 85, 86, 95 Fraenkel, 195, 202, 205 Franke, 528 Freidenthal, 143 Frerichs, 277, 279, 280, 284, 309 Frey, 124 Friedberger, 471, 512, 528 Friederich, 410 Friedrich, 135, 188 Fripp, 410 Frison, 410 Frohner, 471, 528 Fromme, 339, 344 Fuchs, 410 Gachet, 113 Gaeia, 486 Gale, 243, 268 Galea, 473 Galippe, 171, 172 Gallaudet, 65 Gamgee, 123 Gamges, 471 Garre, 345 Gaston, 463 Gaule, 77 Gaylord, 296 Gegenbaur, 50, 63 Generisch, 47, 63 Gentes, 297, 309 Gerhardi, 234, 268 Gerhardt, 188 Giacco, 275 Gilbert, 510 Giordano, 471 1S8, 332, 474, Glaessuer, 108, 109, 124 Glinski, 46, 49, 50, 53, 54, 63 Goldmann, 349, 359 Golgi, 74, 81, 84 Goodall, 181, 188 Goodman, 325, 344 Goppert, 44 Gordon, 226, 268 Gosset, 471 Gotte, 44 Gottlieb, 199, 205 Gould, 325, 344, 483, 486 Gourand, 510 Gow, 79, 80, 95 Graaf, 486 Graham, 510 Greisclius, 411 Greville, 45 8 Groeningen, 349, 359 Griiber, 160 Guidiceandra, 172 486 Guinard, 368, 383 Gulcke, 200, 205 Guleke, 143, 148, 151, 188, 287, 309 Giinther, 19, 27 Giissenbauer, 500, 510 Guyose, 458 Hadra, 351, 359 Hagen, 510 Hagenbach, 175, 188, 510 Hahn, 355, 357, 359, 411 Haldane, 263, 268 Halley, 411 Halliburton, 123, 124 Halstead, 144, 366, 383, 3S9, 400, 411 Ham, 259, 260, 261, 268 Hamburger, 44 Hammar, 44 Hammarschlag, 277, 309 Hammarsten, 97, 124, 125 Hancock, 528 Hanot, 302, 303, 309 Hansemann, 51, 88, 91, 95, 129, 130, 131, 162, 188, 194, 205, 279, 280, 281, 286, 299, 359 Hardin, 471 Harley, 271, 309, 325, 329, 463 Harris, 79, 80, 95 Hartmann, 177, 188, 511 Hartsen, 277, 309 Harvey, 411 Hawkins, 149, 188, 402 Hayes, 471, 528 532 Index of Authors Haynes, 511 Healey, 185, 188, 528 Heaton, 411, 511 Hedon, 236, 271, 309 Heidenhain, 77, 92, 98, 100, 113, 199, 205 Heinricus, 173, 174, 188 Heller, 194, 205 Helly, 44, 52, 54, 57, 63, 144, 257, 319. 365- 383 Hennige, 268 Hennigs, 188 Hennings, 135 Herberg, 528 Hermann, 125 Herter, 108, 125, 207, 224, 233, 240, 267, 268, 302, 309, 500, 511 Hertz, 55 Herxheimer, 300, 309 Herzen, 113 Herzog, 295, 297, 309 Hess, 142, 143, 188 Hewlett, 268 Hildebrand, 196, 197, 200, 201, 205 Hill, 125 Hillier, 181, 188 Hirschfeld, 221, 325, 344 Hlava, 142, 188, 196, 20T, 205 Hodgkin, 188 Hofmeister, 112 Hogarth, 411 Hollander, 511 Holzmann, 332, 344, 480, 486 Hoppe, 176, 188, 511 Hoppe-Seyler, 164, 188, 291, 309 Horrocks, 511 Hufeland, 344 Hyrtl, 49, 63 Jaboulay, 471 Jackson, 195 Jacob, 370, 383 Jacobson, 176, 189, 411 Jaksch, 501, 511 Jankelowitz, 44 Jarotzky, 91, 92, 95 Jastrowitz, 241, 268 Jaun, 349, 352, 359 Jeanselme, 303, 309 Jeffrey, 411 Jephson, 355, 357, 359 Johnston, 474, 475, 486 Jones, 411 Joslin, 295, 310 Jung, 196, 205 Kakels, 185, 189, 528 Karewski, 351, 355, 359 Kasahara, 87, 95, 162, 1S9 Katz, 197, 201, 205, 210, 234, 268 Kauffmann, 277, 309 Kausel, 271 Keenan, 189 Kehr, 471 Keiserling, 387 Keith, 42, 43, 45 Kellock, 511 Kelly, 339, 344 Kempe, 403, 411 Kennan, 149, 411 Keyser, 411 Kilgow, 411 Kindt, 355, 356, 360 Kinnicutt, 172, 1S9, 479, 482, 486 Klebs, 56, 175, 189, 309 Kleburg, 357, 359, 360 Klippel, 165, 189, 471 Klob, so, 51, 53, 63, 411, sii Klobin, 49 Kocher, 66, 466 Kolliker, 95 KoUmann, 41 Korte, 69, 108, 149, 150, 151, 178, 179, 189, 195, 197, 202, 205,355, 359, 360, 400, 411, 471, 481, 486, 500. 511 Kossel, 125 Krehl, 268 Kronlein, 70, 185, 525, 528 Kudrewetzki, 525, 526, 528 Kiihn, 184, 189 Ktihnast, 511 Kiihne, 77, 83, 95, 113, 115, 123, 125, 482 Kiilenkampff, 351, 360 Kuliabko, 294, 309 Kiilz, 242, 268 Kummell, 482 Kuntzmann, 323, 325, 344 Klister, 89, 95, 320, 329, 344, 351, 360 Kutscher, 116 Kiittner, 357, 360 Kyber, 135 Laborderie, 357, 359 Lacher, 56 Laguesse, 45, 85, 87, 88, 89, 93, 95, 292, 309 Lancereaux, 144, 150, 186, 188, 189, 269, 277, 278, 309, 480, 486 Langendorff, 152, 271 Index of Authors 533 92, 195, 196, 201, 202, 205, 240, 174, 286, 289, 290, 292, 293, 294, 95, 296, 297, 299, 300, 303, 366, 374, 396, 417, 426, 447, 449, 515, 527 . Lannois, 165, 189, 291, 309 Lapasset, 370, 383 Lawrence, 528 Le Nobel, 241, 268, 325, 329, 344 Lea, 77, 84, 95, 482 Lefas, 165, 189, 424, 471 Leichtenstern, 332, 344, 479, 486 Leith, 349, 352, 360 Lejars, 471 Leraione, 165, 189, 291, 309, 370, 383 Lenne, 309 Lens, 63 Leonhard, 196 Lepage, 100, 102, 125 Lupine, 165, 241, 268, 274, 275, 297, 309 LetuUe, 52, 53, 59, 181, 189 Leusden, 411 Lewaschew, 84, 88, 92, 93, 95 Lewis, 51, S3 Lichtheim, 329, 344 Lilienthal, 411 Littlewood, 351, 360 Lloyd, 68, 178, 189, 351, 360, 511 Loevenhart, 266, 268 Loewi, 116 Loheac, 526, 528 Lophius, 87 Lorand, 274, 309 Lorrier, 59 Lowe, 471 Lubarsch, 185 Ludolph, 332, 344, 511 Lund, 411 Lynn, 511 Macaigni, 285 Macallum, 78, 95 Mackenzie, 379 Mahomed, 471 Malcolm, 506, 511, 528 Mallory, 72, 86, 95 Malthe, 528 Maly, 125 Mandel, 124 Mankowski, 90, 93, 95 Mann, 355, 360 .04, Manprofit, 511 Marcy, 471 Margain, 303, 309 Markuse, 271 Marseron, 511 Martin, 511, 528 Martinotti, 269 Matani, 479, 486 Mayo, 163, 189, 362, 411, 471 McPhedran, 489, 511 McReynolds, 5 1 1 Meckel, 51, 52 Megnin, 415, 471 Melzer, 112, 125 Mering, 126, 189, 269, 270, 309 Mett,^266, 267 Michelsohn, 185, 189 Mikulicz, 66, 403, 465 Milisch, 197, 199, 205 Milroy, 475 Minkowski, 126, 189, 208, 23 269, 270, 271, 272, 273, 274, 30 . 306, 309 Minnich, 332, 344, 479, 480, 481, 486 Minot, 81 Monprofit, 177, 189, 528 Moore, 261, 306, 309, 339, 344, 411, 471 Morache, 164 Morgagni, 473, 486 Morian, 411 Morton, 511 Mosetig-Moorhof, 471 Motta, 325 Moxon, 348, 360 Moynihan, 160, 179, 189, 351, 387, 411, 471, 475, 483, 486, 509. 511 Moyse, 47 Mraczek, 162 Miiller, 35, 40, 75, 120, 134, 146, 162, 211, 222, 223, 225, 268, 410, 471 Munk, 119, 143, 277, 309 Munster, 411 Murphy, 444, 468, 523 Murray, 511 Muspratt, 403, 411 Myles, 471 Narath, 511 Nash, 288, 308, 309, 351, 359 Nauwerck, 51, 63 Neuberg, 97, 125, 251, 268 Neumann, 51, 53, 63, 511 Neve, 187, 189, 525, 528 534 Index of Authors Nichelsohn, 528 NichoHs, 51, 63, 187, 189, 525, 528 Niemann, 355 Nimier, 172, 189, 528 Nini, 355, 356, 357, 360 Nocard, 415, 471, 512, 528 Nocolaider, 77 Noorden, 238, 301, 304, 309 Norris, 411 Nothnagel, 189, 268, 344, 387, 411, 486, 511, 528 Nussbaum, 463 OcHSNER, 511 Oddi, 38 Oidtmann, 97, 125 Olivier, 186 Olt, 205 Opie, 37, 38, 45, 50, 51, 5 55, 59, 60, 61, 63, 78, 8 92, 95, 131, 135, 136, 145. 147. 153. 157. 161, 165, 166, 189, 197, 198, 201, 204, 205, 263, 266, 280, 287, 290, 292, 295, 300. 303, 309- 335> 336, 366,367,383,386,389, 413, 419, 424, 425, 471 Orth, 149, 186, 189 Oser, 139, 150, 184, 186, 201, 205, 210, 211, 268, 282, 285, 287, 324, 329, 471, 474, 486, 511, 528 Osier, 411 Otis, 355, 357, 359, 360, Owen, 20, 21, 22, 23, 24, 27, 471 Pagenstecher, 467 Paltauf, 511 Panarol, 473, 486 Park, 486, 511, 513, 528 Pauchet, 411 Paulicki, 528 Pavy, 276, 309 Pawlik, 511 Pawlow, 96, 98, 99, 100, loi, 104, 106, 108, 109, no, 113, 120, 121, 123, 125, 152, 189, 266, 267, 268 Pearce, 89, 95, 142, 188 Pearson, 63 Peiser, 411 Pende, 170, 172, 189 Pensa, 80, 81, 95 Pepper, 511 Percival, 140, 189, 362, 383, 411 2, 53 , 54 7. 9c . 91 137- 144 162, 164 199, 200 268, 279 297- 298 342, 344 393- 411 ,528 187, 197 279- 281 344. 387 Pereira-Guimaraes, 359, 360 Perle, 411 Perowoznikoff, 120, 125 Perrin, 471 Peters, 511 Peterson, 527, 528 Phillips, 486, 494, 512, 528 Phulpin, 511 Pisenti, 233, 268 Pitchford, 511 Pitt, 176, 189, 411 Plimmer, 122, 123, 125 Pochon, 113 Ponfick, 195, 205 Popielski, 100, 123, 125 Portal, 140, 362, 383, 4 Portel, 189 Porter, 403 Poucet, 511 Pressel, 348, 360 Prince, 359, 360 Przewoski, 511 Pye-Smith, 511 II QuAiN, 40, 45 Quenu, 157, i^ 471 Rachford, 27, 119, 120, 125 Radziejewski, 120 ' Randall, 351, 352, 354, 360 Ranschoff, 511 Ravant, 299, 308 Raven, 471, 528 Raynes, 471 Recklinghausen, 277, 302, 309 Reed, 309 Reeves, 325, 344 Reichert, 63 Renant, 89 Rennie, 87, 95, 294, 309 Rentoul, 132, 189 Renzi, 234, 268 Reynolds, 411 Rhode, 528 Riboli, 411 Richards, 302, 309 Richardson, 511 Riedel, 154, 155, 189, 471, 511 Rispal, 500 Roaf, 339, 344 Robinson, 163, 189, 511, 527, 528 Robson, 68, 189, 268, 344, 351, 362, 383, 396, 411, 468, 472, 486, 509- 511 Roddick, 411 Index of Authors 535 Rodocanachi, 472 Rokitansky, 135, 189 Rolleston, 51, 63, 161, 358, 360, 520 Rose, 56, 360 Rosenberg, 209, 268 Rosenheim, 241, 268 Ross, 351 Roth, 51 Routier, 528 Rouxj 177, 189, 511 Ruggi, 521, 522 Sahli, 338, 339, 342, 343. 344 Salkowski, 241, 242, 268 Salomon, 220, 268 Salzer, 511 Sandmeyer, 151, 189, 209, 221, 268, 272, 309 Sanitas, 355, 360 Santorini, 44, 52, 55, 60, 61, 62, 145, 152, 156, 157. 171. 173.367. 419, 420, 421, 470, 484, 485, 520 Santos, 47, 48, 63, 203, 346, 415 Sauerbeck, 93, 95, 271, 299, 309 Saunby, 135, 139 Saviotti, 74 Sawyer, 196, 204, 205 Scatterby, 317 Schafer, 45, 125, 292, 309 Schaffer, 74. 