THE FETAL ALCOHOL SYNDROME Public Awareness Campaign Progress Report Concerning The Advance Notice of Proposed Rulemaking on Warning Labels on Containers of Alcoholic Beverages and Addendum The Department of the Treasury and : The Bureau of Alcohol, Tobacco and Firearms February 1979 m— THE FETAL ALCOHOL SYNDROME PUBLIC AWARENESS CAMPAIGN. 1979 PROGRESS REPORT CONCERNING THE ADVANCE NOTICE OF PROPOSED RULEMAKING ON WARNING LABELS ON CONTAINERS OF ALCOHOLIC BEVERAGES AND ADDENDUM Printed for the use of the Department of Treasury, the Bureau of Alcohol, Tobacco and Firearms, and the National Institute on Alcohol Abuse and Alcoholism Washington: Feburary 1979 For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 7 — (p333-HIS PUBLIC HEALTH CONTENTS Page Text of Progress Report: The Fetal Alcohol Syndrome Public Awareness CamPaign ..ueeeeeteeeennnnnneeeeeneennnnnnnn, 1 Addendum to Progress Report A. "Scientific Review of Responses to the Advance Notice of Proposed Rulemaking: Warning Labels on Containers of Alcoholic Beverages," by Gilbert S. Omenn, M.D., Ph.D. Robert Dixon, M.D., and John Ball, M.D., from the Human Resources and Social and Economic Services, Office of Science and Technology Policy, Executive Office of the President, Washington, D.C. vossevcovesssines anes ses 21 B. THE FABRO REPORT: "Alcoholic Beverage Consumption and Outcome of Pregnancy," by Sergio E. Fabro, M.D., Ph.D., from the Fetal-Maternal Division, The George Washington University Medical Center, Washington, D.C. tiiiuiitineenneeeeneeneennnennnens 35 C. THE HALL REPORT, letter to the Bureau of Alcohol, Tobacco and Firearms from Judith G. Hall, M.D., Division of Medical Genetics, Children's Orthopedic Hospital and Medical Center, University of Washington School of Medicine, Seattle, Washington ........... 201 D. THE ETZIONI REPORT: "A Systematic and Graduated Response," by Amitai Etzioni, Ph.D., Center for Policy Research, Inc., New YOrK, New YOUR uc iuissvctecnnsnnnvonsnceonsensse 223 4 bidkealalnl a a” DEPARTMENT OF THE TREASURY Bureau of Alcohol, Tobacco and Firearms [27 CFR Parts 4, 5, and 7] [Notice No. 318; Re: Notice No. 316] THE FETAL ALCOHOL SYNDROME PUBLIC AWARENESS CAMPAIGN Progress Report AGENCY: The Department of the Treasury and the Bureau of Alcohol, Tobacco and Firearms (the Department). ACTION: Progress report on the advance notice of proposed rulemaking (Notice No. 316; 43 FR 2186). SUMMARY : The Department has concluded that there is a need for a public awareness campaign which educates the public about the possible dangers that consumption of liquor, beer, and wine by a pregnant woman can present to an unborn child. Because of the nature of the evidence now available as to the possible dangers, it is not yet clear that warning labels on alcoholic beverage containers would be the best tool to educate the public; and because the Department wants in all instances to avoid unnecessary government regula- tion, it is encouraging the alcoholic beverage industry to work with the Government and private interest groups to educate the public about the possible dangers. If, after a reasonable period of time, the Department does not find that these efforts to provide public information have been successful or that more precise medical evidence becomes available, the proposal to require a warning label on all alcoholic beverage containers will be reconsidered. ADDRESSES: Copies of this progress report are available to interested parties by writing to: Director Bureau of Alcohol, Tobacco and Firearms Post Office Box 385 Washington, D.C. 20044 (Attention: Chief, Regulations and Procedures Division; Notice No. 318) Copies of the addendum to this progress report, which includes the experts' reports and the analysis from the the Office of Science and Technology Policy, the Executive Office of the President, are available for public inspection at the following reading rooms from 8:30 a.m. to 5:00 p.m. BUREAU HEADQUARTERS ATF Reading Room, Room 4408 12th and Pennsylvania Avenue, NW. Washington, D.C. CENTRAL REGION ATF Reading Room, Room 6519 Bureau of Alcohol, Tobacco and Firearms Federal Office Building 550 Main Street Cincinnati, Ohio MID-ATLANTIC REGION ATF Reading Room, Third Floor Bureau of Alcohol, Tobacco and Firearms 2 Penn Center Plaza Philadelphia, Pennsylvania MIDWEST REGION ATF Reading Room Bureau of Alcohol, Tobacco and Firearms 230 South Dearborn Street Chicago, Illinois NORTH-ATLANTIC REGION ATF Reading Room, Sixth Floor Bureau of Alcohol, Tobacco and Firearms 6 World Trade Center New York, New York SOUTHWEST REGION ATF Reading Room, Room 345 Bureau of Alcohol, Tobacco and Firearms Main Tower Building 1200 Main Street Dallas, Texas SOUTHEAST REGION ATF Reading Room Bureau of Alcohol, Tobacco and Firearms 3835 Northeast Expressway Atlanta, Georgia WESTERN REGION ATF Reading Room, Thirty-fourth Floor Bureau of Alcohol, Tobacco and Firearms 525 Market Street San Francisco, California FOR FURTHER INFORMATION CONTACT: Charles N. Bacon or Armida N. Stickney at 202-566-7626. SUPPLEMENTARY INFORMATION: I. BACKGROUND The Bureau of Alcohol, Tobacco and Firearms (ATF) published in the FEDERAL REGISTER on January 16, 1978, an advance notice of proposed rulemaking concerning warning labels on containers of alcoholic beverages. This advance notice requested information to assist ATF in deciding whether the current regulations should be amended to require a warning label on alcoholic beverage containers, concerning the consumption of alcohol by pregnant women and the possible resulting birth defects in their newborn infants. ATF received over 3,000 comments from consumers, special interest groups, industry, doctors, and scientific researchers. The majority of those who wrote said they were opposed to a warning label. The majority of these comments were from con- sumers (2,772). Most of the consumers opposed the warning label, particularly for wine containers. Industry members were also opposed to a warning label on the grounds that, at most, the fetal alcohol syndrome applies only to excessive users of alcohol and that the warning label would be costly and ineffective. The medical profession was divided on the issue; many practitioners felt that moderate use of alcohol during pregnancy has been beneficial while some researchers felt that there is clear evidence linking alcohol use by moderate drinkers to the fetal alcohol syndrome. Several objected, because warning labels would increase feelings of guilt by those women who had consumed alcohol during pregnancy and produced children with birth defects. A general consensus of those opposed was that it is the doctor's responsibility to advise about health hazards rather than the Government's and that any measure by the Government would constitute over-regulation. Many of the comments from doctors and consumers opposing warning labels originated from California, with at least 500 form letters coming from employees of wineries in California. Those in favor of warning labels felt that alcohol consumption by pregnant women is poten- tially dangerous to their children and that a warning label would have a significant effect upon alcohol consumption by these women. They also felt that, in addition, a broad education program for pregnant women should be sponsored by both private concerns and the Government. Others, including women's groups, churches, alcoholism prevention groups, and social service agencies, thought that the proposal was a positive step forward and that ‘a warning label would contribute to better health both for infants and for the public in general. It was also felt that consumers have a right to know what risks they are taking when they use a particular product. Because of the conflicting and sometimes highly technical opinions offered as to the seriousness of the problem and as to the best method for approaching it, the Office of Science and Technology Policy, Executive Office of the President, was asked to evaluate the comments and evidence we received. They submitted an analysis with a recommendation that the Department employ outside, specialized consultants to assess independently the warning label proposal and other approaches that would educate the public. Subse- quently, we engaged three consultants in the following fields of expertise: (a) medical genetics, (b) obstetrics and pharmacology, and (c) social policy. None of these individuals have previously been involved in research or in formulating policy about the fetal alcohol syndrome problem. The experts are Dr. Judith Hall, a medical doctor who is a specialist in genetics and Director of the Division of Medical Genetics at Children's Orthopedic Hospital in Seattle, Washington; Dr. Sergio Fabro, a medical doctor who holds advanced degrees in biological chemistry and pharmacology, and who is Professor and Director of the Fetal-Maternal Medicine Division, George Washington University Medical Center in Washington, D.C.; and Dr. Amitai Etzioni, a sociologist who is Director for the Center for Policy Research, New York, and now a Visiting Fellow at the Brookings Institution. The Department provided two sets of questions for the experts, one of which was directed at the medical experts and the other at the social policy expert. Each of the experts was given copies of the comments received on the advance notice, and the record of the Senate Hearing before the Subcommittee on Alcoholism and Drug Abuse of the Committee on Human Resources, held January 31, 1978. The experts independently evaluated these materials and submitted written reports of their findings to the Department. These written reports were then sent to the National Institute on Alcohol Abuse and Alcoholism; the Food and Drug Administra- tion; the Assistant Secretary for Health, the Department of Health, Education and Welfare; the National Academy of Sciences; and the Office of Science and Technology Policy, Executive Office of the President, for their comments. | \ II. SUMMARY OF EXPERTS' COMMENTS A. FABRO REPORT. 1. Scientific Evidence. Dr. Fabro conludes in this report that the fetal alcohol syndrome does exist. He states that "a series of rather aspecific abnormalities [central nervous system dysfunction; growth deficiencies; cluster of facial abnormalities; and variable major and minor malformations] have been grouped into a 'syndrome' which shows a remarkable reproducibility in that it has been observed only in offspring of chronic alcoholic mothers." He states that some alteration in brain function, growth, and facial appearance are essential for the diagnosis of the full-blown syndrome. He believes that evidence so far indicates that, to produce the full-blown syndrome, the level of alcohol consump- tion must be sufficiently high to cause easily- recognizable chronic alcoholism in the mother. He also states that, while evidence indicates that with lower levels of alcohol the full-blown syndrome is highly unlikely, "some other poor pregnancy outcome (e.g., low birthweight, stillbirth) appears possible." He thinks further, "Prospective epidemiological studies are required to establish whether, for example, wine (2-3 glasses) with dinner or one martini before dinner taken by the mother are devoid of any toxicity for the unborn child." Dr. Fabro believes that "the position taken by the March of Dimes in advising women about alcohol consumption and pregnancy is prudent. Until more data is accumulated in this respect, they advise any woman who is planning to become pregnant to abstain from consuming alcohol-containing beverages." He thinks that heavy alcohol consumption by the chronic alcoholic mother appears essential for the occurrence of infants with the fetal alcohol syndrome, although the mechanism by which these adverse effects are produced is not known. He thinks that it is likely that ethanol in alcoholic beverages is the chemical responsible for producing the fetal alcohol syndrome; however, whether ethanol itself or one of its metabolites is the sole determining factor is not yet clear. Further- more, the importance of other associated factors likely to be present in heavily drinking women such as malnutrition, medical complications (e.g., delirium tremens, cirrhosis), and poor personal habits (cigarette smoking, caffeine intake, and drug abuse) has not yet been defined in regard to establishing the risk of the fetal alcohol syndrome in offspring of chronic alcoholic mothers. As far as the relationship of a woman's drinking pattern to pregnancy outcome is concerned, he thinks it is not known whether there is a critical period during or before pregnancy for the production of the fetal alcohol syndrome, nor is it known whether "binge" drinking is more important in the teratogenicity of alcohol than constant, persistent alcohol abuse, as in chronic alcoholic women. (A /teratogeny/ is a compound which causes malformations in the embryo or fetus when taken by the mother during pregnancy.) He points out that laboratory tests of toxic effects of ethanol given to pregnant animals generally support findings of the toxic effect of ethanol during pregnancy. He cautions, however, against drawing firm conclusions based on animal studies since "at present there is no animal test or battery of tests that can predict human teratogenic risk with complete assurance." He also points out that animal studies were probably the basis for the National Institute on Alcohol Abuse and Alcoholism's statement that "'there is substantial and serious risk for the fetus when the woman chronically drinks 3 or more oz. of absolute alcohol (6 drinks) a day.'" He said the selection of three ounces of absolute alcohol per day as the level when the risk begins to be substantial is an intelligent guess made by extrapolating animal teratological data into humans. As far as the incidence of the fetal alcohol syndrome in children whose mothers are "moderate drinkers" is concerned, he says the full-blown syndrome has only been documented in offspring of chronic alcoholic drinkers although "some toxicity suspicious of, or compatible with, alcohol toxicity during pregnancy has been observed with lower (greater than 1 oz. absolute alcohol per day) consumption." “wx As for the effects of "binge" drinking, he thinks that "according to well-established pharma- cological and toxicological principles, binge drinking, particularly during early pregnancy, may still theoretically be responsible for embryotoxic effects resulting in abortion, stillbirth, intrau- terine growth retardation, and other anomalies "which may be collectively termed ‘'embryotoxic | alcohol related syndrome(s)' (EARS)." “2. Pros and Cons of Warning Labels. Dr. Fabro declines to express an opinion on whether a warning label is justified, citing both incomplete data and his own lack of expertise on the effectiveness of such a measure. He does hink, however, that any warning about the fetal lcohol syndrome should be directed at all women f childbearing age, rather than just pregnant omen, since there are suggestions that alcohol onsumption may exert toxic effects on the human embryo before the mother is aware that she is pregnant. He also thinks it is important to alert he corresponding male population as to possible | toxic effects of alcohol on unborn children. Dr. | Fabro said that effective prevention of the fetal alcohol syndrome may not be possible if doctors are solely responsible for informing women about the risks of drinking primarily because of the possiblity of danger in early stages of pregnancy before she visits the doctor. Ideally, he said, women should know about the dangers as they enter the childbearing years. Dr. Fabro suggests the possibility of explaining the risk during sex education classes in junior and senior high school. 3. Possible Adverse Effects of Warning Labels. While Dr. Fabro does not directly address the question of whether warning labels would have adverse impact, he does say there is "no doubt that our knowledge in this area is scanty and that further research is needed." B. HALL REPORT. l. Scientific Evidence. Dr. Hall finds the evidence "overwhelming" that the fetal alcohol syndrome exists as described by researcher Kenneth L. Jones et al. (Jones' definition includes deficiencies in weight and length; delay of intellectual development and/ or mental deficiency; poor coordination; changes in appearances of face, especially eyes; minor joint anomalies; valve defects of the heart; brain cell myragory abnormalities; and minor genital anomalies.) In addition to the full-blown syndrome, she thinks it is probable that "other more subtle deleterious effects occur in children whose mothers drink during pregnancy," the most serious of which are those which affect the brain and central nervous system. These effects are probably dependent upon the amount of the alcohol ingested, the timing during pregnancy, the genetic background of the mother and fetus, and other environmental factors such as smoking, nutrition, and medication. Dr. Hall says, "This second type of the maternal fetal alcohol spectrum has not yet been fully evaluated or delineated." She cites research studies which indicate "strong hints of behavioral differences, neurologic abnormalities, phychosocial illness...hyperactivity...in the off- spring of moderate and binge drinking mothers; however, long-term follow-up of the infants born to mothers using moderate and minimal amounts of alcohol or those born to binge drinkers has not yet been possible." Based on evidence available both from the studies on the fetal alcohol syndrome and other studies of human teratogens, she believes that "we will see this second category of defects manifested in the children of those women who drank alcohol during pregnancy, perhaps even very small amounts or only during critical periods." Dr. Hall thinks that the relationship between dose and fetal effect - in humans will be difficult to establish in "self- reporting" studies, in part because alcoholics and others notoriously under report the amount they drink, and because there are very likely genetic variations affecting response to dose and timing of consumption. However, since other known human teratogens which cause malformations in large doses usually have an adverse effect in small doses as well, she believes the same kind of dose effect can be anticipated with alcohol. She estimates that the frequency of adverse effects may be as high as 1-2 per 100 births. Dr. Hall believes, "The most susceptible period for detrimental effects of alcohol on the fetus is probably in early pregnancy 10 (the first trimester) ," but she claims that studies have also demonstrated adverse effects in later stages of pregnancy. Dr. Hall also points out, "Since one of the major effects of alcohol is on the developing brain and the brain continues to grow, neurons migrate and cells develop through all three trimesters, it seems likely (and there is some suggestion from the available data) that maternal alcohol consumption at any time during brain development could have an adverse effect." Dr. Hall concludes that, given the present state of knowledge, "no minimum safe level of maternal alcohol consumption can be established at any time during pregnancy." She suggests that the most prudent advice to pregnant women is to avoid alcohol consumption during pregnancy and while nursing. — She believes that the most serious time for alcohol consumption may be during very early pregnancy, prior to making the diagnosis of pregnancy. 2. Pros and Cons of Warning Labels. Dr. Hall recommends that a warning label be required on containers and packaging materials of all alcoholic beverages. She believes women have the right to know that alcohol may endanger the unborn child, and that information about possible danger is not known either by the general public or, more importantly, by the medical profession. For those reasons, she also recommends that a broad educational program for the general public and health professionals be conducted. She states, "There is no question but [that] it is one of the most common known causes of mental retardation," and it is preventable. She thinks it important that an educational program reach all women of childbearing age who may be considering becoming pregnant as well as their family members, husbands, and friends so they can be supportive of a woman's abstinence of alcohol. She thinks the question of previous effectiveness of labels (i.e., cigarettes) is irrelevant since no equivalent situation (a woman altering her behavior to protect her unborn child) has occurred. 3. Possible Adverse Effects of Warning Labels, Dr. Hall finds two issues raised by those opposed to the advance notice which deserve consid- eration. The first issue is whether warning labels would produce unnecessary guilt in women who gave birth to a child (a) which had been damaged by alcohol or (b) which was mentally retarded for some other reason. Dr. Hall thinks this is an "unavoid- able problem" and believes a parent would be "much more incensed and have more guilt if he or she was denied information that would have made it possible to avoid the problem." The second issue is whether giving information directly to the women would interfere with the doctor-patient relationship. She believes most people are now health conscious and want to practice preventive health care, particularly pregnant women. She thinks that the data available from several studies suggest that moderate and "binge" drinking can be controlled and modified by the pregnant woman and that other studies also "suggest that even chronic alcoholics will try to curb their alcohol consumption in the interest of their unborn child." Dr. Hall is "concerned by the number of letters from physicians [in response to the advance notice] who indicated that they do not believe that alcohol can possibly have any adverse effect on the unborn child." She thinks "this attitude reflects the fact that the medical community is not fully informed about the effects of alcohol during gestation and further emphasizes that this information should be directly available to the public." C. ETZIONI REPORT. 1. Scientific Evidence. Dr. Etzioni, who is a sociologist, calls attention to the need for different public policies according to the way the ongoing research on the fetal alcohol syndrome unfolds. He believes different public policies are necessary if it is established that only high dosages, as against both high and low, are harmful; that only continuous drinking is harmful versus continuous and occasional ("binge") drinking; that harmful effects occur relatively late in the fetus 11 12 formation versus during the first two months. In some instances, only pregnant women will have to be reached, in others, all fertile women; in some instances, the target audience would be those who are relatively unresponsive (addicted drinkers), in other circumstances, the relatively more responsive social drinkers. 2. Pros and Cons of Warning Labels. Dr. Etzioni believes the Federal Government should not treat this problem as an isolated issue but should view it as part of a systematic approach to public warnings. He believes warnings required by the Federal Government need to be graduated to differentiate clearly among levels of danger-- according to strength of data available and magnitude of problem caused by the product at issue. He believes high alerts should be used sparingly, taking into account not just the costs which regulations and their enforcement entail, but "costs" in taxing the public's tolerance for government intervention. He also believes as a matter of general principle "the stronger the data, the better the defense one can pose against criticism, and the easier it will be to legitimate the required warnings." He also thinks that "scientific data [need to] go through a sociological process in which they are publicly debated and a slow consensus emerges on the status of the data." He points out that this process is quite different from scientific evaluation, internal to the scientific community. He gives as examples of data that have been processed through the public: the definition of death as a flat brain wave pattern for 48 hours and the relative demerits of heroin versus marijuana. In contrast, he believes data about the fetal alcohol syndrome are not widely known, nor their implications much discussed yet. He, therefore, concludes that "the legitimacy of the claim that alcohol will damage a fetus is not at all well publicized. Regulation would more likely be respected if the data were first 'processed' more publicly...The fact that..data [on the fetal alcohol syndrome] at this stage are poorly processed suggests that the time is not ripe for relatively tough measures from this viewpoint. If higher levels of alert were desired on other grounds, e.g., if new data suggested frequent [fetal alcohol syndrome] risk at early stages of pregnancy, then the public dialogue could be intensified. One of the most effective ways of doing this would be to foster some drastic measures (for instance, sending a registered letter from state health departments to all women aged 14 to 50, or at least to pregnant women, warning that alcohol may harm their fetus.") He points to other considerations which do not favor a high-level alert. One is the effectiveness of warnings as a technique to deter people from behavior that could be harmful. He thinks warnings are more effective when people can switch to some other easily available substitute (from regular to filter cigarettes, birth control pills to other contraceptives, liquid protein diets to other diets). He thinks that warnings would be "more effective if it turned out that drinking wine or beer has no harmful effects on the fetus (a rather unlikely possibility), so that women could be advised to switch to these from hard liquor, or--if it turned out that women could be told to eat before they drink (to dilute the effects of alcohol). We expect warnings to be less effective the fewer 'near' substitutes are available." Dr. Etzioni also thinks that warnings are less effective when both addiction and anxiety are present, which is very likely in the case when a woman is addicted to alcohol and anxious about the health of her unborn child. He goes on, however, to disagree with those who responded to the advance notice, saying that, because of the anxiety problem, information should be withheld so as not to generate more anxiety. "Women are not to be treated as neurotic, infantile creatures who need to be protected from tension. On moral grounds, it would be unethical to prevent millions of pregnant women from making an informed choice on the implausible idea that this bit of anxiety, in a world riddled with anxiety, would be the one that would tax their tolerance to the breaking point." Dr. Etzioni points out that a number of comments submitted by industry suggested that the matter of alert be left largely to the doctor. He says "this reliance would be impossible if early 13 14 and moderate drinking have an effect, because then all women in fertile ages must be notified." He also points to studies which show that doctors have a poor record in transmitting such "preventive" information. He suggests that a broad-based public education program might be funded by adding one penny (or more) to the tax on alcohol and says that "public service announcements on television and radio may do more good than labels, although if one «could have all the aforementioned and labels, they might benefit from the interaction effect...Public service announce- ments can be made to appeal more to motives, senti- ments, and values, quite necessary in addition to, not instead of, informing, if people are to change their minds." He says that he believes further scientific research is needed on the fetal alcohol syndrome before a decision to require warning labels is made. However, he says: If it is considered beneficial for pregnant women, their future children, and the public (which bears a good part of the resulting costs if the warning is ignored) not to consume alcohol, it will not be enough to merely inform women. It will also be necessary to help create a social climate which recognizes that it is unbecoming for a pregnant woman to drink. Thus, just as public service announcements try to break the idea that it is socially well-mannered to offer the driver "one more for the road," it will be necessary to establish that it is asocial to offer a drink to a pregnant women. Public service announcements and public educational campaigns, combined with the advice of physicians and other health professionals, are likely to be relatively more effective than labels or posters in bringing about this change in the social climate, although both of these might help. 3. Possible Adverse Effects of Warning Labels. Dr. Etzioni thinks it important to consider the possible harmful effects that might follow a decision to require warning labels. He thinks it is possible that a "federal label on alcohol may well imply more approval of consuming it (at least by persons not pregnant) than is in the public interest. At least, the 'costs' of such federal legitimization of liquor should be weighed against the benefits of such labels." He also thinks that there are growing segments of the public who ignore warnings, because "they feel they cannot heed the recent avalanche of forbiddens." III. ANALYSIS AND CONCLUSIONS Several conclusions can be drawn from the scientific data: a. Heavy drinking by some women during preg- | nancy can result in a pattern of abnormalities in the offspring, called the fetal alcohol syndrome. This syndrome consists of specific congenital and behavior abnormalities including dysfunction of the central nervous system; growth deficiencies both before— and after birth; cluster of facial abnormalities, especially the eyes; and mental retardation. The full-blown syndrome has only been identified in offspring of those women who are chronic alcoholics. b. There is disagreement among our medical experts--which appears representative of differences within the scientific medical community--about the effects moderate, light, or "binge" drinking may have on the fetus. One expert says it is likely there is a linear relationship between the dose of alcohol and teratogenic effect down to one drink. The other expert does not believe that such a relationship has been proven. On moderate drinking, one expert points to studies which indicate that a mother's moderate drinking can cause a variety of physical and mental abnormalities in her offspring; and the other expert says the evidence is not clear. 15 16 However, both experts say in theory it is possible that "binge" drinking at the "wrong time" can be responsible for birth defects. Moreover, the two medical experts believe that absent clear evidence which establishes a safe level of drinking, the most prudent advice to women planning on becoming preg- nant is to abstain totally from consuming alcohol from the time of conception to birth. C. Studies indicate that pregnant women who are told of the danger alcohol can have on their children are amenable to changing drinking patterns during pregnancy. 4a. There is a low level of awareness of the problem both on the part of the public in general and the medical profession. The Department believes there is sufficient evidence to require further action on the part of industry and Government to educate the population at large, and particularly women of childbearing age, about the problem. The potential consequences are both real and preventable; and most people, including many doctors, are generally unaware of the state of scientific knowledge. We agree with the experts' unanimous opinion that responsibility for informing women of the possible risks should not be left solely to a woman's doctor. The scope of education must be broader than the woman who thinks she may be pregnant and therefore consults a doctor. Initially, risks exist at the earliest stages of pregnancy before a doctor has been consulted and some, particularly low income women, never seek pre- natal care. We also agree with the experts' advice that efforts be made to educate the potential male parent so that he can participate in, and be supportive of, any decision as to alcohol consump- tion during pregnancy. We also agree that special efforts should be made to reach the teenage popula- tion because of the recent increases in teenage pregnancies and drinking. One option which has been considered is to require a warning label concerning the risks of alcohol for the pregnant woman on all alcoholic beverage containers. The Department believes that adequate statutory authority exists for such a requirement and that, in many cases, labeling has served as a vehicle for communication. Such an approach would bring some information about this issue to the attention of large segments of the drinking population. In the case of the fetal alcohol syndrome, however, we have concluded that other forms of education should be attempted before deciding on the necessity of labeling. It seems that warning labels can be most effective when the message sought to be conveyed is simple and easy to state. In the case of the fetal alcohol syndrome, however, the state of scientific evidence does not lend itself to such directness. There are differences of opinion as to the important issues of "binge," moderate, and one- time drinking which cannot be ignored if the message to be communicated is to be as accurate as it should be. Other forms of education--which explain more fully the scientific issues involved--may in this situation be more effective than warning labels and still provide a basis for educated, consumer decision-making. It is also important, generally, that warnings not overstate the danger being discussed. We feel that any action plan should consider the fact that it may be unfair to cause people to think that any level of drinking in pregnancy (i.e., one drink) may have caused problems such as mental retardation in their children. Any approach, therefore, must insure that the information which is disseminated is as accurate as possible. We are also not unmindful of the analysis of Dr. Etzioni discussed above concerning the process, impact, and effectiveness of government warnings. It is possible that education not so directly prescribed by the Government may be as, or more, effective than the simple issuance of what some may consider to be just another government warning. Moreover, we also believe that, where appropriate, Government should first allow the private sector the opportunity to take action before mandating a specific approach by rule. In this case, for example, a sense of corporate responsibility and the desire to avoid potential legal liability and future regulation may motivate sufficient actions. 7 18 IV. PLAN OF ACTION The Department thus intends to work with other appropriate Federal agencies (i.e., the Department of Health, Education and Welfare, including the Food and Drug Administration and the National Institute on Alcohol Abuse and Alcoholism), and with members of all segments of the alcoholic beverage industry, and with other interested groups to develop and implement a program of public education. Among the steps we believe should be included in these education programs are: a. Publishing and distributing this report and those of the retained experts to segments of the public, medical schools, and interested organizations; b. Publishing and disseminating brochures for the public and the medical profession; c. Stimulating education programs in schools; d. Developing public service television and radio announcements; and e. Issuing press releases. The Department intends to monitor carefully the impact of these efforts. We recently conducted an independent public opinion poll to measure the current level of public awareness concerning this problem. In about six months to a year, a new poll will be conducted to compare the level of public awareness. After the two polls are analyzed, a judgement will be made as to the sufficiency and impact of the private and the Government's education programs which have been undertaken. If the evaluation of the joint industry- Government public awareness campaign indicates that there is not a sufficiently high level f public awareness of the problems associat~d with the drinking of liquor, beer, and wine, n=rticularly among women of childbearing age, or If uo: precise medical evidence is developed in current or future studies, then the Department will again consider requiring a warning label as a means of direct communication of this and possibly other health problems to the public. V. DRAFTING INFORMATION Officials from the Department of the Treasury and the Bureau of Alcohol, Tobacco and Firearms jointly participated in developing this progress report. VI. AUTHORITY Accordingly, this progress report is issued under the authority contained in section 5 of the Federal Alcohol Administration Act, 49 Stat. 981, as amended (27 U.S.C. 205). Signed: January 23, 1979 /s/ John G. Krogman John G. Krogman, Acting Director Bureau of Alcohol, Tobacco and Firearms. Approved: January 25, 1979 /s/ Richard J. Davis Richard J. Davis, Assistant Secretary (Enforcement and Operations). 19 ADDENDUM A. "SCIENTIFIC REVIEW OF RESPONSES TO THE ADVANCE NOTICE OF PROPOSED RULEMAKING: WARNING LABELS ON CONTAINERS OF ALCOHOLIC BEVERAGES." Washington, D.C. May 1, 1978 23 Scientific Review of Responses to the Advance Notice of Proposed Rulemaking: Warning Labels on Containers of Alcoholic Beverages Effects of Alcohol on the Fetus For more than two hundred years, articles have appeared in the literature suggesting a correlation between alcohol and mentally defective offspring. In 1968, Lemoine and co-workers published in a French medical journal? work on a malformation syndrome found in offspring of female alcoholics, but this germinal research did not initially find its way into the English language literature and received little attention. The teratogenic effects of alcohol were independently rediscovered in 1973 when Jones et al. reported the effects of alcoholism on human fetal development,’ describing a set of signs and symptoms they termed the "fetal alcoho) syndrome" (Fps)*.4 The principal features of the syndrome are distinctive, and involve central nervous system dysfunction, linear growth deficiency before and after birth, and underdevelopment of the mid-face, especially the eyes. In addition to these malformations or morphologic defects, mental retardation was soon discovered to be a prominent characteristic of children with the syndrome, the degree of retardation generally varying with the degree of physical malformation. > The importance of these findings was obvious. The possibility of defining the etiology of mental retardation of a significant portion of children engendered great interest and, thus, stimulated a number of studies, several of which are still in progress.