95 Schafter, 40 Schepowalnikow, 113 Schieffer, 27 Schiff, 113 Schlagenhaufer, 234, 268, 528 Schlesinger, 161, 162, 189, 528 Schmidt, 106, 107, 108, 109, 124, 125, 205, 212, 222, 268, 293, 297, 300, 309, 330, 344 Schmieden, 472 Schroeder, 263, 268 Schulze, 91, 93, 95, 152, 189, 293, 309 Scott, 93, 94, 95, 192, 193, 240, 268, 300 Seegen, 279, 310 Segond, 472 Segv6, 185, 189, 513, 528 Seitz, 149, 189, 19s, 205, 287 Selberg, 411 Sandler, 163, 189, 472, 526, 528 Senn, 151, 163, 172, 189, 207, 351, 360, 470, 472, 488, 511 Shattock, 475, 486 Shaw, 511 Shea, 411 Sheen, 351, 360, 511 Shirmer, 47, 50, 62, 63 vShupmann, 479 Silver, 277, 310, 325, 344 Simmonds, 355, 360 Simoni, 383 Simonin, 369 Slavsky, 355, 357, 360 Smith, 383, 411, 467 Soxhlet, 211 Soyka, 528 Spalteholz, 72, 84 Ssobolew, 93, 95, 152, 189, 240, 293, 297, 300, 310 Stangl, 297, 298, 310 Starling, 96, 98, loi, 102, 103, 104, 105, 107, III, 113, 117, 124, 125, 207, 307, 310 Statkewitsch, 95 Stefanani, 234, 268 Steinhaus, 78, 95 Sterling, 268 Steven, 195, 205 Stewart, 411, 516, 528 Stibler, 315 Stieda, 511 Stoerk, 347, 360 Stohr, 45 Stokes, 212 Stolnikow, 268 Stoss, 45 Strasburger, 229, 268 Strauss, 270, 310 Symington, 40, 47, 63 Tait, 288 Telling, 50, 51 Terrier, 472, 522, 528 Testut, 34. 35. 37. 39. 40, 57. 58. 66, 365, 383, 491 Thierf elder, 187, 525 Thiroloix, 171, 173, 174. 189, 236, 306, 310, 511 Thomas, 411 Thompson, 27, 359, 360 Thomson, 132 Tiedemann, 47 Tieken, 47 Tilger, 176, 179, 511 Toye, 411 Travers, 347, 360 Trendelenburg, 528 Trevor, 145, 189, 300, 310, 395 Triconi, 528 Truhart, 337, 344> 359 Tuckett, 274, 310 Tuffier, 472 536 Index of Authors Tulpius, 140, 189, 264, 268, 362, 383 Turner, 51, 53, 63 Umber, 275, 310 Underbill, 276, 310 Van den Valden, 339 Vasilieff, 123 Vater, 59, 62, 70, 144, 145, 152, 154, 15s, 156, 174, 315, 319, 365, 366, 367, 379, 385, 389, 393, 394, 395, 418, 419, 422, 481, 483, 485, 486, 489, 520 Vernon, 116, 125 Vidal, 47, 63 Villar, 471, 511 Villi^re, 360 Villiers, 349 Virchow, 63, 95, 175, 188, 189, 268, 309, 310, 498, 511, 528 Vogel, 242, 268, Volhard, 117 Volker, 45 Von Ackeron, 241, 267, 329, 344 Von Bramann, 355, 359 Von Ebner, 79, 81, 83, 85, 95 Von Frey, 124 Von Jaksch, 501, 511 Von Mering, 126, 189, 269, 270, 309 Von Mikulicz, 66, 403, 465 Von Noorden, 238, 301, 304, 309 Von Recklinghausen, 277, 302, 309 Vulpian, 164 Wagner, 50, 63 'Wagstaff, 349, 352, 360 Walker, 226, 268 Wallenfang, 339, 344 Walther, 109, 120, 123, 472 Wandesleben, 358, 360 Watkins, 511 Webber, 411 Weichselbaum, 50, 51, 53, 63, 297, 298, 310 Weinland, 121, 122, 125 Weintraud, 211, 221, 268, 271, 310 Welch, 195, 196 Wertheimer, 100, loi, 102, 105, 125 White, 126, 150, 165, 185, 186, 188, 189, 348, 349, 360, 413, 472, 487, 511. 525. 528 Widal, 160 Wiener, 472 Wilks, 348, 360 Willcox, 262, 268, 340, 344 Williams, 194, 197, 205 Williamson, 131, 189, 279, 280, 281, 284, 286, 300, 310 Willie, 305, 338, 344 Windle, 279, 280, 310 Winkler, 197, 201, 205 Winniwarter, 463 Winslow, 47, 178, 351, 353, 358, 391. 442 Wirsung, 43, 44, 53, 57, 58, 60, 61, 62, 72, 108, 141, 144, 152, 154, 156, 157. 158, 159. 170. 171. 173. 181, 182, 194, 210, 211, 235, 271, 319.367.379.395.396,409,418, 419, 420, 421, 470, 474, 482, 483, 484, 485, 488, 489, 498, 520 Wlassow, 45 Wohlgemuth, 109, 113, 114, 118, 123, 125, 501, 511 Woolsey, 232, 268, 403, 411 Wright, 52, 53, 63, 140, 230, 295, 310 Wyss, 57, 511 Young, 237, 268, 411 Zawadsky, 108, 125 Zawarykin, 119, 125 Zeehuisen, 501 Zeigler, 63, 149 Zeller, 472 Zenker, 51, 54, 63 Ziehl, 210, 268, 325, 344 Ziemssen, 410 Zimmermann, 45 Zukowski, 511 Zung, 93, 95 INDEX Abdominal salivary gland, 71 Abscess of pancreas and suppura- tive pancreatitis, 150 pyemic, of pancreas, 150 Absence of islands of Langerhans in diabetes, 297 Accessory duct, 38 pancreas, 49 pancreatic duct, 44 Acetone bodies in urine in dis- eases of pancreas, 237 Acne pancreatica, 175 Adenoma, fibro-, of pancreas, 187 of pancreas, 187, 525 diagnosis and treatment, 525 Alcohol, influence of, in produc- tion of pancreatitis, 160 relation of, to chronic pancrea- titis, 424 Alcoholism in diabetes, 301 Alimentary glycosuria after par- tial extirpation of pancreas, 271 from disturbances of pan- creas, 304 in diagnosis of diseases of pancreas, 337 relation of pancreas to, 305 Alveoli of pancreas, 76 Amphopeptone, 115 Ampulla of Vater, 37, 39 mode of formation, 59 Amyloid degeneration of pan- creas, 135 Amylopsin, 1 1 1 Anatomical anomalies of pan- creas, 46 variations of pancreatic ducts, 56 Anatomy, comparative, of pan- creas, 17 of pancreas, 28 surgical, of pancreas, 64 Anomalies, anatomical, of pan- creas, 46 Anthropoidea, pancreas of, 26 Antipeptone, 115 Apoplectic cysts of pancreas, 175, 506 Apoplexy, pancreatic, 138 Appetite, alterations of, in dis- eases of pancreas, 320 Aquatic mammals, pancreas of, 24 Areas of Langerhans, 78 Arteries of pancreas, 32, 33 Arteriosclerosis in diabetes, 301 Atrophy of islands of Langerhans in diabetes, 297 of pancreas, 129 cachectic, 130 in diabetes, 129, 280 secondary, 133 Auto-intoxication theory of dia- betes, 273 Azotorrhea in diseases of pancreas, 328 . . Azoturia in diseases of pancreas, 236 Bacteria as cause of chronic pan- creatitis, 151 Bauchspeicheldruse, 71 Bile in urine in diseases of pan- creas, 235 Bile-duct, common, and head of pancreas, operation for explor- ing, 458 Bile-pigments, relation of urobi- linuria to, in diseases of pan- creas, 235 Biliary passages, diseases of, asso- ciation of diseases of pancreas with, 144 Birds, pancreas of, 22 Black jaundice in diseases of pan- creas, 320 Blood changes from diseases of pancreas, 230 in diabetes, 274 in feces in diseases of pancreas, 332 537 538 Index Blood-supply of pancreas, 32 alterations in, as cause of chronic interstitial changes, 164 Bronzed diabetes, 302 Bullet wounds of pancreas, 355 symptoms, 356 treatment, 356 Cachectic atrophy of pancreas, Calcium oxalate in urine in dis- eases of pancreas, 239 Calculi, pancreatic, 169, 473 composition of, 475 glycosuria in, 480 in diabetes, 281 symptoms of, 479 treatment of, 482 Cammidge's reaction as symp- tom of diseases of pancreas, 334 Canals, Saviotti's, 74 Carbohydrates in urine in diseases of pancreas, 240 Carcinoma of liver and chronic pancreatitis, differentiation, 431 of pancreas, 180, 505, 512 and chronic pancreatitis, dif- ferentiation, 340, 429 differential diagnosis, 517 in diabetes, 283 secondary, 186 deposits from, 184 symptoms, 514 treatment, 521 Carnivora, pancreas of, 24 Caruncula major, 36 minor, 39 Casein, pancreatic, 128 Cat, pancreas of, 25 Catarrh of pancreas, 374 pancreolithic, 473 suppurative, of pancreas, 378 Celiac neuralgia, 215 Cell-clumps, intertubular, 78 Cells, centro-acinar, 78 Centro-acinar cells, 78 Chemical pathology of pancreas, 206 Chlorides in urine in diseases of pancreas, 237 Cholecystenterostomy, 463 Cholelithiasis. See Gall-stones. Cirrhosis of liver, pancreas in, 165 of pancreas, 416 Collagen, 117 Comparative anatomy of pan- creas, 17 Connective tissue of pancreas, ar- rangement of, 71 Cystadenoma of pancreas, 177, Cystic disease, congenital, of pan- creas, 177, 509 epithelioma of pancreas, 177, 50s neoplasms of pancreas, 176 simple tumors of pancreas, 505 Cysts, apoplectic, of pancreas, 175, 506 hemorrhagic, of pancreas, 509 hydatid, of pancreas, 177, 510 of pancreas, 172, 487 and trauma, relation of, 179 contents, 500 diagnosis, 497 digestive power of fluid, 500 etiology, 488 in diabetes, 282 physical signs, 492 shape, 498 statistics, 504 symptoms, 490 termination, 502 treatment, 502 statistics on, 504 proliferation, of pancreas, 176, 505 pseudo-, of pancreas, 178 retention, of pancreas, 175 Degeneration, amyloid, of pan- creas, 135 fatty, of pancreas, 134 in diabetes, 281 hyaline, in islands of Langer- hans, in diabetes, 295 of pancreas, 135 Diabete bronze, 302 gras, 278 maigre, 278 Diabetes, 269 absence of islands of Langer- hans in, 297 acute pancreatitis in, 287 after extirpation of pancreas, 269, 270 alcoholism in, 301 arteriosclerosis in, 301 as symptom of diseases of pan- creas, 333 atrophy of islands of Langer- hans in, 297 Index 539 Diabetes, atrophy of pancreas in, 129, 280 auto-intoxication theory of, 273 blood in, 274 bronzed, 302 cancer of pancreas in, 283 chronic interstitial pancreatitis in, 289 cysts of pancreas in, 282 dependence of, upon disease of pancreas, 269 diminished number of islands of Langerhans in, 297 disease of solar plexus as cause of, 272 fatty degeneration of pancreas ' in, 281 gout in, 301 hemorrhage into pancreas in, 287 hyaline degeneration in islands