® A ——————————————————————————— *While FAS has received the most attention in this review, it should be noted that alcohol produces a number of adverse effects on the fetus other than FAS, including decreased birth weight and increased number of still- births. 7 24 Studies using at least six different laboratory animal species have demonstrated that alcohol not only produces congenital malformations but biochemical and behavioral alterations of offspring as well.” In addition, neuropathologic studies of the brains of offspring exposed in utero to high concentrations of alcohol show widespread malformation.8 The findings of Jones et al. were followed by a series of case reports of abnormal physical features suggestive of FAS in offspring of alcoholic mothers. Hanson, in 1976, reported 41 cases of FAS and noted an additional 37 cases identified elsewhere.’ Today, over 200 cases of FAS have been reported from more than 20 medical centers around the world. 10:1 In — published reports, the incidence of FAS has been found to be 1-2 per 1,000 live births; expression of some of the syndrome's characteristics occurs 11 at least twice as frequently. Thus, fetal exposure to alcohol may be a major cause of mental retardation. Relation to Quantity Ingested by the Mother Quantification of the levels of alcohol consumption associated with FAS has been especially difficult. That difficulty lies primarily in reliance on a single retrospective interview assessment of a woman's drink- ing history, and is exacerbated considerably by the general tendency of respondents to underestimate consumption. It is particularly true that heavy drinkers underestimate consumption. In several studies, 3:14,15 only about 5 to 10% of pregnant women report consumption of more than one ounce of absolute alcohol per day. In general, studies define the threshold 25 of "heavy" drinking at this level, although some also include in the definition the consumption of 5 or more drinks on occasion (so-called "binge" drinking).** While it is difficult to make a firm estimate, the risk of the development of the full fetal alcohol syndrome in an offspring of a heavy drinker (as defined above) is probably in the 1-4% range. The relation = of dose to the development of the syndrome or any adverse effect is frustrated by the complexity of such variables as the determination of an accurate history of the amount of consumption, the character of intake (at regular intervals or sporadic), the rate of metabolism, and the effect of other agents, such as cigarettes. Even more difficult is the determination what level is "safe." In fact, both the National Council on Alcoholism and the National Foundation/March of Dimes stress that the only safe decision for expectant mothers is not to drink during pregnancy. 1 9+20 More importantly perhaps, at least one investigator notes that the strongest relationship between maternal alcohol consumption and fetal outcome seems to be drinking behavior in the mother between the time of conception and of recognition of pregnancy. ?] Effectiveness of Labeling No good data exist with regard to the effectiveness of labeling alcoholic beverages. It is, however, generally true that pregnancy is a time when women become most concerned about their bodies and are most **Most studies use the questionnaire developed by Jessor!® to estimate alcohol use. This instrument measures the average daily ounces of absolute alcohol (AA) ingested: this value is termed the “AA score." An "AA score" of 1.00 is equivalent to 1 ounce of absolute alcohol or 2 gynces of 100-proof whiskey. Some use the Cahalan "volume-variability" index: heavy drinkers are those who consume a minimum average of 1% drinks per day and at least 5 or 6 drinks on some occasions. amenable to undertaking good health practices. > In addition, there is evidence that women who drink decrease the amount they drink after learning they are pregnant? and the sporadic or binge drinker may significantly alter this potentially hazardous pattern. Although it is clear that heavy drinking during pregnancy is associated with the fetal alcohol syndrome and that abstinence is not, the data are conflicting with regard to the effect of stopping drinking after the first trimester. Streissguth, for example, has suggested that fetal structural effects may occur very early in pregnancy. 22 On the other hand, preliminary data of Rosett et al. suggest that abstinence or marked reduction of alcohol use during the second half of pregnancy can lower the risk of defective offspring. 1° Thus, some evidence exists to suggest that there is a subgroup of drinkers -- pregnant women -- who are amenable to decreasing consumption, and that,if they in fact decrease consumption, they may decrease the risk of fetal malformation and mental retardation. Whether, however, labeling would have any desired effect is open to question. No evidence which directly addressed the question was presented for review. Anecdotal evidence exists that in a possibly analogous situation -- the labeling of cigarettes -- there was initially a decrease in consumption but that present consumption is now as high as before labeling. Several respondents stated their opinion that labeling would be valueless. A final issue concerns the possible adverse health effects of labeling, two such possibilities being raised by the respondents. One was that labeling represented an intrusion in the doctor-patient relationship. 27 A counterpoint to that concern is that there exists precedent in the requirement by the Food and Drug Administration of "patient package inserts" -- informational literature attached to certain drugs. The second concern was that the existence of labeling would increase feelings of guilt in the woman who does not decrease consumption and who bears a retarded or malformed infant. On the other hand, it is likely that feelings of hostility and frustration would be at least as great if the woman were unaware of the i11 effects of alcohol to the fetus until after birth. No data were presented to support or refute this concern. Conclusions 1. It is clear that heavy drinking by some pregnant women produces signs of fetal alcohol syndrome in their offspring. 2. Data exist that moderate drinking by pregnant women also causes a variety of physical and menta) abnormalities in their offspring. 3. The threshold of alcohol consumption required to produce any change is not certain. Conversely, it is not possible at this time to certify a safe level or drinking pattern. 4. It is likely that pregnant women are amenable to changing poor health habits. 5. The effectiveness of labeling of alcoholic beverages is not certain. Recommendation Because of the clinical, political, and regulatory complexity of the issue of labeling of containers of alcoholic beverages, the Office of Science and Technology Policy recommends that a small group of qualified uninvolved consultants be convened to review the available material concerning the issue, and to make recommendations regarding the proposed labeling. 28 10. 11. 12. References Warner RH & Rosett HL: The effects of drinking on offspring: An historical survey of the American and British 1 J. Studies on Alcohol 36:1395-1420, 1975. Lemoine P. Harrousseau H, Borteyru JP, & Menuet JC: jterature. Les enfants de parents alcooliques. Quest Medical 25:477:82, 1968. (Noted in Streissguth AP: Fetal alcohol syndrome: An epidemiologic perspective. An. J. Epidemiology (in press).) Jones LK, Smith DW, Ulleland CN, & Streissguth AP: Pattern of malformation in offspring of chronic alcoholic mothers. Lancet 1:1267-71, 1973. Jones KL & Smith DW: Recognition of the fetal alcohol syndrome in early infancy. Lancet 2:999-1001, 1973. Streissqguth AP: Psychologic handicaps in children wi syndrome. Work in Progress on Alcoholism, Ann. 140-45, 1976. th fetal alcohol N.Y. Acad. Sci. 273: Data from the National Institute on Alcohol Abuse and Alcoholism (NIAAA), Rockville, MD. Ongoing studies are located at Loma Linda, CA; Seattle, WA; Boston, MA; Baltimore, MD; Charleston, SC; La Jolla, CA; Albany, NY; and Waltham, MA. See Warren KR (ed.): Critical review of the fetal al cohol syndrome. (Presented at NIAAA Press Conference, Washington, D.C., June 1, 1977.) warren KR (ed.): Critical review of the fetal alcohol syndrome. (Presented at NIAAA Press Conference, Washington, D.C., June 1, 1977.) The six species include chicken, albino rat, beagle, sheep, rabbit, and mouse. Clarren SK, Alvord Jr. EC, Sumi SM, Streissguth AP & Smith DW. Brain Malformations Related to Prenatal Exposure to Ethanol. J. Ped. 92:64-67, 1978. Hanson, JW, Jones KL, & Smith DW: Fetal alcohol syndrome: Experience with 41 patients. JAMA 235: 1458-60, 1976. Clarren SK & Smith DW: The fetal alcohol syndrome: Experience With 65 patients and a review of the world literature. (In preparation, 1978). Warren KR (ed.): Critical review of the fetal alcohol syndrome. (Presented at NIAAA Press Conference, Washington, D.C., June 1, 1977.) Little RE: Alcohol consumption during pregnancy as reported to the obstetrician and to an independent interviewer. Sci. 273:588-92, 1976. Ann.N.Y. Acad. 13. 14. 15. 16. 17. 18. 19. 20. el. 22, 2 Little RE: Moderate alcohol use during pregnancy and decreased infant birth weight. Am.J. Public Health 67:1154-56, 1977. Little RE, Schultz FA & Mandell W: Drinking during pregnancy. J.Studies on Alcohol 37:375-79, 1976. Rosett ML, Ouellette EM, Weiner L & Owens E: Therapy of heavy drinking during pregnancy. (In press). (Noted in Warren KR.) Jessor R, Graves T, Hanson R, & Jessor S: Society, Personality, and Deviant Behavior, 1968. Cahalan D, Cisin IH & Crossley HM: American Drinking Practices. Monographs of the Rutgers Center on Alcohol Studies No. 6, 1969. This figure is computed by taking the Clarren (reference 8) estimate of 1-2 per thousand live births and factoring in maternal estimates of alcohol consumption (5-10%). National Council on Alcoholism, Press Conference, New York, N.Y., May 31, 1977. National Foundation/March of Dimes: When you drink, your unborn baby does, too! (pamphlet, 1977). Hanson JW. Streissguth AP & Smith DW: The effects of moderate alcohol consumption during pregnancy on fetal growth and morphogenesis. J.Pediatrics (In press). Streissguth AP: Fetal alcohol syndrome: An epidmiologic perspective. Am. J. Epidemiology (In press). 39 Reviewed Research Reports Borgman RF & Wardlaw FB: Influence of maternal ethanol consumption in rats upon the dams and the offspring. (published; cite unavailable). Borgman RF & Haselden FH: Experimental cholelithiasis and liver lipids in rabbits. Arch. Pathology 83: 411-14, 1967. Buckalew LW: Developmental and behavioral effects of maternal and fetal/neonatal alcohol exposure. Research Comm. in Psychology, Psychiatry & Behavior 2:179-91, 1977. Buckalew LW: Alcohol: A description and comparison of recent scientific vs. public knowledge. (unpublished). Chernoff GF: The fetal alcohol syndrome in mice: An animal model. Teratology 15:223-30, 1977. Clarren SK, Smith DW: The fetal alcohol syndrome: Experience with 65 patients and a review of the world literature. (submitted for publication). Collard ME & Chen CS: Effect of ethanol on growth of neonate mice as a function of modes of ethanol administration. Quart. J. Stud. Alc. 34:1323-1326, 1973. Cross DL: An overview of the fetal alcoholism syndrome (unpublished). Green HG: Infants of alcohlic mothers. Am. J. Obstet. Gynecol, 1974 (full cite unavailable). Hanson JW, Streissquth AP & Smith DW: The effects of moderate alcohol consumption during pregnancy on fetal growth and morphogenesis. J. Pediatrics (In press). Jones KL, Smith DW, Ulleland CN & Streissguth AP: Pattern of malformation in offspring of chronic alcohlic mothers. Lancet 1: 1267-71, 1973. Kochanny TJ: The fetal alcohol syndrome (unpublished). Landesman-Dwyer S, Keller LS & Streissguth AP: Naturalistic observations on newborns: effects of maternal alcohol intake. Alcoholism, Clinical & Experimental Research (In press). Little RE: Moderate alcohol use during pregnancy and decreased infant birth weight. Am. J. Public Health 67:1154-56, 1977. Little RE: Alcohol consumption during pregnancy as reported to the obstetrician and to an independent interviewer. Ann. N.Y. Acad. Sci. 273: 588-92, 1976. Little RE: Schultz FA & Mandell W: Drinking during pregnancy. J. Studies on Alcohol 37:375-79, 1976. 31 2 Martin J. Martin DC, Lund CH & Streissguth AP: Maternal alcohol ingestion and cigarette smoking and their effects on newborn conditioning. Alcoholism: Clinical & Experimental Research 1:243-47, 1977. McDonald JT: The influence of wine vs. ethanol on nitrogen and mineral balance in adult human males. Wine Advisory Board contract M-94, 1974. Ouellette EM: Alcohol in pregnancy and its effects on offspring. (unpublished). Phillips DS & Stainbrook GL: Effects of early alcohol exposure upon adult learning ability and taste preferences. Physiological Psychology 4:473-75, 1976. Rosett HL, Ouellette E, Weiner L & Owens E: The prenatal clinic: A site for alcoholism prevention and treatment. (published; cite not available). Streissguth AP, Herman CS & Smith DW: Stability of intelligence in the fetal alcohol syndrome: A preliminary report. Alcoholism: Clinical & Experimental Research (in press). Streissguth AP: Maternal drinking and the outcome of pregnancy: Implications for child mental health. Am. J. Orthopsychiat, 47:422-31, 1977. Warner RH & Rosett HL: The effects of drinking on offspring: An historical survey of the American and British literature. J. Studies on Alcohol 36:1395-1415, 1975. Warren KR (ed.): Critical review of the fetal alcohol syndrome. Presented at a National Institute on Alcohol Abuse and Alcoholism Press Conference, Washington, D.C., June 1, 1977. Letters Alabama L.W. Buckalew, Normal California Philip R. Alper, Burlingame David S. Bloom, San Rafael Victor J. Bouillon, Napa David P. Buchanan, Hayward Paul B. Carlat, San Francisco Satya N. Chatterjee, Sacramento David L. Chittenden, Day City Grange Coffin, Berkeley Richard J. Cohen, San Francisco Michael N. Cowan, Burlingame G.G. Daggett, Madera James T. English, San Francisco Dennis R. Eyler, Sacramento Ralph E. Faucett, San Diego Paul A. Fine, Oakland Tom M. Fullenlove, San Francisco Robert J. Gilbert, San Francisco Robert S. Hattner, San Francisco Robert Hawkins, Santa Barbara Frank Hinman, San Francisco John F. Huffman, Mill Valley William K. Hummer, Santa Monica Hilliard J. Katz, San Francisco Lowell W. Kearl, Stockton George H. Koenig, Redwood City Edward C. Lawless, San Francisco William R. Nelson, Modesto Jerrod Normanly, Mountain View Frederick K. Ostermann, Mill Valley Lewis J. Richards, Madera Paul Scholten, San Francisco Don Sebastiani, Sonoma Terry J. Turkat Robert L. Waldon, II., San Luis Obispo Donald A. Youngdahl Eugenio A. Zarate Colorado Richard A. Deitrich, Boulder Connecticut Paul J. Coppola, New Haven Georgia Marshall F. Goldberg, Atlanta Preston Lea Wilds, Augusta I111inois Jackson A. Smith, Maywood Robert C. Stepto, Chicago Edward J. Wawszkiewicz, Chicago Iowa Roy M. Pitkin, Iowa City Kentucky Virginia Shannon Cawood, Ashland Louisiana James A. White, III, Alexandria Massachusetts Richard I. Feinbloom, Cambridge Jack H. Mendelson, Belmont Missouri David J. Pittman, St. Louis New Hampshire Robert A. MacCready, Jaffrey J. S. Vazifdar, Meredith New York F.L. Armstrong, Penn Yan LeClair Bissell, New York Joseph P. Giardino, Gloversville Charles H. Lynch, Syracuse Alfred A. Smith, Valhalla North Carolina John A. Ewing, Chapel Hill Ohio Wendall W. Adams, Cleveland Thomas W. Wykoff, Cleveland Pennsylvania Ralph H. Meyer, Sayre Tennessee Coy Freeman, Knoxville Elvin B. Noxon, Knoxville Texas William S. Horn, Fort Worth Washington John D. bagdade, Seattle Earl P. Benditt, Seattle Robert J. Dramer, Pasco Sharon Landesman-Dwyer, Seattle Ruth E. Little, Seattle Ann P. Streissguth, Seattle England Frank M. Sullivan, London Other George J. Hummer Don Q. Mitchell 33 34 Miscellaneous Materials Chafetz ME: Statement before the Senate Subcommittee on Alcoholism and Drug Abuse, Feb. 13, 1978. Randall CL: Statement before the Senate Subcommittee on Alcoholism and Drug Abuse, Jan. 31, 1978. Rosett HL: Statement before the Senate Subcommittee on Alcoholism and Drug Abuse, Jan. 31, 1978. Fetal-Alcohol Syndrome Workshop (summaries by Brody J, Kuzma JW, LaBlanc AE, Lester D, Phillips RL, & Schenker S). National Council on Alcoholism (extensive material). National Institute on Alcohol Abuse and Alcoholism (Listing of current research projects). Wine Institute (extensive material). ADDENDUM B. THE FABRO REPORT: "ALCOHOLIC BEVERAGE CONSUMPTION AND OUTCOME OF PREGNANCY." Washington, D.C, September 25, 1978 37 PREFACE This document is a written report to the Bureau of Alcohol, Tobacco, and Firearms on the scope of ATF's Fetal Alcohol Syndrome Program, as defined in Contract ATF-78-C-1086, In reviewing the twelve volumes of materials received from the Department of the Treasury, as well as additional available data on maternal alcohol intake and pregnancy out- come, it became evident that there are number of problematic areas in this subject, Although most of them are interrelated, I have attempted to analyze each problem individually, at the expense of some repetition, To this end, after a general introduction, the material presented is divided into 4 main sections, The first section (Section A) contains a review of the available data relating to pregnancy outcome in chronic alco- holics, In addition, this section summarizes evidence obtained thus far in laboratory animals on the toxicity of ethanol during pregnancy. Section B deals with an evaluation of the available data according to known pharmacological, toxicological, teratolo- gical, and epidemiological principles, followed by the Summary and Concluding Remarks (Section C), Section D deals with the answers to the FASP Question- naire, and is followed by the Bibliography (Section E), 38 ACKNOWLEDGEMENTS The help and assistance in many phases of this project of Mrs. Catherine Milton and Mrs, Armida Stickney, Enforcement and Operations Divison of the Department of Treasury are gratefully acknowledged, Thanks are also due to Dr, K, R. Warren of the NIAAA for providing useful information regarding blood ethanol levels and risk for the unborn child; and to Mr. Richard Bast, of the National Clearinghouse for Alcohol Information, for his kindness in supplying pertinent infor- mation and in allowing me to use the Clearinghouse library in Gaithersburg, Maryland, I am also in debt to the secre- tarial staff of Enforcement and Operations, especially Elise Jones, Dorothy Jackson, Kathy Dye, Susy Page, Jody Cothran and Debbie Aboudraah, for the magnificant job they did in typing the manuscript, INDEX Preface Acknowledgements Index Introduction SECTION A-1 Pregnancy Outcome in Women with Medically - Recognized Chronic Alcoholism A-1l, a The Fetal Alcohol Syndrome - A-1l, b Epidemiological Studies in Chronic Alcoholic Women A-2 Epidemiological Studies on the Effects of Maternal Alcohol Consumption on Pregnancy Outcome A-3, a Studies in Laboratory Animals ° A-3, b Comments A-4 Summary v B~1 Pharmacological and Toxicological Principles B-1l, a Pharmacology of Alcoholic Beverages B-1, b Clinical Manifestations of Alcoholism B-1, c¢ Pharmacological and Toxicological Principles. of Alcohol Consumption and Pregnancy Outcome B-2 Teratological Principles B-3 Epidemiological Principles C Summary and Concluding Remarks D FASP Questionnaire E References PAGE 37 38 39 40 43 43 51 53 65 77 82 83 23 96 107 111 126 147 151 176 39 40 INTRODUCTION Intemperance in the use of alcoholic beverages creates many . problems in modern society. These problems may be divided into three categories: psychologic, medical, and sociologic. The chief psychologic problem is why a person drinks excessively, often with full knowledge that such action will result in physical injury to himself and irreparable harm to his family. The medical problem embraces all aspects of alcoholic habituation as well as the diseases which result from over-indulgence in alcohol. The sociologic problem comprises the effects of sustained inebriety on the family and community. Alcoholism has been defined as both a chronic disease and a disorder of behavior, characterized in either context by drinking alcohol to an extent that surpasses the social drinking customs of the community and that interferes with the drinker's health, inter- personal relations, or means of livelihood. Reduced to pharmaco- logic terms, it is addiction to alcohol. TABLE I. ESTIMATED PATTERNS OF AMERICAN ADULT DRINKING PRACTICES IN TOTAL MALE FEMALE TR TAIN POPULATION| PER | POPULATION| PER [POPULATION| PER wad (Millions)2 CENT 3 (Killions) CENT | (Millions) CENT All Persons Age 18+! 149.5 100 7.2 100 78.2 100 Abstainers 47.8 32 16.4 23 31.3 40 Infrequent Drinkers 4 15.0 10 5.7 8 8.6 1] Light Drinkers 9 46.3 31 21.4 30 25.8 33 lloderate Drinkers 6 26.9 18 17.1 24 9.4 12 Heavy Drinkers 7 13.5 9 10.7 15 3.1 4 categories by sex and type of drinker do not add to totals due to rounding. Zpoputation estimates for July 1, 1976 from Bureau of Census. proportionate drinking category estimates for adults age 18+ from Harris Surveys 1972-1974. Drink less than cnce per month. Sorink Tess than 0.22-0z. absolute alcohol per day. _ NCALI 6prink 0.22 - 1.0-0z. absplute alcohol per day. " AL, 7orink more than 1.0-0z. absolute alcohol per day. Compiled by the National Clearinghouse for EXAMPLES OF DAILY CONSUMPTION Alcohol Information. Beer Wine Liquor Light Drinking less than 6-0z. OR less than 2-0z. OR less than 0.5-0z. tloderate Drinking 6 ~- 27-0z. OR 2 - 9-o0z. OR 0.5 - 2.5-0z. } Heavy Drinking more than 27-0z. QR more than 9-0z. OR more than 2.5-0z. RS 77-03 42 The precise number of such persons, commonly called alcoholics in the United States is not known. In 1971, the Department of Health, Education, and Welfare estimated that about 9 million men and women (7 percent of the adult population) "manifested the behavior of alcohol abuse and alcoholism." According a recently issued report by the National Institute of Alcohol Abuse and Alcoholism, as of July 1, 1976, there were 13.5 million heavy drinkers in this country. In the late '60's and early '70's, attention turned to a specific aspect of the multiple problems created by alcohol abuse. Lemoine et al. (1) in 1968, and Jones and his associat in 1973 (2), described a specific dysmorphic syndrome - the Fetal Alcohol Syndrome (FAS) - in offspring of chronic alcoholic mothers and this dramatically sharpened the focus on the rela- tionship between ethanol intake by the pregnant woman and the health of the infant. SECTION A-1: PREGNANCY QUTCOME IN WOMEN WITH MEDICALLY RECOGNIZED CHRONIC ALCOHOLISM It has been known for centuries that the infants of chronic alcoholics are likely to be born "weak and sickly and often look shrivel'd and old, as though they have numbered many years" (3). However, no precise account was given until recent times when Lemoine (1) in France in 1968 and Jones and Smith (2) in 1973 in the U. S. A. described a specific dysmorphic condition, the Fetal Alcohol Syndrome (FAS) characteristic of offspring of chronic alcoholic mothers. SECTION A-1, a: The Fetal Alcohol Syndrome (FAS) By 1976, only three years after Jones' original article, well over 200 cases of infants with FAS had been described in the literature (4), and in almost all cases chronic alcoholism was recognized in the mother during pregnancy. The account of the FAS given below is taken from the reports originating from the 'Dysmorphology Unit of the University of Washington in Seattle. According to Clarren and Smith (4), the abnormalities most typically associated with the FAS can be grouped into four categories (Table 2): 1) central nervous System dysfunction; 2) growth deficiency; 3) a characteristic cluster of facial abnormalities; and 4) variable major and minor malformations. 44 Table 2: Principal Features of the Fetal Alcohol Syndrome Observed in 245 Persons Affected (4) FEATURE MANIFESTATION Central nervous system dysfunction: Intellectual Mild to moderate mental retardation* Neurologic Microcephaly* Poor co-ordination, hypotoniat Behavorial Irritability in infancy* Hyperactivity in childhood? Growth deficiency: Prenatal 2 SD for length & weight* Postnatal <2 SD for length & weight* Disproportionately diminished adipose tissue? Facial characteristics: Eyes Short palpebral fissures* Nose Short, upturned? Hypoplastic philtrum* Maxilla Hypoplastic?t Mouth Thinned upper vermilion* Retrognathia in infancy* Micrognathia or relative prognathia in adolescence?’ *Feature seen in > 80% of +Feature seen in >» 50% of patients. patients. Central nervous system dysfunction - Mental retardation has been one of the most common and serious problems associated with ‘ethanol teratogenicity. In 126 patients with FAS in which performance was evaluated according to standardized testing procedures, 107 (85%) scored more than two standard deviations below the mean (5, 6, 7, 8, 9, 11, 12, 13). While not all affected individuals were retarded, rarely did an affected patient display average or better than average mental ability. Whether the mental deficiency in these patients is the result of prenatal exposure to ethanol or is the consequence of post- 45 natal life with an alcoholic mother must be carefully considered. However, evidence has now accumulated to support a prenatal origin for much of the problem. Streissguth et al. (13) have evaluated intelligence in 20 patients with FAS who had varying degrees of growth deficiency and dysmorphic features. The patients ranged in age from 9 months to 21 years. The average I. Q. was 65 with a range from 16 to 105. In general, the more phenotypically involved in- dividuals had the lower I. Q. scores, suggesting that the prenatal insult which produced the dysmorphic features also produced the mental deficiency. Jones et al. (14) found in a survey of the offspring of 23 chronically alcoholic women from the Perinatal Collaborative Project of the National Institutes of Neurologic Diseases and Stroke that 44% of the surviving offspring had I. Q. scores which were below 80. Intelligence scores were not statistically different between those children raised with their alcoholic parents and those raised by relatives or in foster homes. Clarren et al. (15) have recently demonstrated structural alterations in the brains of infants exposed to alcohol in utero. Four brains showed similar malformations related to failure or interruption of neuronal and glial migrations. While the type of malformation was similar in each case, the location of the malformations varied from subject to subject. The most consistent anomalies were cerebellar dysplasias and heterotopic cell clusters, especially on the brain surface. In 46 one instance, the malformations were primarily in the cerebrum and there was associated microcephaly. Subtentorial anomalies produced hydrocephalus in two cases but had no untoward effects on head size in another patient. Similar malformations have been described by Majewski in one patient (16). Microcephaly has been an important feature of the FAS. Generally it has been of prenatal onset, although occasionally it has only become apparent with time. Microcephaly reflects deficient brain growth but as the neuropathologic and psychologic studies demonstrate, normocephaly does not necessarily predict normal brain structure or function after intrauterine alcohol exposure. Furthermore, hydrocephaly can be an occasional variant in FAS if the malformations which usually cause limited brain growth also interfere with cerebrospinal fluid dynamics (4). Neurologic abnormalities may be present from birth in FAS, indicating the prenatal nature of this condition. Newborns are usually irritable, tremulous, have a poor suck and have apparent hyperacusis (5, 17). These abnormalities usually last several weeks or months. Older children have most frequently shown hypotonicity (5, 8, 13, 20, 21). However, severe hypertonicity has been observed in at least one older patient (22) and mixed tonicity with hypotonic arms and hypertonic legs has been noted in one other case (23). Seizures beyond the neo- natal period have been surprisingly rare, although neonatal seizures have been occasionally observed (17, 22, 24). 47 Hyperactivity is a frequent component of FAS in young children. The extent to which this behavior is organic versus environmentally deter- mined has not been established. Growth deficiency - Most infants with FAS Are growth deficient at birth for both length and weight. Jones and Smith observed greater deficits in prenatal linear growth than in weight at birth in their initial Seattle patients (2), but it has now been recognized that many affected infants have greater deficits postnatal weight than length. Few infants have demonstrated postnatal catch-up growth. With their poor suck and poor growth, many affected infants have been initially and repeatedly evaluated for failure to thrive. In general, children with FAS remain more than two standard deviations below the mean for height and weight with weight being more severely limited. Decreased adipose tissue is a nearly constant feature of children with FAS. A major complaint of parents bringing their children to clinic for follow-up has been their inability to "fatten up" their "skinny little kid" (4). Although failure to grow at a normal rate has prompted some endo- crinologic studies, appropriate levels of growth hormone, cortisol and gonadotropins have been demonstrated in FAS children (11, 25). Distinctive facial abnormalites - there is a rather typical facial appearance in persons with the fetal alcohol syndrome. While many disorders include mental deficiency and growth deficiency, it is the facial similarities among FAS children which unite them into a discernible entity. While these similarities are clear from the 48 photographs of affected children published by numerous authors, the written descriptions have not always emphasized the same features, and this has led to some confusion. The fetal alcohol syndrome facies is characterized by a few key features: short palpebral fissures, a hypoplastic upper lip with thinned vermi- lion, and diminished to absent philtrum. Frequently the face is further altered by midfacial growth deficiency and mandibular growth deficiency. The growth of the eye, as of the rest of the nervous system, is adversely affected by fetal alcohol exposure. On rare occasions, eye growth has been so deficient that frank micro- phthalmia has been seen (5). Typically, modest growth defi- ciency of the eye is reflected in shortened palpebral fissures. Unfortunately, standards for palpebral fissure size in children are based on rather old data and the means and standard devia- tions are inadequately established (26, 27). Clarren and Smith (4) have viewed short palpebral fissures as one of the most important diagnostic criteria; thus, 59 of their FAS patients (91%) have had this facial feature. Twelve other authors have described short palpebral fissures in 54 of 74 patients (73%). Only Majewski did not find this abnormality to be a frequent feature. He reports short palpebral fissures in less than 107% Of his 76 patients (16). Strabismus and myopia are frequent a— eye prblems. Majewski (20) has reported ptosis in 50% of his patients and blepharophimosis in 107%, but these abnormalities ho have been less frequently noted by other observers. Structural alterations within the eye such as cataracts, optic atrophy, and tortuous retinal vessels have been observed in individual cases. The face, in general, has a drawn appearance. This is produced principally by the thinned upper vermilion and hypo- plastic philtrum, and is further accentuated by the frequent additional feature of midface hypoplasia. The retrusive maxilla contributes to the accompanying flattened profile and occasional downslanting of the palpebral fissures. The nose is frequently short with a low bridge and associated epicanthal folds and anteverted nostrils. The short upturned nose gives the real or apparent impression that the distance from the alae nasae to the upper lip is long. Cleft lip and palate have been occasionally observed. The ear and mandible are involved in some patients. Posterior rotation of the helix is common and alteration in conchal shape occurs occasionally. The mandible is generally small at birth. In some children, micrognathia remains with increasing age, in others the jaw grows relatively better than the midface and an apparent prognathism may be seen in adolescence. Taken as a whole, the face of fetal alcohol syndrome is as distinctive as that of Down's syndrome and is as readily appreciated in the newborn period as in later life. However, the important abnormalities, taken individually, are subtle and not likely to nN be found in standard listings of malformations. 50 Associated major and minor anomalies - While there is an increased frequency of malformations in children with FAS, no one major malformation occurs in the majority of cases. Table 3 lists the major and minor malformations which have been found in at least 2 of the available 245 reports. Table 3: Associated Features of the Fetal Alcohol Syndrome Observed in 245 Persons Affected (4) AREA FREQUENT * occasIoNaLt Eyes Ptosis, strabismus, Myopia, clinical microph- epicanthal folds thalmia, blepharophimosis Ears Posterior rotation Poorly formed concha Mouth Prominent lateral Cleft lip or cleft palate, palatine ridges small teeth with faulty enamel Cardiac Murmurs, especially Ventricular septal defect in early child- great-vessel anomalies, hood, usually tetralogy of Fallot atrial septal defect Renogenital Labial hypoplasia Hypospadias, small rotated kidneys, hydronephrosis Cutaneous Hemangiomas Hirsutism in infancy Skeletal Aberrant palmar Limited joint movements, creases, pectus especially fingers & elbows, excavatum nail hypoplasia, especially 5th, polydactyly, radioulnar synostosis, pectus carinatum, bifid xiphoid, Klippel-Feil anomaly, scoliosis / Muscular Hernias of diaphragm, umbilicus of groin, diastasis recti * Reported in between 26 and 50% of patients. + Reported in between 1 and 25% of patients. Differential diagnosis - In a few severely affected children, a superficial facial resemblance to delLange syn- drome has been noted (4, 6). Two children with gestational histories of substantial ethanol exposure have also been reported to resemble patients with Noonan syndrome 130, 31). Generally, however, the FAS dhenciyne is distinct and not readily confused with other recognized patterns of malformation. SECTION A-1, b: Epidemiological Studies in Chronic Alcoholic Women Two studies have been reported on the outcome of alcoholic pregnancies, but the sample size in both is small. Shurygin (32) described 42 children born to 18 alcoholic mothers in Russia. The 19 offspring born prior to the mothers' alcoholism displayed disorders that were primarily "vegetative, emotional and be- havioral"” with symptom onset at 9-10 years of age and symptom remission with improved social circumstances. On the other hand, many of the 23 children born after full-fledged alcoholism in the mothers had "profound impairments of the CNS that were manifest early in infancy." Fourteen of these 23 were mentally retarded. Unfortunately, little is known regarding the socio-economic or educational status of the families, but it was noted that there were not serious somatic diseases in the mothers, nor materially inadequate circumstances for the children. The second systematic study of offspring of alcoholic mothers was obtained by a review of the charts of the National Institute of Neurologic Disease and Stroke's Perinatal Project 52 (14). This has been a prospective study of about 55,000 women and their offspring who had been observed up to seven years postnatally in 12 medical centers. Unfortunately, direct questions about alcohol use during pregnancy had not been in- cluded in the original research design. However, whenever alcoholism was mentioned in the clinical record, a retrospective entry had been made in the chart summary. Charts were reviewed and in 23 cases there was evidence of chronic alcoholism before and during the pregnancy. Each of these was compared with charts of two non-alcoholic control women matched for socio- economic status, education, race, age, parity, marital status, and the institution where the mother and child were followed. Charts of the offspring were reviewed by another investigator who had no information about the mothers' drinking status. Among the offspring of the alcoholic women, there were four who died in the perinatal period and six who had physical findings con- sistent with the FAS. Thus, the rate of adverse outcome was 43% as compared with a 2% rate in the control group. Offspring of the alcoholic women had smaller growth parameters in the newborn nursery as well as at seven years of age. 1I.Q. testing at age seven was available in 12 offspring of the alcoholic women. Of the 12, the six who had lived with their mothers had a mean I.Q. of 73, while the six who had spent some time with relatives had a mean I.Q. of 84. 