of Langerhans in, 295 infectious natvire of , 2 7 7 inflammatory changes in pan- creas in, 286 interacinar pancreatitis in, 292 nervous system in production of, 276 pancreatic calculi in, 281 extracts in, 306 pathological changes in interaci- nar islets in, 295 secretin in, 306 swelling and increase in size of islands of Langerhans in, 299 syphilis in, 301 treatment of, 305 Diabetes bronze, 166 Diagnosis, general, of diseases of pancreas, 311 Diastase, pancreatic, in Diastatic ferment, method of testing for, in diseases of pan- creas, 267 Digestion glands of animals, comparative anatomy, 17 impaired, of starchy foods, re- duction or failure of pancrea- tic secretion as cause, 223 of fat by pancreatic juice, 117 Digestive disturbances from ab- sence or diminution of pan- creatic secretion, 208 enzymes of pancreatic juice, in functions of pancreas, 97 symptoms in diseases of pan- creas, 320 Diverticulum of Vater, 37, 39 Dog, pancreas of, 25 Duct of Santorini, 38, 44 of Wirsung, 36, 44 pancreatic, accessory, 44 Ducts of pancreas, 36 pancreatic, anatomical varia- tions of, 56 Dyspepsia in diseases of pancreas, 320 Dyspeptic disturbances in chronic pancreatitis, 428 Emaciation in diseases of pan- creas, 320 Embryology of pancreas, 41 Enterokinase, 1x3 Enzymes, digestive, of pancreatic juice, III pancreatic, detection of, 264 Epithelioma, aj'stic, of pancreas, 177' 505 Erepsin, 116 Erythrocytes in pancreatitis, 231 Ethereal sulphates in urine in diseases of pancreas, 233 _ Extirpation of pancreas, diabetes after, 270 partial, alimentary glycosu- ria after, 272 glycosuria after, 271 utilisation of proteids after, Fat, digestion of, by pancreatic juice, 117 in feces in diseases of pancreas, 210, 211 necrosis in diagnosis of diseases of pancreas, 337 of pancreas, 190 and bacterial invasion, 195 experimental production of, 196 Fat-splitting ferment in diseases of pancreas, method of testing for, 266 in urine in diagnosis of diseases of pancreas, 335 in diseases of pancreas, 263 Fatty degeneration of pancreas, 134 in diabetes, 281 feces in diseases of pancreas, 323 infiltration of pancreas, 134 Feces, blood in, in diseases of pan- creas, 332 540 Index Feces, fat in, in diseases of pan- creas, 2IO, 211 fatty, in diseases of pancreas, . 323 . in chronic pancreatitis, 432 in diseases of pancreas, 223, 322 muscle fibres in, in diseases of pancreas, 328 nuclei in, in diseases of pan- creas, 330 white appearance of, in diseases of pancreas, 225 Ferment, diastatic, method of testing for, in diseases of pan- creas, 267 emulsif, 119 fat-splitting, in diseases of pan- creas, method of testing for, 266 in urine, in diagnosis of dis- eases of pancreas, 335 in diseases of pancreas, 263 milk-curdling, of pancreatic juice. III, 128 proteolytic, method of testing for, in pancreatic diseases, of pancreatic juice, 114 Ferments of pancreatic jtiice, in Fever in diseases of pancreas, 313 Fibres, muscle, in feces, in diseases of pancreas, 328 Fibro-adenoma of pancreas, 187 Fish, pancreas of, 20 Fitz's rule in acitte pancreatitis, 399 Focal necrosis of pancreas, 137 Frenum carunculse, 36 Gall-stones, acute pancreatitis from, 389 and chronic pancreatitis, dif- ferentiation, 43 1 pancreatitis from, 365 relation of chronic pancreatitis to, 154 Gangrenous pancreatitis, 147 Gland, abdominal salivary, 71 salivaire abdominale, 71 Glycosuria after partial extirpa- tion of pancreas, 271 alimentary, after partial extir- pation of pancreas, 271 from disturbances of pan- creas, 304 in diagnosis of diseases of .pancreas, 337 Glycosuria, alimentary, relation of pancreas to, 305 as symptom of diseases of pan- creas, 333 association of, with acute pan- creatitis, 288 in diseases of pancreas, 241 in pancreatic calculi, 480 transitory, 276 Gout in diabetes, 301 Hematoma of pancreas, 175 Hemipeptone, 115 Hemochromatosis, 302 chronic interstitial pancreatitis in, 166 Hemoglobin in pancreatitis, 232 Hemorrhage in diseases of pan- creas, 317 into pancreas, pancreatitis from, 368 of pancreas, 137 in diabetes, 287 relation of, to acute pan- creatitis, 149 Hemorrhagic cysts of pancreas, 509 pancreatitis, 138, 146, 149 pancreas in, 147 Hernial sacs, pancreas in, 56 Histology of pancreas, 71 Hyaline degeneration in islands of Langerhans in diabetes, 295 of pancreas, 135 Hydatid cyst of pancreas, 177, 510 Incisura pancreatis, 29 Indicanuria in diagnosis of dis- eases of pancreas, 336 in diseases of pancreas, 233 Induration of pancreas from for- eign substances, 150 Infantilism, pancreatic, 131 Infectious diseases, pancreatitis in, 159 nature of diabetes, 277 Infiltration, fatty, of pancreas, 134 Inflammatory affections of pan- creas; 140, 361 etiology, 363 historical references, 361 changes in pancreas in diabe- tes, 286 Influenza, pancreatitis and, 160 Index 541 Injuries of pancreas, 345 sequels, 358 Interacinar islands, 78 islets, pathological changes in, in diabetes, 295 pancreatitis in diabetes, 292 Interstitial pancreatitis, chronic, in diabetes, 289 Intertubular cell-clumps, yS. Intestine, effect on exclusion of pancreatic secretion on, 210 Islands, interacinar, 78 of Langerhans, 78 absence of, in diabetes, 297 atrophy of, in diabetes, 297 diminished number of, in diabetes, 297 hyaline degeneration in, in diabetes, 295 in production of internal secretion of pancreas, 292 position, 87 size and distribution, 87 swelling and increase in size of, in diabetes, 299 primary, 88 Islet, principal, 88 Islets, interacinar, pathological changes in, in diabetes, 295 Jaundice, black, in diseases of pancreas, 320 in chronic pancreatitis, 428 in diseases of pancreas, 319 Lacerations of pancreas, 346 symptoms and diagnosis, 352 treatment, 353 Lactase, iii, 121 Langerhans, islands of, 78 absence of, in diabetes, 297 atrophy of, in diabetes, 297 diminished number of, in diabetes, 297 hyaline degeneration in, in diabetes, 295 in production of internal secretion of pancreas, 292 position, 87 size and distribution, 87 swelling and increase in size of, in diabetes, 299 Lesser pancreas, 30 Leukocytosis in diseases of pan- creas, 232 Lipuria in diagnosis of diseases of pancreas, 337 in diseases of pancreas, 264 Liver, cancer of, and chronic pan- creatitis, differentiation, 431 cirrhosis of, pancreas in, 165 Lobules of pancreas, 73 Lymphadenoma of pancreas, 188 Lymphatics of pancreas, 34 Maltase, 112 Maltosuria as symptom of diseases of pancreas, 333 in diseases of pancreas, 241 Mammalia, pancreas of, 22 Metabolic symptoms of diseases of pancreas, 333 Milk-curdling ferment of pancrea- tic juice, III, 128 Milk-tube, Schmidt-Stokes, 212 Mumps and pancreatitis, relation of, 160, 369 Muscle fibres in feces in diseases of pancreas, 328 Nausea and vomiting in diseases of pancreas, 321 Necrosis, fat, in diagnosis of dis- eases of pancreas, 337 of pancreas, 190 and bacterial invasion, 195 experimental production of, 196 focal, of pancreas, 137 Nerves of pancreas, 34 Nervous system in production of diabetes, 276 Neuralgia, celiac, 315 Nuclei in feces in diseases of pan- creas, 330 Omental tuberosity of pancreas, Operation for exploring head of pancreas and common bile-duct, 458 Opie's test for fat-splitting fer- ment in urine in diagnosis of diseases of pancreas, 335 Oxaluria in diagnosis of diseases of pancreas, 337 in diseases of pancreas, 239 Pain and tenderness in diseases of pancreas, 314 542 Index Pancreas, abscess of, and suppu- rative pancreatitis, 150 accessorium, 49 adenoma of, 187, 525 diagnosis and treatment, 525 alveoli of, 76 amyloid degeneration of, 135 anatomical anomalies of, 46 anatomy of, 28 comparative, 17 apoplectic cysts of, 175, 506 arteries of, 32, 33 as seat of secondary deposits of malignant growths, 186 atrophy of, 129 cachectic, 130 in diabetes, 129, 280 secondary, 133 blood changes from diseases of, 230 blood-supply of, 32 alterations in, as cause of chronic interstitial changes, 164 body of, 28, 30 bullet wounds of, 355 symptoms, 356 treatment, 356 calculi of, 473. See also Pan- creatic calculi. cancer of, 180, 505, 512 and chronic pancreatitis, dif- ferentiation, 340, 429 differential diagnosis, 517 in diabetes, 283 secondary, 186 deposits from, 184 symptoms, 514 treatment, 521 catarrh of, 374 chemical pathology of, 206 cirrhosis of, 416 color of, 28 comparative anatomy of, 17 congenital cystic disease of, 177, 509 connective-tissue of, arrange- ment, 71 consistency of, 28 cystadenoma of, 177, 505 cystic epithelioma of, 177, 505 neoplasms of, 176 simple tumors of, 505 cysts of, 172, 487 and trauma, relation of, 179 contents, 500 diagnosis, 497 digestive power of fluid, 500 Pancreas, cysts of, etiology, 488 in diabetes, 282 physical