53 SECTION A-2: EPIDEMIOLOGICAL STUDIES ON THE EFFECTS OF MATERNAL ALCOHOL CONSUMPTION ON PREGNANCY OUTCOME A number of epidemiologic studies have been initiated in an attempt to establish a relationship between level of alcohol consumption by the mother and pregnancy outcome. Unfortunately, these studies utilized a variety of different techniques to measure alcohol consumption in relation to light, moderate, and heavy drinkers. Not all of the studies are epidemiologically well controlled for confounding factors, and different end-points were sometimes measured. Compari- sons among these studies are therefore difficult and dis- cordant results may be expected. Of course, this also raises the question of the validity of applying the results of such studies to entire populations. In a fairly well-controlled study, Little (28) used a quantity-frequency-variability questionnaire to determine average ounces of absolute alcohol consumed per day. She found that maternal alcohol use in the six months prior to pregnancy and in the fifth through the eighth month of pregnancy was significantly related to birth weight of off~- spring after controlling for maternal age, height, parity, and smoking, as well as gestational age and sex of offspring. She used a sample of 263 women with single live births without medical complications, all middle class, predominantly white, and participating in a health maintenance organization. Daily consumption of one ounce of absolute alcohol per day in the 54 pre-pregnancy period was associated with an average weight decrement of 90.8 grams in the offspring. This amount con- sumed in late pregnancy was associated with a 160 gram decre- ment in birth weight. Kaminski, Rumeau-Rouquette and Schwartz (33) reported similar finding of 9,236 pregnant women studied in Paris. Women consuming over 0.4 liters of wine or an equivalent daily (approximately 1.5 oz. of absolute alcohol) had more stillbirths, higher rate of small-for-date babies, lower average placental weight and infants with lower average birth weight. These findings were significant even when other risk factors related to alcohol consumption were controlled for. Beer drinking seemed to be particularly implicated in the increased risk, compared tc wine drinking. Ouellette, Rosett, Rosman and Wiener (34) in a prospective study of drinking and the outcome of pregnancy in 322 predomi- nantly non-white, high risk, inner-city women, also reported a significant relationship between heavy alcohol use and intra- uterine growth retardation and abnormality in the offspring. Thus, infants born to heavy drinkers had twice the risk of an abnormality as compared to those born to abstinent or moderate drinkers. Thirty-two percent of infants born to heavy drinkers demonstrated congenital anomalies, as compared to 9 per- cent in the abstinent and 14 percent in the moderate group. Microcephaly and multiple congenital anomalies were much more frequent in this group. Women classified as heavy drinkers in this sample had a much higher alcohol intake than those of Little or Kaminski. 55 Two epidemiological studies failed to find a relationship between alcohol intake during pregnancy and birth weight of off- spring (35, 36). Mau and Netter (35) did, however, report a significant relationship between coffee ingestion (in the first trimester) and birth weight of offspring and also between alcohol consumption ( in the first trimester) and shorter gestations. They attempted to control for related factors including maternal age, parity, socioeconomic status, and body weight, by analyzing appropriate subsamples. They reported that the alcohol and coffee effects appear independent of most such factors. A retrospective epidemiologic study of intrauterine growth was conducted by Russell (37), who reviewed records of women attending a county hospital outpatient clinic. She hypothesized that women with an alcohol related psychiatric diagnosis were more likely to have been drinking heavily during a prior preg- nancy than either women selected at random from the general pop- ulation, or women with a psychiatric diagnosis that was not alcohol related. Birth weights of 223 offspring of 81 women with an alcohol related diagnosis were obtained from birth certificates and compared with the birth weights of 276 off- spring of 94 women with a non-alcohol related diagnosis. Each of these offspring were matched with a control selected from birth certificate files employing criteria of sex, county of birth, year of birth, maternal age, race, and education. The birth weight of each infant born to an alcoholic woman was sub- tracted from that of the matched control infant and these 56 differences were analyzed with respect to maternal drinking histories. Russell found that the average birth weight of infants born to women with an alcohol related psychiatric disorder was significantly lower than that of infants born to other women seen in the clinic. No congenital malformations were recorded on the birth certificates of the study children, and only one malformation was recorded on those of matched controls. Drinking histories could be estimated from clinical records for 95 women, not all of whom had been diagnosed as having alcohol related psychiatric problems. When mothers started drinking more than ll years before delivery, the mean birth weight was 898 grams less than matched controls (p<£.01); 6-10 years of maternal drinking, 403 grams (pg .01); 0-5 years, 319 grams (p< .05). Drinking by fathers was not associated with lower birth weights. Three prospective studies on alcohol and pregnancy outcome have been underway in the United States for the past three years, and two of these have focused on a variety of individual infant assessments in addition to the usual Pregnancy outcome variables used in epidemiological studies. The Boston University Study (34, 38) - A collaborative program was initiated in May 1974, by the Departments or Obstetrics and Gynecology, Pediatrics, Pediatric Neurology, and Psychiatry of the Boston University School of Medicine and the Boston City Hospital. In this study, the sample was a high risk inner-city group with a high proportion of non- white and non-married subjects. 57 Cahalan's volume-variability index was used to evaluate alechol consumption. Separate inquiry was made about the use of wine, beer, and liquor. The patients were divided into three groups on the basis of their drinking practices. Heavy drinkers were defined according to Cahalan's criteria for "high- volume high-maximum'; they consumed at least 5 or 6 drinks on some occasions and a minimum average of 1-1/2 drinks per day. In this population the heavy drinkers had a mean consumption of 4 ounces of absolute alcohol per day with 25% drinking between 8 and 16 ounces of absolute alcohol per day. The rare drinkers used alcohol less than once a month. All women who drank more than once a month but did not meet the criteria for heavy drinkers were classified as moderate drinkers. Of the first 559 women surveyed, 52 (10%) were heavy drinkers, 226 (40%) were moderate drinkers and 281 (50%) were rare drinkers. The results so far obtained confirmed that heavy drinking is associated with heavy smoking. Sixty percent of the heavy drinkers smoked a pack or more a day in contrast to 15% of the rare drinkers. Eleven percent of the total population reporting using psychoactive drugs at some point in their lives. In the heavy drinking group, 35% had earlier experience with barbituates, amphetamines, heroin or hallucinogens. Almost all of the patients who had used drugs had switched to alcohol before they became pregnant. The results indicated that women classified as heavy drinkers, as contrasted with moderate and rare drinkers, have a significantly higher chance of bearing abnormal children. Their infants are at a higher risk for congenital anomalies, growth abnormalities and functional disturbances. The complete FAS has not been observed in any of these infants and no specific pattern of anomalies has been found. Four of the 37 infants (11%) born to heavy drinking women were microcephalic as compared with 1 of 245 (0.4%) in the other two groups. However, as Streissguth points out "Unfortunately other risk factors that were also significantly related to maternal drinking in this sample (smoking, maternal age and parity) were not controlled for in the statistical analyses of the outcome variables." (39) Preliminary intervention attempts among 42 heavy drinking women indicated that the 15 women who reported they abstained or significantly decreased their alcohol intake had healthier offspring than the mothers who continued as heavy drinkers during pregnancy. Loma Linda University Study — At the 1977 Fetal Alcohol Syndrame Workshop (San Diego, California) , this large prospective study being conducted at Loma Linda University in Southern California re- ported preliminary descriptive findings on 5,878 women attending the prenatal clinics of four hospitals (40). This study group is representative of the ethnic distribution of the general population of Southern California. Only 2.1% of this study population were classified as heavy drinkers (> 1.0 oz. absolute aloohol/ day). The report focused primarily on the characteristics of the mothers since analysis of the findings among their infants is still in progress. A woman who drinks heavily during pregnancy shows a character- istic profile. She is likely to be an older (>35 years) unmarried woman from the white, Hispanic or Indian group who has had pre- vious abortions. She tends to be a smoker and have a high caf- feine, sugar, and refined food intake. She is more likely to have used hard drugs and to have had binge drinking episodes. Smoking was found to be an important confounding variable. Heavy drinking women tend to be heavier smokers. The University of Washington Study - The University of Washington prospective study was initiated in October 1974. A sample of 1,529 unselected, predominantly middle class, pregnant women receiving prenatal care at two large Seattle hospitals was interviewed in their homes during their fifth month of pregnancy regarding the quantity and frequency of alcohol consumed, diet, smoking, drugs and medications taken during and prior to pregnancy. Eighty percent of the sample drank in the month prior to pregnancy, and eighty-one percent of the sample drank during the first five months of pregnancy. Women whose average daily consumption exceeded 1.0 ounce of absolute alcohol were identified as "heavy" drinkers. 7.2% of the sample drank at that level prior to preg- nancy and 2.47 continued during pregnancy. During pregnancy, 22% of the heavy drinkers were also heavy smokers, and 57% of the heavy somers were also heavy drinkers. In various portions of the study, different subsamples of the 1,529 women were studied. The first subsample (21) consisted of 60 82 women who had been identified as consuming 1.0 ounces or more of absolute alcohol per day in the month prior to and/or during the first months of pregnancy, and had delivered their infants. These infants were paired with control infants randomly selected from among the offspring of light drinking and abstaining study mothers born the same day at the same hospital. Each pair of infants was referred for clinical evaluation without knowledge of maternal drinking history, Of 163 infants examined, 11 were judged as showing clinical signs of a possible prenatal effect of alcohol on growth and morphogenesis (Table 4). Characteristics of Infants Classified as Abnormal (21) Clinical Features¥* Table 4 Average Maternal Alcohol Intake (oz/day) Case # Short Month Prior Small Micro- Palpebral FAS Dysmorphic to Recognition First 5 months Size cephaly Fissures Features (22) of Pregnancy of Pregnancy 1%% + + + + 25.8 1.2 2%% + + + + 5.4 8.6 3 4 + + = 6.7 0.1 4 1 - + + 1.9 0.9 5 oe N.A. N.A. + 1.8 0.6 6 + = + - 1.8 <0.1 + + + - 1.4 0.1 8 - + oe + 1.4 0.7 9 + + + - 1.0 1.2 10 + - + T 0.0 0.0 11 + - + + 0.0 0.0 *small size = weigh/or length <3rd 7 for sex and gestational age; microcephaly = head circumference <3rd 7 for sex and gestational age; short palpebral fissures = width < 1.8 cm. at gestational ages 236 weeks. **infants with fetal alcohol syndrome. +=abnor- mality present, -= abnormality absent, + = borderline abnormality (measurement at 3rd %). N.A. = data not available or norms not applicable. aN — 62 Nine of tnese 11 had mothers with AA*Z1.0 oz. Only two of these infants were classified as having the fetal alcohol syndrome, and both of their mothers were chronic alcoholics (AA 5.0 oz. pre-recognition of pregnancy). The other seven infants, who showed lesser al- terations of growth and morphogenesis "suggestive" of fetal alcohol syndrome, were born to women who reported drinking an average of 1.0 to 6.7 ounces of absolute alcohol per day in the month before recognition of pregnancy. However, the same mothers reported much lower alcohol consumption during the first five months of pregnancy (AA between<< 0.1 to 1.2 oz. per day). Two of the eleven patients who delivered babies with features suggestive of FAS reported no alcohol consumption in the month prior to the recognition of pregnancy and during the first five months of gestation. In examining these data, it is difficult to evaluate the role played by maternal alcohol consumption. In reporting two additional cases of FAS, this interesting paper confirms that chronic alcoholism in the mother is associated with abnormal newborns. However, it is inconclusive as far as the toxicity of "moderate" doses of alcohol consumption is concerned. * (AA = Average Absolute Alcohol Consumption per day; see Section B-3 for methodology and definitions.) 63 The sample size is small and some of the parameters measured (e.g., birth weight and length) are known to be influenced by a variety of other environmental, medical and obstetrical con- founding factors not well controlled for in this study. The statistical significance of the results have been calculated by including in the "moderate-heavy'" drinkers (AA>1.0 ounce/ day) two mothers who were definitely chronic alcoholics and had children with the full-blown FAS. The large difference reported in alcohol consumption between the month prior to recognition of pregnancy and during the first five months of gestation may not be reliable. As has been pointed out (41), the difficulty lies primarily in reliance on a single, retro- spective interview assessment of a woman's drinking history, and is exacerbated by the general tendency of respondents to under estimate consumption (33); it is particularly true that heavy drinkers tend to hide or underestimate alcoholic con- sumption. In a second subsample, 500 neonates of interviewed mothers were examined using the Brazelton Behavioral Scale (42). Two hundred and fifty were offspring of the heaviest drinking women, and 250 were randomly selected from the non-drinkers bl after feeding. The woman's estimate of her l-month pre- pregnancy alcohol consumption was predictive of the infant's: performance deficit, whereas her estimate of her alcohol consumption during pregnancy was not predictive. Maternal alcohol and nicotine consumption in combination resulted in poor neonatal performance on day 2 of life. The authors themselves emphasized that these results should be taken with caution since, at present time, there is no evidence that performance during the neonatal period is in any way predictive of later learning ability. Section A-3, a: STUDIES IN LABORATORY ANIMALS Many investigations of the toxic effects of ethanol on prenatal development in animals have been performed with somewhat contradictory results (for recent reviews see 3, 50, 51). Sandor and Elias (50) studied the effects of ethanol in chick embryos. Early maldevelopment and mortality occurred in a considerable portion of the embryos and the remaining subjects showed a significantly reduced body weight toward the end of incubation. Ethanol exerts a variety of toxic effects of reproduc- tion in eutherian mammals. For example, Badr et al. (52) found that alcohol is mutagenic in the mouse dominant lethal test, exerting the most marked effects against epididymal spermatozoa and late spermatids. Ethyl alcohol has also been shown to affect fertilization and early embryonic development in the golden hamster (60). In this species, the dynamics of the ova's passage along the tube after ovulation is modified by ethanol, so that normal fertilization is disturbed and polyspermia and zygote death follow. 66 Sandor and Amels (53) investigated the effects of ethanol on the intra-uterine development of albino rats. Ethanol (1.5 grams per kg and 2 gms per kg; i.v.) was given at six, seven and eight days of gestation in one group, and at six and seven days of gestation in another group. Ethanol, at 2 gms per kg body weight, induced twice the number of malformations than were found at 1.5 gms per kg. Skeletal abnormalities of the extremities and facial areas were recorded. Skosyreva (54) administered daily 5 ml of 40% alcohol per kg of body weight orally to rats on days 8 to 14 of gestation. Intra-uterine mortality was 25% in the alcohol- treated animals versus 2% in the controls. There was an increased resorption rate and decreased mean fetal birth weight, but no gross skeletal or organic anomalies were noted. Tze and Lee (55) fed female rats water containing ethanol at 30 grams per 100 ml as their only available fluid, in a balanced powdered diet. After one month, their blood alcohol concentration was 61 + 23 mg. A pair-fed group received a 150-calorie diet lacking ethanol. An additional control group had food and water available at all times. All three groups delivered at 21-22 days. Only 50% of the ethanol-fed mothers known to have copulated 67 delivered litters, and their average litter size was significantly lower than the mean litter size of eight pair- fed or control mothers. Some offspring of the alcohol fed mothers exhibited microcephaly; cracked, dry, loose skin; reddened areas on the head and body; and a generally shriveled appearance. Kronick (56) injected pregnant mice interperitoneally with a 25% solution of ethanol in saline at a dosage of 0.03 ml per gm of body weight during different periods of organogenesis. Fetal mortality rate was substantially increased following alcohol treatment on gestation days 9 through 12. The incidence of fetal anomalies was significantly increased following alcohol treatment only on days 8, 9, and 10. Coloboma of the iris was most frequently observed in animals injected on days 8 and 9; absence of the forepaw was seen only in offspring of mothers injected on day 10. Critical periods for other anomalies were not apparent. Chernoff (57) administered alcohol orally via an all- liquid diet (Metrecal plus ethanol, containing 15-35% ethanol-derived calories) to two highly inbred strains of mice (CBA and C3H) which differed in ethanol preference, alcohol dehydrogenase activity, and ethanol sleep time. 63 These diets, JHLOh resulted in alcohol blood levels of 73-398 mg/100 ml in non-pregnant females, were the sole sustenance for the females for at least 30 days before and throughout gestation. A pattern of malformations similar to those observed in human children with the fetal alcohol syndrome was noted in this study. Prenatal death and maldevelopment increased with the level of alcohol intake. At low ethanol levels, deficient occiput ossification, neural anomalies, and decreased fetal weight occurred; cardiac and eyelid dysmorphology were found in fetuses exposed to higher ethanol levels. The pattern of malformations, which was characterized by a dose- response effect and strain differences in susceptibility, indicates that chronic maternal alcoholism is embryolethal and teratogenic in mice. Similar results were obtained by Randall et al, (58) who administered ethanol on days 5 and 10 of gestation to C57BL/6 mice as part of a nutritionally balanced diet. Twice as many fetuses were resorbed in the alcohol-fed group than in the control groups. Experimental animals had at least 1 malformed fetus per litter, while malformed fetuses were evident in only 5 of 29 control litters. While certain experimental pups had multiple anomalies, no more than 1 anomaly was noted in any affected control fetus. The experimental fetuses weighed significantly less DN Oo than controls, and had a significantly higher incidence of skeletal, heart, and abdominal/urogenital anomalies. The results of this study clearly demonstrate the teratogenicity of ethanol in C57BL/6J mice, and add empirical support to the suggestion that ethanol is the direct teratogenic agent in alcoholic beverages. Ellis and Pick (59) have reported efforts to develop an experimental model of the fetal alcohol syndrome in the beagle. A high dose group received 5.3 gms per kg of ethanol, which is equivalent to the daily consumption of about 24 ounces of 100 proof distilled beverage (12.2 ounces of absolute alcohol) by a person weighing 135 pounds. The moderate dose group received 4.5 gms per kg (9.5 ounces of alcohol per 135 pound woman), and a low dose group received 3.0 gms per kg (6.3 ounces per 135 pound woman). The high dose of ethanol induced severe physical dependence in the pregnant beagles and completely suppressed intrauterine tissue differentiation and development of the fertilized and implanted ovum. The moderate dose produced dependence of a milder degree and permitted a more advanced intrauterine development with spontaneous abortion at 6-7 weeks gestation, or in retention within the uterus of abnormal dead fetuses. A low dose produced no detectable dependence and was followed by the birth of normal pups at term. 70 As far as the behavioral effects of intrauterine exposure to ethanol are concerned, Yanai et al. (61) reported that ethanol treatment of pregnant mice had little or no effect on agnostic behavior of their offspring. However, they found that ethanol administered transplacentally significantly increased the cricket-predation behavior of DBA/1BG offspring, but not that of offspring of the less predatory C57BL/10BG strain. Similarly Elis et al. (62) found that alcohol given to pregnant mice throughout gestation resulted in more aggressive male offspring with more locomotor activity and markedly lower concentrations of 5-hydroxytryptamine in the brain. No differences between experimental and control (receiving the same volume of water) animals were noted with regard to either sociability or timidity, or in duration of gestation, number of litters, or body weight. Yanai et al. (63) also showed in two inbred strains of mice (C57 and DBA) that the offspring of parents fed alcohol were more susceptible to audiogenic seizures and had shorter latency to seizure than either pair-fed or normally fed controls. Hyperactivity, small size with unimpaired growth rate and slow maturation of the righting reflex was observed in kittens born to mothers receiving alcohol during the last 2 weeks of pregnancy (63). 71 Elton et al. (6U4) have recently reported the results of some experiments in monkeys. Three of 4 ethanol-habituated pregnant pigtailed macaques voluntarily reduced their consumption of ethanol at conception and increased their consumption after delivery of apparently normal neonates. The offspring of the 4th monkey, whose intake averaged 5.64 g/kg daily from preconception through delivery, was assessed at birth as tremulous, overreactive, and hyperactive. Similar results were obtained in the rat also. Bond et al. (65, 66, 67) found that prenatal ethanol administration leads to an increase in open-field activity of offspring prior to puberty and that these animals display shock-avoidance learning deficits when tested as adults. Increased spontaneous locomotor activity in rats treated prenatally with ethanol was observed by Branchey et al. (68), who also reported increased brain tyrosine hydroxylase activity in these animals. Malakhovskii et al. (69) gave male and female white rats a 15 percent alcohol solution as the only drinking fluid for 3 months; the rats were then mated with control rats 24 hours after alcohol withdrawal. The progeny of the "alcoholized" and control animals were tested at the age of 1 month for a series of behavioral parameters. The frequency of passive and aggressive responses, the speed of acquisition of the 72 conditioned defense reflex, and of choosing a correct escape route were all 2 to 3 times lower in the offspring of "alcoholized" parents than in the controls. Alcoholization of the mother had a slightly stronger effect than that of the father. Microscopy did not reveal any changes in brain structure. However, Druse et al. (70) reported CNS myelin abnormali- ties in rats born to mothers fed ethanol chronically for 1 month prior to conception and during gestation. The ethanol exposed pups showed elevated synthesis of myelin (particularly heavy myelin) at 18 and 25 days and decreased synthesis of all myelin subfractions (heavy, medium, and light) at 54 days. They also demonstrated increased incorporation of (3H)-leucine and (14C)-glucose into the myelin subfractions at 18 and 25 days and decreased in- corporation at 54 days. Although the quantity and rate of synthesis of individual myelin subfractions was ab- normal in the ethanol treated pups, the protein composi- tion of the separated subfractions was normal. Similar findings have been reported by Rosman and Malone (71). Martin et al. (72) also reported results on the toxic effects of ethanol in the pregnant rat. Chronic ethanol consumption by gravid rats produced certain maternal changes as well as changes in the survival, development, and operant performance of offspring. Differences between ethanol and control groups were found for both maternal weight gain and length of gestation, while differences specific to the offspring included: number of neonatal deaths; growth retardation to postnatal day 72; date of eye opening; distance tranversed on day 15; contingent performance in the first week on continuous reinforcement, fixed ratio, and timing appetitive schedules; and number of shock initiations and ability to discriminate contingencies on punishment schedules. No gross abnormalities were observed in the major organs of ethanol-fed animals at necropsy. Possible mechanisms of the various effects were discussed. Rawat (73) studied the effect of maternal ethanol consumption on fetal hepatic metabolism in the rat. The first measurable changes were observed after 13 days of pregnancy, when there was a steady linear increase in hepatic cytoplasmic and mitochondrial NADH reductase activity in ethanol exposed fetuses; adult levels were reached 12 days after birth. Kesaniemi (74, 75) found that the elimination of ethanol was equal in pregnant and non-pregnant rats, but that the acetaldehyde content of peripheral blood after ethanol administration was higher in pregnant than in non-pregnant 4 animals. Since no differences were found in the liver alcohol and acetaldehyde dehydrogenases, a difference in the extrahepatic metabolish of acetaldehyde was suggested. Sze et al. (76) found that mice exposed to ethanol in utero showed increased alcohol dehydrogenase activity as well as an increase in hepatic microsomal mixed function oxidase activity. Henderson and Schenker (77) investigated the effects of maternal alcohol ingestion on fetal brain, as well as on heart, liver and kidney in rat. An all-liquid diet was employed in both the experimental animals (containing 6% w/v ethanol) and pair-fed controls. The chronic oral in- take of alcohol resulted in a maternal blood alcohol concen- tration of 67 mg per 100 ml. The authors noted a signifi- cant increase in mortality (30%) and a decrease in body weight of alcohol-exposed pups compared to pair-fed controls. Protein concentration was unchaged in heart, liver and kidneys, but was slightly elevated in brain of the 3 day- old animals exposed to ethanol in utero. DNA synthesis rates were normal in all four organs, but DNA concentra- tion was significantly lower in the liver of the alcohol- exposed group. In the alcohol group the total RNA levels were significantly depressed by about 10-30 percent (p<0.05) in all four organs studied. In related studies, Rawat (78) observed the effects 75 pf long-term ethanol consumption by pregnant rats on the incorporation of 14C-leucine into ribosomal protein of fetal and neonatal brain. Fetal cerebral ribosomes incorporated 30% less 14C-leucine than controls, and neonatal rats suckling on ethanol-fed mothers showed about a 60% decrease as compared to the control group. Rawat (80) also studied rates of protein synthesis by livers of fetal and neonatal rats exposed in utero to ethanol. Rates of incorporation of labelled leucine into hepatic proteins were significantly lower in the offspring of the ethanol fed rats compared with the control group. Maternal ethanol consumption resulted in decreases in total hepatic RNA content, hepatic RNA/DNA ratio, and ribosomal protein content of the fetal liver. The retarded growth of the offspring of alcoholic mothers could be directly related to inhibition of protein synthesis. A variety of biochemical changes have been observed in fetuses and offspring of rats exposed to ethanol during pregnancy. Maternal ethanol consumption resulted in increased levels of ¥ -aminobutyric acid, but decreased the acetylcholine content of the fetal brain (79) and inhibited proteolytic and tryptophan oxidase activity in the fetal liver (80). A significant, though reversible, retardation in the maturation of adrenal catecholamine 76 stores (81) and elevated synaptosomal tyramine uptake (82) was noted in offspring of rats treated with ethanol during pregnancy. Chronic ethanol ingestion by pregnant rats caused alterations in the development pattern of ornithine decarboxylase activity in the heart and brain of offspring, suggesting that ethanol interferes with fetal polyamine metabolism (83, 84). 77 Section A-3, b: SINT From the experiments summarized above, it is clear that ethanol exerts a variety of toxic effects on reproduction in laboratory animals. For example, ethanol produces dominant lethal effects on mouse spermatozoa (52), interferes with normal tubal function and zygote development in the hamster (60), and causes intrauterine death and/or growth retardation in the developing rat and mouse conceptus (54,56). Moreover, in some species and under certain experimental conditions, ethanol possesses teratogenic activity. Thus, when ethanol was injected into the air space of chicken eggs at early stages of incubation, the major effects were growth re- tardation with generalized malformations of the central nervous system leading to increased chick mortality (50). Intravenous injection of rats with ethanol on days 6 to 8 of gestation caused increased resorption, retardation of skeletal development and some abnormalities of the central nervous system (53). Injecting mice intraperitoneally in midgestation increased the resorption rate and produced offspring with coloboma of the iris and ectrodactyly (56). The administration of ethanol in the drinking water of mice over a 10-day period in mid- to late gestation resulted in small fetuses with an increased rate of minor skeletal variants (57). /8 Administration of ethanol to rats by a gastric cannula on days 8 to 14 of gestation resulted in an increased resorp- tion rate and decreased fetal body weight, but no gross skeletal or visceral abnormalities were produced (54); how- ever, ethanol administered in a liquid diet to mice 5 to 10 of gestation produced an increased indidence of structural abnormalities in addition to embryolethality and growth retardation (58). In general, the toxic effects of ethanol within a single experiment appear to be dose-related, but it is difficult to compare results from different experiments because of variations in ethanol dosages, routes of ethanol administra- tion, and animal species. Although oral ethanol intake by pregnant animals during the period of organogenesis has produced abnormalities in offspring (58), this has not been a constant finding (54). This could be related to the varying modalities of ethanol administration leading to variations in the level and duration of ethanol exposure of the developing embryo. For example, significant congenital abnormalities were not observed in the experiments of Skosyreva (54), who administered single daily oral doses of ethanol (5 ml of 407% ethanol per kg body weight) to 13 rats between the 8th and 14th day of pregnancy. On the contrary, Randall et al. (58) have produced convincing results indi- cating that ethanol supplied as part of a liquid diet to pregnant mice during morphogenesis is significantly 79 teratogenic, In examining these results, one is tempted to hypothesize that the embryo's duration of alcohol exposure may be an important factor in determining the teratogenicity of this compound. Thus, when given in single daily doses high levels of alcohol may be attained, but only for a relatively short time, since ethanol is rapidly metabolized. In experiments in which ethanol is incorporated in a liquid diet, toxicologically important blood levels of ethanol are likely to be maintained throughout most of the day. As will be discussed later in Section B-2, it is not known whether peak blood levels are more important in producing teratogenic effects than is the duration of embryonic exposure to a given chemical. In the case of ethanol, this issue is extremely critical, since the compound has a relatively short half-life due to its rapid metabolic degradation. It may very well be that an occasional high level of alcohol is not sufficient to produce significant toxic effects on the developing conceptus, and that significant teratogenic effects can be obtained only with persistent toxicologically effective levels of ethanol during morphogenesis. In my opinion, this hypothesis bears an important relationship to the possible teratogenicity of "binge" drinking, and it should be actively investigated. SE—— 80 Furthermore, it is likely that threshold values for ethanol teratogenicity vary depending upon the route of administration and the animal species considered. Although a dose-effect relationship has been demonstrated in some experi- ments, in none has a threshold value for the embryolethality, growth-retarding effects and teratogenicity of ethanol been firmly established, nor has an animal experiment clearly defined the most sensitive period of gestation for ethanol toxicity. The well-performed experiments of Chernoff (57) are in- triguing in this regard, since he obtained significant terato- genic effects in two strains of rats which had been made chronically alcoholic for at least one month prior to pregnancy. These experiments in animals, as well as the human experience, raise the possibility that the development of a significant degree of tolerance to ethanol, as in a chronic alcoholic mother, may be an important contributing factor in the tera- togenicity of ethanol. In these cases, whether the fetus is exposed to higher or more frequent peak levels of ethanol because of the maternal alcohol tolerance remains to be assessed. The results of Randall et al., (58) however, indicate that at least in the mouse, this is not an essential feature of ethanol teratogenicity, since congenital abnormalities were produced by an ethanol-containing liquid diet given only during the period of organogenesis. 81 The results obtained in experiments measuring the effects of ethanol during pregnancy on behavioral and biochemical parameters in the offspring are also difficult to Shterpret. There are many inconsistencies in regard to the effects of ethanol during pregnancy on the behavior of offspring (85), but this is understandable in view of the fact that in different experiments different end-points were measured in the offspring; different doses of ethanol were used at different stages of pregnancy, and different strains and animal species were used. Finally, an extrapolation of the results to the human is difficult if not impossible; as will be discussed in more detail in Section B-2, results obtained in animals on the toxicity of chemicals during pregnancy are not fully predictive for the human. 