signs, 492 shape, 498 symptoms, 490 termination, 502 treatment, 502 statistics on, 504 digestive functions of, 97 symptoms in, 320 diseases of, 126 acetone bodies in urine in, 237 alimentary glycosuria from, 304 in diagnosis of, 337 alterations of appetite in, 320 association of, with diseases of biliary passages, 144 azotorrhea in, 328 azoturia in, 236 bile in urine in, 235 black jaundice in, 320 blood in feces in, 332 calcium oxalate in urine in, 239. carbohydrates in urine in, 240 chlorides in urine in, 237 dependence of diabetes upon, 269 diabetes as symptom of, 333 diagnosis, 311 dyspepsia in, 320 emaciation in, 320 ethereal sulphates in urine in, 233 fat in feces in, 210, 211 necrosis in diagnosis of, 337 fat-splitting ferment in, method of testing for, 266 in urine in, 263 in diagnosis of, 335 fatty feces in, 323 feces in, 322 fever in, 313 glycosuria as symptom of, 333 glycosuria in, 241 hemorrhage in, 317 increased flow of saliva in, 33 2 indicanuria in, 233 indicanuria in diagnosis of, jaundice in, 319 leukocytosis in, 232 lipuria in, 264 lipuria in diagnosis of, 337 Index 543 Pancreas, diseases, maltosuria as symptom, 333 maltosuria in, 241 metabolic symptoms, 333 method of testing for diasta- tic ferment in, 267 for proteolytic ferment in, 265 muscle fibres in feces in, 328 nausea and vomiting in, 321 nuclei in feces in, 330 oxaluria in, 239 oxaluria in diagnosis of, 337 pain and tenderness in, 314 pancreatic reaction in urine as symptom of, 334 pentosuria as symptom of, 333 pentosuria in, 241 phosphaturia in, 236 phosphaturia in diagnosis of, 336 physical signs, 312 pressure symptoms in, 316 relation of urobilinuria to bile -pigments in, 235 Sahli's test in, 338 steatorrhea in, 323 stercobilin in, 331 symptomatology, 311 test meals in, 339 urine in, 233 vomiting in, 321 white appearance of feces in, 225 di visum, 49 ducts of, ^6 embryology of, 41 extirpation of, diabetes after, 269, 270 partial alimentary glycosuria after, 271 gl5^cosuria after, 271 utilisation of proteids after, 221 fat necrosis of, 190 and bacterial invasion, 195 experimental production of, 196 fatty degeneration of, 134 in diabetes, 280 infiltration of, 134 feces in diseases of, 223 fibro-adenoma of , 187 focal necrosis of, 137 head of, 28, 29 and common bile-duct, opera- tion for exploring, 458 Pancreas, hematoma of, 175 hemorrhage of, 137 in diabetes, 287 pancreatitis from, 368 relation of, to acute pancrea- titis, 149 hemorrhagic cysts of, 509 histology of, 7 1 hyaline degeneration of, 135 hydatid cyst of, 177, 510 in cirrhosis of liver, 165 in hemorrhagic pancreatitis, 147 in hernial sacs, 56 in syphilis, 161 induration of, from foreign sub- stances, 150 inflammatory affections of, 140, 361 classification, 362 etiology, 363 historical references, 361 changes in, in diabetes, 286 injuries of, 345 sequels, 358 internal secretion of, islands of Langerhans in production of, 292 lacerations of, 346 sj^mptoms and diagnosis, 352 treatment, 353 lesser, 30 lobules of, 73 lymphadenoma of, 188 lymphatics of, 34 methods of exposing, 69 minus, 46 neck of, 28, 30 nerves of, 34 of anthropoidea, 26 of aquatic mammals, 24 of birds, 22 of carnivora, 24 of cat, 25 of dog, 25 of fish, 20 of mammalia, 22 of reptiles, 21 of rodents, 23 of ruminants, 24 omental tuberosity of, 32 parvum, 47 pathology of, 126 penetrating wounds of, 357 treatment, 358 peritoneum of, 35 physiology of, 96 position of, 28 proliferation cysts of, 176, 505 544 Index Pancreas, proximity of, to stom- ach, 65 pseudo-cysts of, 178 pyemic abscess of, 150 relation of, to alimentary glyco- suria, 305 to peritoneum, 68 retention cj'-sts of, 175 retroperitoneal position of, 68 sarcocarcinoma of, 185 sarcoma of, 185, 523 secondary, 186 shape of, 28 size of, 28 structure of, 71 suppurative catarrh of, 378 surgical anatomy of, 64 syphilis of, 526 treatment, 527 tail of, 29, 30, 32 triangle of infection of, 152 tuberculosis of, 163, 525 symptoms and diagnosis, 526 treatment, 526 tumors of, 179, 512 diagnosis and symptoms of, 312 uncinate process of, 29 veins of, 32, 33, 34 wounds of, 345 Pancreatic acne, 175 apoplexy, 138 calculi, 169, 473 composition of, 475 glycosuria in, 480 in diabetes, 281 symptoms of, 479 treatment of, 482 casein, 128 diastase, 11 i duct, accessory, 44 ducts, anatomical variations of, enzymes, detection of, 264 extracts in diabetes, 306 infantilism, 131 juice, digestion of fat by, 117 digestive disturbances from absence or diminution of, 208 enzymes of, 1 1 1 effect of exclusion of, from intestine, 210 on proteid, 114 excessive, 207 external, composition and characters, 106 ferments of, 1 1 1 Pancreatic juice, mechanism of flow, 98 milk-curdling ferment of, 128 proteolytic ferment of, 114 reduction or failure of, as cause of impaired digestion of starchy foods, 223 steapsin of, 118 total daily output, 108 ranula, 175, 498 reaction, 243 in urine as symptom of dis- eases of pancreas, 334 sialorrhea, 332 Pancreatitis, acute, 384 association of glycosuria with, 288 diagnosis of, 398, 406 etiolog}^ of, 387 Fitz's rule in, 399 gall-stones as cause of, 389 in diabetes, 287 relation of pancreatic hemor- rhage to, 149 symptoms of, 384 treatment of, 400 and influenza, 160 and typhoid fever, 159 chronic, 412 and cancer of liver, differen- tiation, 431 of pancreas, differentia- tion, 340, 429 and gall-stones, differentia- tion, 431 bacteria as cause of, 151 diagnosis, differential, 429 dyspeptic disturbances in, 428 etiology of, 412 feces in, 432 from extension of inflamma- tory process from neigh- boring organs, 166 interacinar form, 416 interlobular form, 416 interstitial, in diabetes, 289 in hemochromatosis, 166 interacinar type, 169 interlobular type, 167 jaundice in, 428 pain and tenderness in, 427 prognosis of, 433 relation of, to cholelithiasis, 154 alcohol to, 424 symptoms of, 426 treatment of, 433 Index 545 Pancreatitis, chronic, treatment of, surgical, 435 results from, 468 urine in, 428 erythrocytes in, 231 from hemorrhage into pancreas, 368 gall-stones as cause, 365 gangrenous, 147 hemoglobin in, 232 hemorrhagic, 138, 146, 149 pancreas in, 147 in infectious diseases, 159 influence of alcohol on produc- tion of, 160 interacinar, in diabetes, 292 mumps and, 160 relation of, 369 subacute, 404 treatment of, 407 suppurative, 147 and abscess of pancreas, 150 syphilitic, 161 Pancreolithic catarrh, 473 Pancreolithotomy, 485 Pancreon, 218 Papilla major, 36 minor, 39 Paranuclein, 237 Parotitis and pancreatitis, rela- tion of, 160, 369 Pathology, chemical, of pancreas, 206 of pancreas, 126 Penetrating wounds of pancreas, 357 treatment, 358 Pentosuria as symptom of dis- eases of pancreas, 333 in diseases of pancreas, 241 Peritoneum of pancreas, 35 relations of pancreas to, 68 Phosphaturia in diagnosis of dis- eases of pancreas, 336 in diseases of pancreas, 236 Physiology of pancreas, 96 Pialyn, in Plasmosomes, 78 Plexus, solar, disease of, as cause of diabetes, 272 Plica longitudinalis, 36 Polypeptides, 115 Pressure symptoms in diseases of pancreas, 336 Primary islands, 88 Principal islet, 88 Proliferation cysts of pancreas, 176- 505 35 Prosecretin, 104 Proteids, effect of pancreatic juice on, 114 utilisation of, after extirpation of pancreas, 221 Proteolytic ferment, method of testing for, in pancreatic diseases, 265 of pancreatic juice, 114 Pseudo-cysts of pancreas, 178 Pyemic abscess of pancreas,^! 