32 Section A-4: SUMMARY In section A the medical and scientific evidence as to the toxic effects of alcoholic beverages and ethanol consumption during pregnancy in the human and in laboratory animals is reviewed and a few comments made. The evidence points to a possible cause—-effect relationship between abuse of alcoholic beverages during pregnancy and toxicity to the unborn child. Such a relationship is clearer when the level of alcohol consumption by the mother represents an easily recognizable chronic alcoholic condition. Thus, most of the cases of full-blown FAS have been reported in offspring of mothers who had been chronic alcoholics for a significant period of time. However, at the present time, it is less clear whether lower maternal doses of ethanol are also associated with toxic effects on the progeny, and whether other associated high risk conditions that are likely to occur in the drinking mother contribute to the toxicity of ethanol. It is also not known during which period of gestation the conceptus is most sensitive to the embryotoxic effects of ethanol exposure and whether a definite dose-effect relation- ship exists. In the following section (Section B), an attempt is made to evaluate the available data on the toxicity of alcohol during pregnancy, using known pharamcological, texicological, teratological and epidemiological principles, 83 SECTION B-1l: PHARMACOLOGICAL AND TOXICOLOGICAL PRINCIPLES SECTION B-1, a: PHARMACOLOGY OF ALCOHOLIC BEVERAGES An understanding of the pharmacological properties of alcohol and alcoholic beverages is essential in order to evaluate the effects of alcohol consumption during pregnancy and its possi- ble damage to the unborn child. An outline of the most important aspects of the pharmacology of alcohol and alcoholic beverages is given below. For a com- prehensive review of this subject, the reader is referred to a series of excellent reviews (86, 87, 88, 89, 90, 91). Pharmacokinetics of Ethanol Ethyl alcohol, or ethanol, is the pharmacologically active ingredient in beer, wine, whiskey, gin, brandy, and other less common alcoholic beverages. In addition to ethanol, alcoholic beverages contain a variety of other chemicals, but as far as we know they have no important pharmacologic properties. A comprehensive review of the chemical composition of alcoholic beverages has recently been published (92). Ethanol is absorbed from both the stomach and the small intestine. Its presence may be detected in the blood within 5 minutes after ingestion, and the maximum concentration is reached in 30 to 90 minutes. The ingestion of milk and fatty foods impedes, and water facilitates, its absorption. There is no indication that high-alcohol beverages are more toxic than low-alcohol beverages, based on equivalent amounts of alcohol, or that they contain any peculiarly toxic component. On the other hand, there are some results indicat- ing that following ingestion of equivalent amounts of ethanol, vodka, gin and whiskey produce higher blood levels of ethanol than beer and wine (93). After entering the bloodstream, alcohol reaches various organs, as well as spinal fluid, urine, and pulmonary alveolar air, attaining concentrations which bear a constant relationship to those in the blood. Alcohol distributes in the total body water (TBW), and in fact, blood ethanol concentrations reflect the TBW volume. TBW in women is 45.5% of body weight, whereas it is larger (51.5%) for men. Thus, after ingesting equal amounts of ethanol, higher blood ethanol levels will be present in women than in men. Lester (94, 95), on the basis of his own work and a review of the literature concluded that man and most mammals produce approximately 12-39 g of endogenous ethanol/24 hr. Krebs and Perkins (96) showed in rats and Blamstrand (97) in a patient that this endogenous alcohol is formed in the gut. Alcohol is metabolized principally through oxidation by alcohol dehydrogenase (ADH), first to acetaldehyde and then to acetate, which is released into hepatic venous blood. The acetate is then metabolized mainly in peripheral tissues to acetyl coenzyme A, which enters the Krebs cycle, with eventual release of water 85 and carbon dioxide. Less than 10% is excreted chemically unchanged in urine, sweat and breath. These metabolic pathways are well known and widely described. Although 80% or more of ingested alcohol is metabolized in the liver, small amounts of ADH are found in the intestinal and gastric mucosa, lungs, kidney, spleen, prostate, testis, and retina; this extra-hepatic ADH activity may play a minor metabolic role, which could increase during excessive or long-continued use (98). Alcohol is oxidized and eliminated at a fairly constant rate. The First Special Report to the U. 8S. Congress on Alcohol and Health (99) states "In a 150-pound man, alcohol is metabolized at approximately one drink per hour." A typical drink is de- fined as one containing 3/4 oz. (17.5 g) of ethanol, the amount in 1.9 oz. of 80-proof spirits. If, therefore, a man weighing 1038, , 70 kg (154 1bs.) drank this amount every hour and there was a prompt absorption, the blood alcohol level should theoretically be zero at the end of each hour. That pattern of drinking, of course, bears little re- semblance, 40 what happens in real life. Studies in man by Mezey and Febon (100) have shown that the average metabolism is about .15 g/kg/hr, or 252 g in 24 hours in a 70 kg man. Newman et al. (101, 102), on the basis of limited studies in dogs and humans, concluded that the maximal amount of alcohol that can ordinarily be metabolized is about 300 g per day. Beerstecher (103) and Rohan (104) cite instances, however, in which actual daily intake has exceeded this amount for months or years. 86 Blood alcohol levels reflect the amount ingested, absorption from the gut, distribution in body water and metabolism. Early studies (105) showed that after rapid ingestion of 22 g ethanol in the form of undiluted whiskey or a martini on an empty stomach, the blood level rises to about 45 mg/100 ml in about 30 minutes, then begins to fall, becoming undetectable at 3 hours. After ingestion of 44 g in the same form, blood levels reach about 80 mg/100 ml in 30-45 minutes, then fall steadily but are usually detectable for 6 hours, with small amounts remaining in the stomach during most of this period. There is considerable individual variation, due mainly to unequal absorption from the stomach. Mellanby (106), using dogs, and Haggard et al. (107), employing human subjects, found that when the same amount of alcohol is diluted or taken with or after a meal, blood levels reached are not as high, probably owing to slower absorption; elimination, however, occurs at about the same rate. The essential validity of these observations has been confirmed by many workers. Within wide ranges, therefore, blood levels can be related to type of intake and time. However, intake and blood levels are more difficult to correlate consistently with acute effects. The phenomenon of tolerance to alcohol is not well under- stood, although it is an important element in the body's reaction to alcohol and may be a key factor in permitting the daily ingestion of very large amounts without apparent harm for many years. The linkage of tolerance to physical dependence and its attendant withdrawal syndrome is likewise not well understood. Very few factors are capable of increasing the rate of alcohol metabolism. However, it seems well established that 8/7 chronic alcoholics metabolize alcohol faster than normal individuals. Amino acids (especially alanine) and fructose also enhance ethanol metabolism, but their clinical utility is limited. Alcohol also reduces the intestinal absorption of nutrients such as glucose, amino acids, calcium, folate, and vitamin Bis: This inhibition of absorption may contribute to the malnutrition frequently found in alcoholic subjects. Starvation slows the rate of alcohol metabolism in the liver, although this varies greatly in degree from one person to another. Pharmacological Effects of Ethanol (108) Cardiovascular System. There appears to be a direct action of alcohol on the excitability and contractility of heart muscle. With intoxicating doses there is a rise in cardiac rate and output and in systolic and pulse pressures, and a cutaneous vasodilatation at the expense of splanchnic constriction. Some authors have stated that prolonged intoxication may have a damag- ing effect on cardiac and skeletal muscle, a degeneration of fibers supposedly due to suppression of myophosphorylase activity. Increased sweating and vasodilatation cause a loss of body heat and a fall in body temperature. Stomach. In low concentration ({10%) by whatever route it is administered, alcohol stimulates the gastric glands to pro- duce acid, apparently by antral activation of gastrin release, and possibly by causing the tissues to form or release histamine. With the ingestion of alcohol in concentrations of over 10 to 15 percent, the secretion of mucus is increased, the stomach 838 mucosa becomes congested and hyperemic, and the secretion of acid may then become depressed. This is a state of acute gastritis, from which recovery may be relatively rapid. The increase in appetite following ingestion of alcohol is due to the stimulation of the end organs of taste and to a general sense of well-being. Metabolic Effects. Alcohol has a number of metabolic effects. In the area of lipid metabolism it can cause hyper- triglyceridemia as well as lead to a fatty liver. It inter- feres with carbohydrate metabolism and can produce hypoglycemia by impairing gluconeogenesis; however, a significant degree of hypoglycemia will occur only if hepatic glycogen stores are depleted. Under certain conditions, alcohol can also interfere with the peripheral utilization of glucose and produce hyperglycemia. When ethanol is oxidized, there is a simultaneous generation of reduced nicotinamide adenine dinucleotide; as a result pyruvate is converted to lactate. Thus alcoholism may result in increased levels of serum lactate, occasionally lactic acidosis and also hyperuricemia secondary to the in- hibitory action of lactic acid in the renal excretion of uric acid. Renal System. Alcohol drinkers frequently exhibit low serum levels of phosphate and magnesium presumably because of increased renal excretion of these ions. There are also well- recognized effects on water excretion. The ingestion of 4 oz. of 100-proof bourbon whiskey may result in a diuresis comparable to that which follows the drinking of large amounts of water. 89 This diuresis is most likely due to the transient suppression of the release of antidiuretic hormone from the supraopticohypo- physeal system. However, diuresis occurs only during the initial phase of alcohol administration and does not persist during prolonged drinking. There is also increased urinary excretion of ammonium and a titratable metabolic and respiratory acidosis. The former is presumably due to an accumulation of acid metabolites and the latter to the direct action of alcohol on the respira- tory center. Nervous System. It is now generally accepted that alcohol is not a central nervous system (CNS) stimulant, but a depress- ant. Some of the early and apparent stimulatory effects of alcohol, manifested as garrulousness, agressiveness, excessive activity, and increased electrical excitability of the cerebral cortex, are due to the inhibition of certain subcortical struc- tures (high brainstem reticular formation) which ordinarily modulate cerebral cortical activity. Similarly, the initial hyperactivity of tendon reflexes may represent a transitory escape of spinal motor neurons from higher inhibitory centers. With increasing amounts of alcohol, however, the depressant action spreads to involve the cerebral neurons as well as other brainstem and spinal neurons. All types of motor performance, whether the simple main- tenance of a standing posture, the control of speech and eye movements, or highly organized and complex motor skills, is 90 adversely affected by alcohol. The movements involved in these acts are not only slower than normal but also more inaccurate and random in character and therefore less adapted to the accomplishment of specific ends. Alcohol also impairs the efficiency of mental function by interfering with the learning process, which is slowed and rendered less effective. The facility of forming association, whether of words or figures, tends to be hampered and the power of attention and concentration is reduced. The person is not as versatile as usual in directing thought along new lines appropriate to the problems at hand. Finally, alcohol impairs the faculties of judgment and discrimination and, all in all, the ability to think and reason clearly. A scale relating the various degrees of clinical intoxica- tion to the blood alcohol levels in nonhabituated persons has been constructed. At blood alcohol levels of 30 mg per 100 ml, a mild euphoria exists, and at 50 mg per 100 ml, a mild incoordina- tion. At 100 mg per 100 ml, ataxia is obvious; at 300 mg per 100 ml, the patient becomes stuporous; and a level of 400 mg per 100 ml is accompanied by deep anesthesia and could prove fatal. These figures are valid, provided that the alcohol content rises steadily over a 2 hour period. It should be emphasized, however, that such a scale has no value in the chronic alcoholic. It does not take into account two adaptive changes that every organism makes to alcohol: an increased rate of alcohol metabolism by the liver and the development of tolerance. N It is common knowledge that an habituated individual can drink more alcohol and show fewer of its nervous system effects than a moderate drinker or abstainer. In the chronic alcoholic subject the ingestion of a given amount of alcohol will result Vv in a lower blood alcohol level than in a non-alcoholic individual; furthermore, for a given blood alcohol level one will observe lesser degrees of "drunkenness" or inebriation. The organism is capable of adapting to alcohol after a very short exposure. If the alcohol concentration in the blood is raised very slowly, no symptoms appear, even at quite high levels. It would appear that the important factor in this rapid adaptability is not so much the rate of increment or the height of the blood alcohol level, but the length of time the alcohol has been present in the body. It has also been shown that if the dosage of alcohol which causes blood levels to reach a certain height is held constant, the blood alcohol concentration falls and clinical evidence of intoxication disappears. The cause of this fall in alcohol concentration is not clear. This type of tolerance has been termed "metabolic" and refers to the adjustments made by the nervous system to long-continued exposure to alcohol. There must be a subtle alteration in the metabolism of the neurons such that they can function in the face of high tissue alcohol levels. This alteration requires some time for its establishment. Removal of alcohol from the habituated nervous system results in another disturbance in neuronal function, presumably an overactivity. Pharmacogenetics of Ethanol Theoretically, genetic factors might influence both the metabolism of alcohol and the effects of alcohol on physiological functions. It has not yet been determined to what extent alcohol metabolism is conditioned by genetic factors. According to von Wartburg and Schurch (109), alcohol dehydrogenase, the enzyme which catalyses alcohol metabolism, occurs in two forms, a typical and an "atypical" form. The frequency of these two types was investigated in various population groups and found to vary greatly. In Sweden, the "atypical" form was present in 4%, in Switzerland in 20% and in Japan in 90-95% of the population, The significance of the difference between countries is obscure. Fenna et al. (110) observed that the rate of alcohol metabolism was more rapid in white individuals than in Indians and Eskimoes. Similar racial differences have been found in the field of pharmacogenetics. In glycogenosis type I, i.e., an "inborn error of metabolism," the alcohol elimination rate was found to be three to four tines more rapid than normal. The absence since individuals lacking this enzyme have so far not been encountered. The investigation of ethnic differences in alcohol sensitivity carried out by Wolff is of special interest (111). He gave alcohol to young and adult Japanese, Taiwanese, Koreans and Caucasians and found that their responses to alcohol showed great variations: of the white children only 5% ry 93 responded with marked flushing of the face as compared with 74% of the Asian children. This also applied to adults in which great differences were also noted in their heart rates and in subjective intoxication. Pharmacokinetics and Pharmacologic Effects of Ethanol during Pregnancy The information on the pharmacokinetics and pharmacological effects of ethanol described above were obtained in adult males and less frequently in nonpregnant females. Whether these pharmacological effects are fully valid when applied to preg- nant females remains to be assessed. Significant physiological changes take place during the course of human pregnancy (112, 113, 114). Alternations in maternal weight and increased extra- and intracellular fluid are not due to the growing conceptus alone, but are also attributable to profound changes in maternal physiology (Figure 1, Table 5). TABLE 5 FIG.1. Components of weight gain in 12 normal pregnancy. (Hytten FE, Leitch I: - Maternal The Physiology of Normal Pregnancy. 10 stores Oxford, Blackwell, 1964) J Tissue 8 fluid = 2 Blood 6 { Uterus | Breasts 4H 3 I: Tz 8 (Fetus a Es Sy (Paces 2 - y 58 \Liquor amnii 10 20 30 40 Weeks of pregnancy TABLE 5. Estimate of Extracenular and Intracellular Water Added During Pregnancy Total water (ml) Extracellular (ml) Intracellular (ml) Fetus 2343 1360 983 Placenta 540 260 280 Liquor amnii 792 792 0 Uterus 743 490 53 Mammary gland 304 148 156 Plasma 920 920 0 Red cells 163 0 163 Extracellular extravascular water 1395 1195 0 Total 7000 5165 1835 Hytten FE, Leitch I: The Physiology of Human Pregnancy. Oxford, Blackwell, 1964 116 95 Thus, the physiological adaptive mechanisms of pregnancy involve marked changes in the respiratory, cardiovascular, gastrointestinal and urinary systems of the mother as well as metabolic changes (Table 6). Table 6: SUMMARY OF MEASURABLE PREGNANCY CHANGES Parameter Respiratory System Tidal volume Resistance of tracheobronchial tree Expiratory reserve Residual volume Functional residual capacity Respiratory minute volume Cardiovascular System Heart rate Stroke volume Cardiac output Peripheral blood flow Blood volume Gastronintestinal System Cardiac sphincter tone Acid secretion Motility Gallbladder emptying Urinary Tract Renal plasma flow Glomerular filtration rate Ureter tone Metabolism Oxygen consumption Percent* Increase Decrease 30-40 36 40 40 25 40 0-40 X 20-30 600 48 X X X X 25-50 50 Xx 14 * change from non-pregnant values 96 These maternal changes are likely to modify the physiologic disposition of, and response to, ethanol so that the pharmacologica. effects of drinking equivalent amounts of alcohol may differ in pregnant versus nonpregnant females. It is unfortunate that little, if any, information is available on this subject. SECTION B-1, b: CLINICAL MANIFESTATIONS OF ALCOHOLISM (115) Although alcohol may alter the function of practically every organ system, the most important clinical effects are on the digestive organs and on the nervous system. Effect on the Digestive Organs Gastritis. Morning nausea and vomiting are symptoms common in alcoholics. Characteristically, the patient can suppress these symptoms by taking a drink or two, after which he is able to consume large quantities of alcohol without the recurrence of the symptoms until the following morning. Other complaints referable to the gastrointestinal system are abdominal distention, epigastric distress, belching, typical or atypical ulcer symptoms, and hematemesis. The most common pathologic basis for these symptoms is a superficial gastritis, which is an almost invariable sequel to prolonged drinking. Most instances of gastritis are benign, and the symptoms sub- side after a few days of abstinence, but more severe forms are associated with mucosal erosions or ulcerations and may be the source of serious bleeding. The incidence of peptic ulcer is exceptionally high in alcoholics. A less frequent but serious cause of hematemesis is the so-called Mallory-Weiss syndrome, which is characterized by lacerations of the mucosa at or just below the gastroesophageal junction. Hepatitis. Patients admitted to the hospital following a period of prolonged drinking and severe dietary depletion almost invariably show enlargement of the liver because of infiltration of the parenchymal cells with fat. This fatty liver is essentially reversible provided that the patient remains abstinent and receives a nutritious diet. A form of hepatocellular necrosis or alcoholic hepatitis is observed frequently in chronic alcoholics, especially following a severe drinking bout. About 8 percent of patients with severe alcoholism develop a permanent form of liver disease, i.e., cirrhosis, in which a diffuse proliferation of fibrous tissue replaces the normal lobular architecture of the organ. Pancreatitis. The excessive use of alcohol is also a significant factor in the causation of pancreatitis. The mildest form of this disorder may be attributed to gastritis Or may go unnoticed. In more severe form, pancreatitis presents as an acute abdominal catastrophe, i.e., with epigastric pain, vomiting, and rigidity of the upper abdominal muscles. In these circumstances the pancreas appears tense and edematous, often with a serosanguinous exudation of fluid on its surface. The most severe form is that of hemorrhagic pan- creatitis. Alcoholics may also develop a chronic relapsing form of pancreatitis, often associated with irregular calcification of the pancreas. 