50 Ranula pancreatica, 175, 498 Reptiles, pancreas of, 2 1 Retention cysts of pancreas, 175 Rodents, pancreas of, 23 Ruminants, pancreas of, 24 Sahli's test in diseases of pan- creas, 338 Saliva, increased flow of, in dis- eases of pancreas, 332 Salivary gland, abdominal, 71 Santorini, duct of, 38, 44 Sapocrinin, 105 Sarcocarcinoma of pancreas, 185 Sarcoma of pancreas, 185, 523 secondary, of pancreas, 186 Saviotti's canals, 74 Schmidt-Stokes milk-tube, 212 Secretin, 103 in diabetes, 306 Sialorrhoea pancreatica, 332 Sinusoids, 81 Solar plexus, disease of, as cause of diabetes, 272 Starchy foods, impaired digestion of, reduction or failure of pan- creatic secretion as cause, 223 Steapsin, in of pancreatic juice, 118 Steatorrhea in diseases of pan- creas, 323 Stercobilin in diseases of pancreas, 331 . . , Stomach, proximity of pancreas to, 65 Structure of pancreas, 71 Svilphates, ethereal, in urine, in diseases of pancreas, 233 Suppurative catarrh of pancreas, 378 . . pancreatitis, 147 and abscess of pancreas, 150 Surgical anatomy of pancreas, 64 Swordfish, pancreas of, 20 546 Index Symptoms, general, of diseases of pancreas, 311 Syphilis in diabetes, 301 of pancreas, 526 treatment, 527 pancreas in, 161 Syphilitic pancreatitis, 161 Tenderness and pain in diseases of pancreas, 314 Test meals in diseases of pancreas, 339. Transitory glycosuria, 276 Trauma, cysts of pancreas and, relation of, 179 Triangle of infection of pancreas, 152. Trypsin, in, 113 Tuberculosis of pancreas, 163, 525 symptoms and diagnosis, 526 treatment, 526 Tumors, cystic, of pancreas, 176 simple, of pancreas, 505 of pancreas, 179, 512 diagnosis and symptoms of, 312 Typhoid fever, pancreatitis and, 159 Uncinate process of pancreas, 29 Urine, acetone bodies in, in dis- eases of pancreas, 237 bile in, in diseases of pancreas, 23s Urine, calcium oxalate in, in dis- eases of pancreas, 239 carbohydrates in, in diseases'of pancreas, 240 ethereal svilphates in, in diseases of pancreas, 233 fat-splitting ferment in, in diag- nosis of diseases of pan- creas, 335 in diseases of pancreas, 263 in chronic pancreatitis, 428 in diseases of pancreas, 233 pancreatic reaction in, 243 as symptom of diseases of pancreas, 334 Urobilinuria, relation of, to bile- pigments, in diseases of pan- creas, 235 Vater, ampulla of, 37, 39 mode of formation, 59 Veins of pancreas, 32, 33, 34 Vomiting and nausea in diseases of pancreas, 321 White appearance of feces in dis- eases of pancreas, 225 Wirsung, duct of, 36, 44 Wounds, bullet, of pancreas, 355 j symptoms, 356 treatment, 356 of pancreas, 345 penetrating, of pancreas, 357 treatment, 358 SAUNDERS' BOOKS on ■ Pathology, Physiology Histology, Embryology and Bacteriology W. B. SAUNDERS COMPANY 925 WALNUT STREET PHILADELPHIA 9, HENRIETTA STREET COVENT GARDEN, LONDON LITERARY SUPERIORITY 'W^HE excellent judgment displayed in the publications of the house * at the very beginning of its career, and the success of the mod- ern business methods employed by it, at once attracted the attention of leading men in the profession, and many of the most prominent writers of America offered their books for publication. Thus, there were produced in rapid succession a number of works that imme- diately placed the house in the front rank of Medical Publishers. One need only cite such instances as Keen's five-volume work on Sur- gery, Kelly and Noble's Gynecology and Abdominal Surgery, Fow- ler's Surgery, Ashton's Gynecology, Moynihan's works on Abdominal Operations and on Gallstones, Sahli, Kinnicutt, and Potter's Diagnostic Methods, Stengel's Pathology, Hirst's Obstetrics, Anders' Practice, DaCosta's Surgery, and the International Text-Book of Surgery, edited by Warren and Gould. These books have made for themselves a place among the best works on their several subjects. A Complete Catalogue of our Publications will be Sent upon Request SAUNDERS' BOOKS ON American Text-Book of Pathology American Text=Book of Pathology. Edited by Ludvig Hektoen, M. D., Professor of Pathology, Rush Medical College, in affiliation with the University of Chicago ; and David Riesman, M. D., Professor of Clinical Medicine, Philadelphia Polyclinic. Handsome imperial octavo, 1245 pages, 443 illustrations, 66 in colors. Cloth, ^7.50 net; Sheep or Half Morocco, 1^9.00 net. MOST SUMPTUOUSLY ILLUSTRATED PATHOLOGY IN ENGLISH The present work is the most representative treatise on the subject that has appeared in Enghsh. It furnishes practitioners and students with a comprehensive text-book on the essential principles and facts in General Pathology and Pathologic Anatomy, with especial emphasis on the relations of the latter to practical medicine. The illustrations are nearly all original, and those in color are printed directly in the text. In fact, the pictorial feature of the work forms a complete atlas of patho- logic anatomy and histology. Quarterly Medical Journal. Sheffield, England "As to the illustrations, we can only say that whilst all of them are good, most of them are really beautiful, and for them alone the book is worth having. Both colored and plain, they are distributed so profusely as to add very largely to the interest of the reader and to help the student. McConneirs Patholo|[y A Manual of Pathology. By Guthrie McConnell, M. D., Path- ologist to the Skin and Cancer Hospital, St. Louis. i2mo of 523 pages, with 170 illustrations. Flexible leather, ;^2.50 net. RECENTLY ISSUED Dr. McConnell has discussed his subject with a clearness and precision of style that render the work of great assistance to both student and practitioner. The illustrations, many of them original, have been introduced for their practical value. PATHOLOGY. Wells* Chemical Pathology Chemical Pathology. Ik^ng a Discussion of General Pathology from the Standpoint of the Chemical Processes Involved. By H. Gideon Wei.ls, Ph.D., M.D,, Assistant Professor of Pathology in the University of Chicago. Octavo of 549 pages. Cloth, $3.25 net; Half Morocco, ^4.75 net. JUST ISSUED Dr. Wells here presents the latest work systematically considering the subject of general pathology from the standpoint of the chemical processes involved. It is written for* the physician, for those engaged in research in pathology and phys- iologic chemistry, and for the medical student. In the introductory chapter are discussed the chemistry and physics of the animal cell, giving the essential facts of ionization, diffusion, osmotic pressure, etc., and the relation of these facts to cellular activities. Special chapters are devoted to Diabetes and to Uric-acid Metabolism a7id Gout. Wm. H. Welch, M.D., Professor of Pathology, Johns Hopkms University . " The work fills a real need in the English literature of a very important subject, and I shall be glad to recommend it to my students." McFarland*s Pathology A Text=Book of Pathology. By Joseph McFarland, M.D., Pro- fessor of Pathology and Bacteriology in the Medico-Chirurgical Col- lege of Philadelphia. Octavo of 818 pages, with 350 illustrations, many in colors. Cloth, ^5.00 net; Half Morocco, ^6.50 net. RECENTLY ISSUED— BEAUTIFULLY ILLUSTRATED Unlike most works on pathology, this work treats the subject, not from the professor's point of view, but from that of the student. The text is admirably elucidated by numerous excellent illustrations, many of them having been especi- ally drawn. American Medicine " We feel confident in saying no other recent treatise, not encyclopedic in character on any subject, contains so much direct and correlated information on the branch with which it deals." SAUNDERS' BOOKS ON Dtirck and Hektoen*s General Pathologic Histolo^ Atlas and Epitome of General Pathologic Histology. By Pr. Dr. H. DiJRCK, of Munich. Edited, with additions, by Ludvig Hek- TOEN, M.D., Professor of Pathology in Rush Medical College, Chicago. 172 colored figures on yy lithographic plates, 36 text-cuts, many in colors, and 353 pages. Cloth, ^5 .00 net. In Saunders' Hand- Atlas Series. A NEW VOLUME— RECENTLY ISSUED This new Atlas will be found even more valuable than the two preceding volumes on Special Pathologic Histology, to which, in a manner, it is a com- panion work. The text gives the generally accepted views in regard to the signifi- cance of pathologic processes, explained in clear and easily understood language. The lithographs in some cases required as many as twenty-six colors to reproduce the original painting. Dr. Hektoen has made many additions of great value. W. T. Councilman, M. D., Professor of Pathologic Anatomy, Harvard University. " I have seen no plates which impress me as so truly representing histologic appearances as do these. The book is a valuable one." HowelFs Physiology A Text=Book of Physiology. By William H. Howell, Ph.D., M. D., Professor of Physiology in the Johns Hopkins University, Balti- more^ Md. Octavo of 915 pages, 275 illustrations. Cloth, $4.00 net. JUST READY— NEW (2d) EDITION Dr. Howell has had many years of experience as a teacher of physiology in several of the leading medical schools, and is therefore exceedingly well fitted to write a text-book on this subject. Main emphasis has been laid upon those facts and views which will be directly helpful in the practical branches of medicine. At the same time, however, sufficient consideration has been given to the experimen- tal side of the science. The entire literature of physiology has been thoroughly dio^ested by Dr. Howell, and the important views and conclusions introduced into his work. Illustrations have been most freely used. " The Lancet, London " This is one of the best recent text-books on physiology, and we warmly commend it to the attention of students who desire to obtain by reading a general, all-round, yet concise survey of the scope, facts, theories, and speculations that make up its subject matter." PATHOLOGY. Stengel's Text-Book of Pathology Just Issued— The New ^5th) Edition A Text-Book of Pathology, By Alfred Stengel, M. D., Professor of Clinical Medicine in the University of Pennsylvania. Octavo volume of 979 pages, with 400 text-illustrations, many in colors, and 7 full-page colored plates. Cloth, ^5.00 net; Sheep or Half Morocco, $6.50 net. WITH 400 TEXT-CUTS. MANY IN COLORS. AND 7 COLORED PLATES In this work the practical application of pathologic facts to clinical medicine is considered more fully than is customary in works on pathology. While the subject of pathology is treated in the broadest way consistent with the size of the book, an effort has been made to present the subject from the point of view of the clinician. In the second part of the work the pathology of individual organs and tissues is treated systematically and quite fully under subheadings that clearly indicate the subject-matter to be found on each page. In this edition the section dealing with General Pathology has been most extensively revised, several of the important chapters having been practically rewritten. A very useful addition is an Appendix treating of th technic of pathologic methods, giving briefly the most important methods at present in use for the study of patholog5^ including, however, only those methods capable of giving satisfactory results. The book will be found to maintain fully its popularity. PERSONAL AND PRESS OPINIONS William H. Welch, M. D.. Professor of Pathology, Johns Hopkins University, Baltimore, Md. " I consider the work abreast of modern pathology, and useful to both students and practi- tioners. It presents in a concise and well-considered form the essential facts of general and special pathologic anatomy, with more than usual emphasis upon pathologic physiology." Ludvig Hektoen, M. D., Professor of Pathology, Rush Medical College, Chicago. " I regard it as the most serviceable text-book for students on this subject yet written by an American author." The Lancet, London "This volume is intended to present the subject of pathology in as practical a form as pos- sible, and more especially from the point of view of the 'clinical pathologist.' These subjects have been faithfully carried out, and a valuable text-book is the result. We can most favorably recommend it to our readers as a thoroughly practical work on clinical pathology." SAUNDERS' BOOKS ON GET A • THE NEW THE BEST iX m 6 r 1 C Si n standard Illustrated Dictionary Just Issued— New (4th) Edition The American Illustrated Medical Dictionary. A new and com- plete dictionary of the terms used in Medicine, Surgery, Dentistry, Pharmacy, Chemistry, and kindred branches ; with over lOO new and elaborate tables and many handsome illustrations. By W. A. Newman Borland, M. D., Editor of " The American Pocket Medical Diction- ary." Large octavo, nearly 850 pages, bound in full flexible leather. Price, ^4.50 net; with thumb index, ^5.00 net. Gives a Maximum Amount of Matter in a Minimum Space, and at the Lowest Possible Cost WITH 2000 NEW TERMS The immediate success of this work is due to the special features that distin- guish it from other books of its kind. It gives a maximum of matter in a mini- mum space and at the lowest possible cost. Though it is practically unabridged, yet by the use of thin bible paper and flexible morocco binding it is only i}( inches thick. The result is a truly luxurious specimen of book-making. In this new edition the book has been thoroughly revised, and upward of two thousand new terms that have appeared in recent medical literature have been added, thus bringing the book absolutely up to date. The book contains hundreds of terms not to be found in any other dictionary, over 100 original tables, and many hand- some illustrations, a number in colors. PERSONAL OPINIONS Howard A. Kelly, M. D., Professor of Gynecology, Johns Hopkins University , Baltimore. " Dr. Borland's dictionary is admirable. It is so well gotten up and of such convenient siee. No errors have been found in my use of it." J. Collins Warren. M.D.. LL.D., F.R.C.S. (Hon.) Professor of. Surgt-}y, Harvard Medical School. " I regard it as a valuable aid to my medical literary work. It is very complete and of convenient size to handle comfortably. I use it in preference to any other." KMnRYOLOCY. Heisler's Text-Book qf Embryology Just Issued — The New fsdj Edition A Text=Book of Embryology, By John C. Heisler, M.D., Pro- fessor of Anatomy in the Medico-Chirurf^ical College, Philadelphia. Octavo volume of 435 pages, with 212 illustrations, 32 of them in colors. Cloth, $'^^.00 net. WITH 212 ILLUSTRATIONS, 32 IN COLORS The fact of embryology having acquired in recent years such great interest in connection with the teaching and with the proper comprehension of human anatomy, it is of first importance to the student of medicine that a concise and yet sufficiently full text-book upon the subject be available. This new edition represents all the latest advances recently made in the science of embryology. Many portions have been entirely rewritten, and a great deal of new and impor- tant matter added. A number of new illustrations have also been introduced and these will prove very valuable. The previous editions of this work filled a gap most admirably, and this new edition will undoubtedly maintain the reputation already won. Heisler' s Embryology has become a standard work. PERSONAL AND PRESS OPINIONS G. Carl Huber. M. D., Professor of Histology and Embryology, University of Michigan, Ann Arbor. " I find the second edition of ' A Text-Book of Embryology' by Dr. Heisler an improve- ment on the first. The figures added increase greatly the value of the work. I am again recommending it to our students." William Wathen. M. D.. Professor of Obstetrics, Abdominal Surgery, and Gynecology, and Dean, Kentucky School of Medicine, Louisville , Ky. " It is systematic, scientific, full of simplicity, and just such a work as a medical student will be able to comprehend." Birmingham Medical Review, England " We can most confidently recommend Dr. Heisler's book to the student of biology or medicine for his careful study, if his aim be to acquire a sound and practical acquaintance with the subject of embryology." SAUNDERS' BOOKS ON Mallory and Wright's Pathologic Technique Recently Issued— Third Edition, Revised and Enlar£>ed Pathologic Technique. A Practical Manual for Workers in Patho- logic Histology, including Directions for the Performance of Autopsies and for Clinical Diagnosis by Laboratory Methods. By Frank B. Mallory, M. D., Associate Professor of Pathology, Harvard Univer- sity ; and James H. Wright, M. D., Director of the Clinico-Pathologic Laboratories, Massachusetts General Hospital. Octavo of 469 pages, with 138 illustrations. Cloth, ^3.00 net. WITH CHAPTERS ON POST-MORTEM TECHNIQUE AND AUTOPSIES In revising the book for the new edition the authors have kept in view the needs of the laboratory worker, whether student, practitioner, or pathologist, for a practical manual of histologic and bacteriologic methods in the study of patho- logic material. Many parts have been rewritten, many new methods have been added, and the number of illustrations has been considerably increased. Among the many changes and additions may be mentioned the amplification of the de- scription of the Parasite of Actinomycosis and the insertion of descriptions of the Bacillus of Bubonic Plague, of the Parasite of Mycetoma, and Wright's methods for the cultivation of Anaerobic Bacteria. There have also been added new staining methods for elastic tissue by Weigert, for bone by Schmorl, and for con- nective tissue by Mallory. The new edition of this valuable work keeps pace with the great advances made in pathology, and will continue to be a most useful laboratory and post-mortem guide, full of practical information. PERSONAL AND PRESS OPINIONS Wm. H. Welch. M. D., Professor of Pathology, Johns Hopkins University , Baltimore. " I have been looking forward to the pubHcation of this book, and I am glad to say that I find it a most useful laboratory and post-mortem guide, full of practical information and well up to date." Boston Medical and Surgical Journal " This manual, since its first appearance, has been recognized as the standard guide in patho- logical technique, and has become well-nigh indispensable to the laboratory worker." journal of the American Medical Association " One of the most complete works on the subject, and one which should be in the library of every physician who hopes to keep pace with the great advances made in pathology." HISTOLOGY. Bohm, Davidoff, anb Huber's Histology A Text=Book of Human Histology. Including Microscopic Tech- nic. By Dr. A. A. Bohm and Dr. M. von Davidoff, of Munich, and G. Garl Huber, M. D., Professor of Histology and Embryology in the University of Michigan, Ann Arbor. Handsome octavo of 528 pages, with 361 beautiful original illustrations. Flexible cloth, ^3.50 net. RECENTLY ISSUED-NEW (2d) EDITION, ENLARGED The work of Drs. Bohm and Davidoff is well known in the German edition, and has been considered one of the most practically useful books on the subject of Human Histology. This second edition has been in great part rewritten and very much enlarged by Dr. Huber, who has also added over one hundred origi- nal illustrations. Dr. Huber' s extensive additions have rendered the work the most complete students' te.xt-book on Histology in existence. Boston Medical and Surgical Journal " Is unquestionably a text-book of tlie first rank, having been carefully written by thorough masters of the subject, and in certain directions it is much superior to any other histological manual." DrewV Invertebrate Zoolo^ A Laboratory Manual of Invertebrate Zoology. By Oilman A. Drew, Ph.D., Professor of Biology at the University of Maine. With the aid of Members of the Zoological Staff of Instructors of the Marine Biolog- ical Laboratory, Woods Holl, Mass. i2mo of 200 pages. Cloth, ^1.25 net. JUST READY The author has had extensive experience in the classroom and the laboratory, being in charge of the Marine Biological Laboralorv at Woods Holl, Massachusetts. This training has fitted him most admirably to write such a book as this. The subject is pre- sented in a logical way, and the type study has been following, as this method has been the prevailing one for many years. SAUNDERS' BOOKS ON McFarland's Pathogenic Bacteria The New (5th) Edition, Revised A Text=Book Upon the Pathogenic Bacteria. By Joseph McFar- LAND, M. D., Professor of Pathology and Bacteriology in the Medico- Chirurgical College of Philadelphia, Pathologist to the Medico-Chirur- gical Hospital, Philadelphia, etc. Octavo volume olf 647 pages, finely illustrated. Cloth, 1^3.50 net. JUST ISSUED This book gives a concise account of the technical procedures necessary in the study of bacteriology, a brief description of the life-history of the important patho- genic bacteria, and sufficient description of the pathologic lesions accompanying the micro-organismal invasions to give an idea of the origin of symptoms and the causes of death. The illustrations are mainly reproductions of tl e best the world affords, and are beautifully executed. In this edition the entire work has been practically rewritten, old matter eliminated, and much new matter inserted. H. B. Anderson, M. D., Professor of Pathology and Bacteriology, Trinity Medical College, Torotito. " The book is a satisfactory one, and I shall take pleasure in recommending it to the students of Trinity College." The Lancet, London " It is excellently adapted for the medical students and practitioners for whom it is avowedly written. . . . The descriptions given are accurate and readable." HilFs Histology and Organography A Manual of Histology and Organography. By Charles Hill, M. D., Professor of Histology and Embryology, Northwestern Univer- sity, Chicago. i2mo of 463 pages, 313 illustrations. Flexible leather, ^2.00 net. RECENTLY ISSUED Dr. Hill's fifteen years' experience as a teacher of histology has enabled him to present a work characterized by clearness and brevity of style and a completeness of discussion rarely met in a book of its pretensions. Particular consideration is given the mouth and teeth ; and illustrations are most freely used. BA CFERIOLOG Y AND FA THOLOG Y. Eyre*s Bacteriologic Technique The Elements of Bacteriologic Technique. A Laboratory Guide for the Medical, Dental, and Technical Student. By J. W. H. Eyre, M. D., F. R. S. Edin., Bacteriologist to Guy's Hospital, London, and Lecturer on Bacteriology at the Medical and Dental Schools, etc. Octavo volume of 375 pages, with 170 illustrations. Cloth, ;^2.50 net. FOR MEDICAL. DENTAL. AND TECHNICAL STUDENTS This book presents, concisely yet clearly, the various methods at present in use for the study of bacteria, and elucidates such points in their life-histories as are debatable or still undetermined. It includes only those methods that are capable of giving satisfactory results even in the hands of beginners. The illus- trations are numerous and practical. The work is designed with the needs of the technical student generally constantly in view. The Lancet, London " Stamped throughout with evidence that the writer is a practical teacher, and the directions are more clearly given . . . than in any previous work." Warren's Pathology and Therapeutics Surgical Pathology and Therapeutics. By John Collins Warren, M. D., LL.D., F. R. C. S. (Hon.), Professor of Surgery, Harvard Medical School. Octavo, 873 pages, 136 relief and lithographic illustrations, 33 in colors. With an Appendix on Scientific Aids to Surgical Diagnosis and a series of articles on Regional Bacteriology. Cloth, $5.00 net; Sheep or Half Morocco, $6.50 net. SECOND EDITION. WITH AN APPENDIX In the second edition of this book all the important changes have been em- bodied in a new Appendix. In addition to an enumeration of the scientific aids to surgical diagnosis there is presented a series of sections on regional bacteriology, in which are given a description of the flora of the affected part, and the general principles of treating the affections they produce. Roswell Park. M. D., In the Harvard Graduate Magazine. " I think it is the most creditable book on surgical pathology, and the most beautiful medica! illustration of the bookmakers' art that has ever been issued from the American press. SAUNDERS' BOOKS ON Dtirck and Hektoen's Special P&tholo^ic Histolog'y Atlas and Epitome of Special Pathologic Histology. By Dr. H. DiJRCK, of Munich. Edited, with additions, by Ludvig Hektoen, M. D., Professor of Pathology, Rush Medical College, Chicago. In two parts. Part I. — Circulatory, Respiratory, and Gastro-intestinal Tracts. 120 colored figures on 62 plates, and 158 pages of text. Part II. — Liver, Urinary and Sexual Organs, Nervous System, Skin, Muscles, and Bones. 123 colored figures on 60 plates, and 192 pages of text. Per part : Cloth, ;^3.oo net. In Saunders' Hand-Atlas Series. The great value of these plates is that they represent in the exact colors the effect of the stains, which is of such great importance for the differentiation of tissue. The text portion of the book is admirable, and, while brief, it is entirely satisfac- tory in that the leading facts are stated, and so stated that the reader feels he has grasped the subject extensively. William H. Welch, M. D.. Professor of Pathology, Johns Hopkins University, Baltimore. " I consider Diirck's 'Atlas of Special Pathologic Histology,' edited by Hektoen, a very useful book for students and others. The plates are admirable." Sobotta and Huber*s Human Histolo^ Atlas and Epitome of Human Histology. By Privatdocent Dr. J. Sobotta, of Wiirzburg. Edited, with additions, by G. Carl Huber, M. D., Professor of Histology and Embryology in the University of Michigan, Ann Arbor. With 214 colored figures on 80 plates, 68 text-illustrations, and 248 pages of text. Cloth, ^^4.50 net. /« Saunders' Ha?id- Atlas Series. INCLUDING MICROSCOPIC ANATOMY The work combines an abundance of well-chosen and most accurate illustra- tions, with a concise text, and in such a manner as to make it both atlas and text- book. The great majority of the illustrations were made from sections prepared from human tissues, and always from fresh and in every respect normal specimens. The colored lithographic plates have been produced with the aid of over thirty colors. Boston Medical and Surgical Journal " In color and proportion they are characterized by gratifying accuracy and lithographic beauty." PHYSIOLOGY. 13 American Text- Book of Physiology American Text=Book of Physiology. In two volumes. Edited by William H. Howell, Ph.D., M. D., Professor of Physiology in the Johns Hopkins University, Baltimore, Md. Two royal octavo volumes of about 600 pages each, fully illustrated. Per volume : Cloth, $i.QO net; Sheep or Half Morocco, $4.25 net. SECOND EDITION, REVISED AND ENLARGED Even in the short time that has elapsed since the first edition of this work there has been much progress in Physiology, and in this edition the book has been thoroughly revised to keep pace with this progress. The chapter upon the Cen- tral Nervous System has been entirely rewritten. A section on Physical Chem- istry forms a valuable addition, since these views are taking a large part in current discussion in physiologic and medical literature. The Madical News " The work will stand as a work of reference on physiology. To him who desires to know the status of modern physiology, who expects to obtain suggestions as to further physiologic inquiry, we know of none in English which so eminently meets such a demand." Stewart's Physiology A Manual of Physiology, with Practical Exercises. For Students and Practitioners. By G. N. Stewart, M. A., M. D., D. Sc, Professor of Physiology in the University of Chicago, Chicago. Octavo volume of 911 pages, with 395 text-illustrations and colored plates. Cloth, ^4.00 net. RECENTLY ISSUED— NEW (5th) EDITION This work is written in a plain and attractive style that renders it particularly suited to the needs of students. The systematic portion is so treated that it can be used independently of the practical exercises. In the present edition a con- siderable amount of new matter has been added, especially to the chapters on Blood, Digestion, and the Central Nervous System. Philadelphia Medical Journal " Those familiar with the attainments of Prof. Stewart as an original investigator, as a teacher and a writer, need no assurance that in this volume he has presented in a terse, concise, accurate manner the essential and best established facts of physiology in a most attractive manner." 14 SAUNDERS' BOOKS ON Levy and Klemperer's Clinical Bacteriology The Elements of Clinical Bacteriology. By Drs. Ernst Levy and Felix Klemperer, of the University of Strasburg. Translated and edited by Augustus A. Eshner, M. D., Professor of Clinical Medicine, Philadelphia Polyclinic. Octavo volume of 440 pages, fully illustrated. Cloth, ^2.50 net. S. Solis-Cohen, M. D., Professor of Clinical Medicine, Jefferso7t Medical College, Philadelphia. " 1 consider it an excellent book. I have recommended it in speaking to my students." Lehmann, Neumann, and Weaver's Bacteriology Atlas and Epitome of Bacteriology : including a Text-Book of Special Bacteriologic Diagnosis. By Prof. Dr. K. B. Lehmann and Dr. R. O. Neumann, of Wiirzburg. Front the Second Revised and Enlarged German Edition. Edited, with additions, by G. H. Weaver, M. D., Assistant Professor of Pathology and Bacteriology, Rush Medical College, Chicago. In two parts. Part I. — 632 colored figures on 69 lithographic plates. Part II. — 511 pages of text, illustrated. Per part: Cloth, ^2.50 net. In Saunders' Hand-Atlas Series. Lewis' Anatomy and Physi- ology for Nurses Anatomy and Physiology for Nurses. By LeRoy Lewis, M.D., Surgeon to and Lecturer on Anatomy and Physiology for Nurses at the Lewis Hospital, Bay City, Michigan. i2mo of 317 pages, with 146 illustrations. Cloth, $1.^]^ net. JUST ISSUED Nurses Journal of the Pacific Coast " It is not in any sense rudimentary, but comprehensive in its treatment of the subjects in hand." PATHOLOGY, BACTERIOLOGY, AND PHYSIOLOGY. 15 Senn'S Tumors second RevUed Exlhion Pathology and Surgical Treatment of Tumors. By Nicholas Senn, M. D., Ph. D., LL.D., Professor of Surgery, Rush Medical Col lege, Chicago. Handsome octavo, 718 pages, with 478 engravings, including 12 full-page colored plates. Cloth, $5.00 net; Sheep or Half Morocco, ^6.50 net. "The most exhaustive of any recent book in English on this subject. It is well illus- trated, and will doubtless remain as the principal monograph on the subject in our language for some years." — Journal of the Atnerican Medical Association. Stoney's Bacteriology and Technic ^T^^.'^^, BACrRRlOLOGY AND SURGICAL TeCHNIC FOR NURSES. Rv EmILY M. A. Stoney, Superintendent, Carney Hospital, Mass. Revised by Frederic R. Griffith, M.D., Surgeon, N. Y. i2mo of 278 pages, illustrated. $1.50 net. "These subjects are treated most accurately and up to date, without the superfluous reading which is so often employed. . . . Nurses will find this book of the greatest value." — The Trained Nurse and Hospital Review. Clarkson's Histolo^ A Text-Book of Histology. Descriptive and Practical. For the Use of Students. By Arthur Clarkson, M. B., C. M. Edin., formerly Demonstrator of Physiology in the Owen's College, Manchester, Eng- land. Octavo, 554 pages, with 174 colored original illustrations. Cloth, ^4 00 net. " The volume in the hands of students will greatly aid in the comprehension of a sub- ject which in most instances is found rather difficult. . . . The work must be considered a valuable addition to the list of available text-books, and is to be highly recommended.'" . — New York Medical Journal. Gorhatn's Bacteriology A Laboratory Course in Bacteriology. For the Use of Medical, Agricultural, and Industrial Students. By Frederic P. Gorham, A. M., Associate Professor of Biology in Brown University, Providence, R. I., etc. i2mo of 192 pages, with 97 illustrations. Cloth, ^1.25 net. " One of the best students' laboratory guides to the study of bacteriology on the mar- ket. . . . The technic is thoroughly modern and amply sufficient for all practical pur- poses. " — American Journal of the Medical Sciences, Raymond's Physiology NeTsSfSn Human Physiology. By Joseph H. Raymond, A. M., M. D., Pro- fessor of Physiology and Hygiene, Long Island College Hospital, New York. Octavo of 685 pages, with 444 illustrations. Cloth, I3.50 net. " The book is well gotten up and well printed, and may be regarded as a trustworthy guide for the student and a useful work of reference for the genera; practitioner. The illustrations are numerous and are well executed." — The Lancet, London. i6 BACTERIOLOGY, PHYSIOLOGY, AND HISTOLOGY. Ball's Bacteriolo^ Recently Issued— Fifth Edition, Revised Essentials of Bacteriology : being a concise and systematic intro- duction to the Study of Micro organisms. By M. \'. Ball, M. D., Late Bacteriologist to St. Agnes' Hospital, Philadelphia. i2mo of 236 pages, with 96 illustrations, some in colors, and 5 plates. Cloth, |i.oo net. In Saunders" Question- Compend Series. " The technic with regard to media, staining, mounting, and the hke is culled from the latest authoritative works." — The Medical Times, New York. Budgett's Physiology NeT(2d)'E;diS?n Essentials of Physiology. Prepared especially for Students of Medi- cine, and arranged with questions following each chapter. By Sidney P. BuDGETT, M. D., Professor of Physiology, Medical Department of Washington University, St. Louis. i6mo volume of 233 pages, finely illustrated with many full-page half-tones. Cloth, ^i.oo net. In Saunders' Question- Compend Series. "He has an excellent conception of his subject. . . It is one of the most satisfactory books of this class" — University of Pennsylvania Medical Bulletin. V • »»• J « . Recently Issued Leroy s Histology New od) Edition Essentials of Histology. By Louis Leroy, M. D., Professor of Histology and Pathology, Vanderbilt University, Nashville, Tennessee. i2mo, 263 pages, with 92 original illustrations. Cloth, ^i.oo net. In Saunders' Question- Compend Series. " The work in its present form stands as a model of what a student's aid should be ; and we unhesitatingly say that the practitioner as well would find a glance through the book of lasting benefit." — Tke Medical World, Philadelphia. Bastin's Botany Laboratory Exercises in Botany. By the late Edson S. Bastin, M. A. Octavo, 536 pages, with 87 plates. Cloth, ^2.00 net. Barton and Wells* Medical Thesaurus A Thesaurus of Medical Words and Phrases. By Wilfred M, Barton, M. D., Assistant Professor of Materia Medica and Therapeutics, and Walter A. Wells, M.D., Demonstrator of Laryngology, Georgetown University, Washington, D. C. i2mo, 534 pages. Flexible leather, $2.50 net; thumb indexed, $3.00 net. A e rk tj^'Tk.*!* Fifth Revised Edition American Pocket Dictionary just issued Dorland's Pocket Medical Dictionary. Edited by W. A. New- man DoRLAND, M. D., Assistant Obstetrician to the Hospital of the University of Pennsylvania. Containing the pronunciation and defini- tion gf the principal words used in medicine and kindred sciences, with 64 extensive tables. Handsomely bound in flexible leather, with gold edges, $1.00 net; with patent thumb index, $1.25 net. " I can recommend it to our students without r«a«i-ve." — J. H. Holland, M. D., ^the Jefferson Medical College, Philade^)liia. COLUMBIA UNIVERSITY This book is due on the date indicated' below, or at the expiration of a definite period after the date of borrowing, as provided by the rules of the Library or by special ar- rangement with the Librarian in charge. DATE BORROWED DATE DUE DATE BORROWED DATE DUE ■'■' - ■: >^r 1 i 1 CZS(63S)M50