98 Effect on the Nervous System A large number of neurologic disorders are associated with alcoholism, as listed in Table 7. The factor common to all of them is the abuse of alcohol, but the mechanism by which alcohol produces its effects varies widely from one group of disorders to another. 11. 1xt. Iv. Table 7 NEUROLOGIC DISORDERS ASSOCIATED WITH ALCOHOLISM Alcohol intoxication - drunkenness, coma, excitement ("pathologic intoxication") The abstinence or withdrawal syndrome - tremulousness, hallucinosis, "rum fits," delirium tremens Nutritional diseases of the nervous system secondary to alcoholism A. Wernicke - Korsakoff syndrome B. Polyneuropathy C. Optic neuropathy ("tobacco-alcohol amblyopia") D. Pellagra Diseases of uncertain pathogenesis associated with alcoholism A. Cerebellar degeneration B. Marchiafava-Bignami disease C. Central pontine myelinolysis D. Cerebral atrophy E. "Alcoholic" cardiomyopathy and myopathy Neurologic disorders consequent upon Laennec's cirrhosis and portal-systemic shunts A. Hepatic stupor and coma B. Chronic hepatocerebral degeneration 99 I. Alcohol Intoxication. The signs of drunkenness consist of varying degrees of exhilaration and excitement, loss of restraint, irregularity of behavior, loquacity, slurred speech, incoordination of movement and gait, irritability, drowsiness, and, in advanced cases, stupor and coma. On rare occasions acute intoxication is characterized by an outburst of irrational, combative and destructive behavior, which terminates when the patient falls into a deep stupor and for which he may later have no memory. This state has been referred to as "pathologic intoxication" or "acute alcoholic paranoid state." Allegedly this reaction may follow the ingestion of relatively small amounts of alcohol, and it has been variously ascribed to constitutional differences in susceptibility to alcohol, previous cerebral injury, and "an underlying epileptic predisposition." However, there are no critical data to support any of these contentions. Alcohol acts on nerve cells in a manner akin to the general anesthetics. Unlike the latter, however, the margin between the dose of alcohol that produces surgical anesthesia and that which dangerously depresses respiration is a very narrow one, a fact which accounts for the occasional fatality in cases of alcoholic narcosis. Alcoholic coma is characterized by serious depression of respiration, heralded by a loss of corneal and pupileary reflexes. In fatal cases periperal vascular collapse preceeds death. 1100 II. The Abstinence or Withdrawal Syndrome. A second category of alcoholic neurologic disease comprises the tremulous, hallucinatory, epileptic, and delirious states. Although a sustained period of chronic intoxication is the underlying factor in each of these disorders, the symptoms become manifest only after a period of relative or absolute abstinence from alcohol - hence the designation abstinence or withdrawal syndrome. The prototype of the patients afflicted with these symptoms is the binge or periodic drinker, although the steady drinker is not immune if, for some reason, he or she stops drinking. By far the most common manifestation of the abstinence syndrome is a state of tremulousness, commonly referred to as "the shakes" or "the jitters" combined with general irritability and gastrointestinal symptoms, particularly nausea and vomiting. The symptoms first show themselves after several days of drinking, usually in the morning, after the short period of abstinence that occurs during sleep. Generalized tremor is an outstanding feature of this illness. It is of fast frequency (6 to 8 oscillations per second), slightly irregular, and variable in its severity, tending to diminish when the patient is in quiet surroundings and to increase with motor activity or emotional stress. The tremor may be so violent that the patient cannot stand without help, speak clearly, or feed himself. Although the flushed facies, anorexia, tachycardia, and tremor subside to a large extent within a few days, the patient does not regain his full composure for a much longer time. The overalertness, tendency to startle easily, and jerkiness of movement may persist for a week or longer; the feeling of uneasiness may not leave the patient completely for 10 to 14 days. Symptoms of disordered perception (hallucinosis) occur in about one-quarter of the tremulous patients. The patient may complain of "bad dreams" -- nightmarish episodes associated with disturbed sleep, which are difficult to separate from real experience. Sounds and shadows may be misinterpreted, or familiar objects may be distorted and assume unreal forms. Although these are not hallucinations in the strict sense of the term, they represent the most common forms of disordered sense perception in the alcoholic. Hallucinations may be purely visual or auditory in type, mixed visual and auditory, and occasionally tactile or olfactory. There is little evi- dence to support the popular belief that certain visual hallu- cinations are specific to alcoholism. They are more commonly animate than inanimate and may comprise various forms of human, animal, or insect life. In this particular setting (i.e., where relative or absolute abstinence follows a period of chronic inebriation) there is a marked tendency to develop convulsive seizures (withdrawal seizures or "rum fits"). Over 90 percent of 102 seizures occur during the 7- to 48-hour period following the cessation of drinking, with a peak incidence between 13 to 24 hours. During the period of seizure activity the electro- encephalogram may be abnormal, but it reverts to normal in a matter of days, even though the patient may go on to develop delirium tremens. Almost one-third of the patients with general- ized seizure activity go on to develop delirium tremens, in which case the seizures invariably precede the delirium. Delirium tremens is the most dramatic and grave of all the alcoholic complications. It is characterized by profound confusion, delusions, vivid hallucinations, tremor, agitation, and sleeplessness, as well as by increased activity of the autonomic nervous system, i.e., dilated pupils, fever, tachy- cardia, and profuse perspiration. Delirium tremens develops in one of several settings. The patient, an excessive and steady drinker of many years' duration, may have been admitted to the hospital for an unrelated illness, accident, or operation, and 3 to 4 days later becomes delirious. Or following a prolonged spree, he may have already experienced several days of tremulousness and hallucinosis, or one or more seizures, and may even be recovering from these symptoms, when he suddenly develops delirium tremens. In the majority of cases delirium tremens is benigh and short-lived, ending as abruptly as it begins. Consumed by the relentless activity and wakefulness of several days' duration, 103 the patient falls into a deep sleep; he awakens lucid, quiet, and exhausted, with virtually no memory for the events of the delirious period. About 15 percent of cases of delirium tremens, as defined above, end fatally. In many of these there is an associated infectious illness or injury, but in a few no complicating illness is discernible. Patients frequently die in a state of hyperthermia or peripheral circulatory collapse; in some, death comes so suddenly that the nature of the terminal events cannot be determined. IIT. Nutritional Diseases of the Nervous System. The Wernicke- Korsakoff syndrome, alcoholic polyneuropathy, optic neuropathy ("tobacco-alcohol amblyopia") and pellagra are considered under this heading. These diseases compose a relatively small but serious group of illnesses in chronic alcoholics. In contrast to the role of alcohol in intoxication and abstinence syndromes, its role in these nutritional diseases is purely secondary, serving mainly to displace food in the diet. IV. Alcoholic Diseases of Uncertain Pathogenesis. Included in this category are several diverse disorders which are prac- tically always encountered in alcoholic patients. Their relationship to the excessive use of alcohol is not fully understood and probably is not crucial, since all of them have been described in nonalcoholic patients. There is a considerable amount of indirect evidence that these disorders 104 are nutritional in origin, but as yet this relationship must be regarded as unproved. Alcoholic Cerebellar Degeneration. This term is applied to a nonfamilial type of cerebellar ataxia which occurs in adult life against a background of prolonged ingestion of alcohol. The symptoms frequently evolve in a subacute fashion (several weeks or months), after which they remain stationary for many years. The signs are those of cerebellar dysfunction, affecting stance and gait predominantly. Once established, the signs change very little, although some improvement of gait (due mainly to recovery from complicating polyneuropathy) may follow cessation of drinking. The essential pathologic changes consist of degeneration of varying severity of all the neurocellular elements of the cerebellar cortex, particularly of the Purkinje cells, with a striking topographic restriction to the anterior and superior aspects of the vermis and hemispheres. Marchiafava-Bignami Disease (Primary Degeneration of the Corpus Callosum). This is a rare complication of alcoholism. The symptoms are diverse and include pyschic and emotional dis- orders, delirium and intellectual deterioration, convulsive seizures, and varying degrees of tremor, rigidity, paralysis, apraxia, aphasia, and sucking and grasping reflexes. The duration is variable, from several weeks to months, and recovery is possible. The pathologic picture consists of symmetrically placed areas of demyelination in the corpus callosum, particu- larly the middle lamina. 105 Pontine Myelinolysis. This term refers to a unique pathologic change affecting the center of the basis pontis, in which the medullated fibers are destroyed in a single symmetric focus of varying size. The disease may manifest itself by pseudobulbar palsy and quadriplegia, but usually the lesion is so small that it causes no symptoms and is found only at postmortem examination. The relationship of this condition to either alcoholism or malnutrition is obscure, but most of the cases have occurred in patients with prolonged and severe nutritional depletion. Cerebral Atrophy. The pathologic examination of relatively young alcoholic patients not infrequently discloses an unexpected degree of convoluntional atrophy, most prominent in the frontal lobes, and a symmetric enlargement of the lateral and third ventricles. The ventricular enlargement may also be found on pneumoencephalography. In some patients these findings are associated with overt complications of alcoholism, such as the Wernicke-Korsakoff syndrome, but in many of them no other abnormalities can be found, and the history discloses no symptoms of neurologic disease. The nature of this disorder is quite unclear. "Alcoholic" Myopathy. Attention has been drawn to several disorders of skeletal and cardiac muscle, apparently primary in nature, in association with chronic alcholism. One type of myopathic syndrome, which may be generalized or focal, is characterized by the acute onset of severe pain, tenderness, 106 and edema of muscles, accompanied by myoglobinuria, renal damage, and hyperpotassemia in severe cases. In other cases, diffuse muscle weakness is associated with hypopotassemia and vascular necrosis of muscle. It should be noted that these acute forms of muscle weakness are not confined to alcoholics. Another type of myopathy is characterized by the subacute development of weakness and atrophy of the proximal limb and girdle muscles, with "myopathic" changes in the electromyogram and elevated creatine phosphokinase levels in serum, but without local pain or edema. Muscle power is slowly restored in these patients following abstinence from alcohol and improvement in nutrition. That this disorder represents a primary affection of muscle has not been established beyond doubt. Muscle biopsies from such patients suggest that it may represent a proximal form of polyneuropathy, despite relatively mild clinical signs of peripheral nerve disease. "Alcoholic cardiomyopathy" is the name given to a nonspecific affection of cardiac muscle which has a higher incidence in patients with chronic alcoholism than in the non-alcoholic popu- lation. The role of alcohol, malnutrition, or some hitherto unsuspected factor in the genesis of these disorders is not known, and their structural and biochemical bases requires further study. 107 V. Neurologic Disorders Consequent Upon Cirrhosis and Portal- System Shunts. Hepatic coma refers to an episodic disorder of consciousness which frequently complicates (or terminates) advanced liver disease and/or portal-system shunts. It is associated with typical electroencephalographic abnormalities and intermittency of sustained muscular contraction which presents as an irregular flapping movement of the outstretched limbs (asterixis). Less frequently, cirrhosis is complicated by a chronic and largely irreversible form of hepatocerebral disease, the main symptoms of which are dementia, dysarthria, ataxia, and athetosis. A failure to metabolize ammonium, or perhaps some other substance absorbed from the bowel, may be the result of either hepatocellular disease or of shunting of blood around the liver. Presumably, the acute and rapidly developing effect of this toxin on the brain is episodic stupor or coma, which is reflected pathologi- cally by a diffuse astrocytic hyperplasia in the CNS; a prolonga- tion of this effect may lead to irreversible neurologic symptoms and parenchymal lesions. Section B-1, c: PHARMACOLOGICAL AND TOXICOLOGICAL PRINCIPLES OF ALCOHOL CONSUMPTION AND PREGNANCY OUTCOME. This brief review emphasizes that the pharmacological and medical implications of alcohol consumption are complex. The historic difficulty in clearly distinguishing the alcohol user from the alcohol abuser, and the lack of a generally accepted definition of 108 alcoholism (116) is an indication of this complexity. Alcohol drinkers differ from each other in a number of respects, including amount of ethanol ingested on each occasion, pattern of drinking, length of alcohol consumption, and dependence and tolerance to alcohol. As a result of these differences, alcohol use may result in acute recurrent intoxications or in a number of chronic, some of them progressive, disorders with different degrees of social and health ramifi- cations. Hore (117) describes two phases in drinking behavior: a) the pre-dependent phase, where drinking behavior is best understood as the result of a combination of psychological and social factors, and in which the sequelae of physical dependence are absent, and b) the dependent phase proper, with evidence of physical dependence. Since most would agree that alcoholism is defined as dependence (addiction) to alcohol, then the pre-dependent phase has been termed "normal", or "social", drinking. Jellinek (116) however, differentiates five species of alcoholism: alpha - representing a purely psychological con- tinued dependence without loss of control or inability to abstain; beta - that form in which physical complications occur without either physical or psychological dependence; gamma - alcoholism in which there is acquired tissue tolerance, adaptive cell metabolism, physical dependence and loss of control; delta - alcoholism with inability to abstain instead of loss of control; and episilon alcoholism (dipsomania) - which is periodic alcoholism. 109 These classifications and the disease concept of alcoholism have been the subject of much debate (118), clearly pointing out that "alcoholics" are an ill-defined group which is difficult to identify. Furthermore, since they may differ in a number of respects from each other, they are an heterogeneous popu- lation and therefore difficult to characterize. Except for extreme cases of chronic alcoholism, the alcoholic (alcohol addict) can hardly be differentiated from the "social drinker" (non-alcohol addict). The alcoholic is rarely identified by physicians or other health or law officials before the disease becomes a significant medical or social problem (117). In examining the pregnancy outcome of women using (or abusing) alcoholic beverages, careful consideration should be given to the multiple pharmacological, toxicological, medical and social implications of alcohol use. This problem has been considered and attempts have been made to better characterize alcohol drinking mothers with poor pregnancy outcome. For example, Jones et al. (2), in reporting 8 cases of FAS in children of chronic alcoholic mothers, attempted to better define the degree of maternal alcoholism by examining the duration of maternal alcoholism and the presence or absence of delirium tremens, cirrhosis and nutritional anemia (Table 8). 110 TABLE 8: Maternal History of Alcoholism in Eight Mothers with FAS Children (2) History of alcoholism Patient Number All cases 1 2 3 4 5 6 7 8 Duration (years) 7 3 4 11 2+ 10 23 15 9.4% Delirium tremens + ? + + ? - + + 5/6 Cirrhosis - ? - + ? - + - 2/6 Nutritional anemia - ? - + ? - + - 2/6 *Mean + = present; - = absent; ? - unknown The clear delineation of the degree of duration of maternal alcoholism may be very important in assessing the role played by alcohol in the FAS. This may be particularly true in assessing the possible teratogenicity of "low" doses of ethanol. Thus, from a pharmacological and toxicological point of view, the effects of drinking 2 or 3 ounces of absolute alcohol per day markedly differs, depending upon the number of doses (drinks) in which this amount is administered during the day, the type of alcoholic beverage used, the degree of maternal alcohol toler- ance, and whether this amount of alcohol is ingested during meals or on an empty stomach. As pointed out by Streissguth (119), the identification and detailed description of the drinking woman are important aspects of any study involving drinking habits in the mothers and pregnancy outcome. It may very well be that the method of classifying women into arbitrary groups (e.g., light, moderate, and heavy drinkers) by determining only the average daily alcohol intake will prove too insensitive for establishing a clear relationship between alcoholic beverage consumption in women and poor pregnancy outcome. 111 SECTION B-2: TERATOLOGIC PRINCIPLES An evaluation of the data presented in Section A reveals that there is a relationship between alcoholic beverage consumption by the mother and poor pregnancy outcome. Although this relationship has been suspected for a long time, only recently it has been fimmly established by the outstanding contributions of Iemoine et al. (1) and Jones et al. (2). It was through the efforts of these authors, working independently, that clinical detection of the toxicity of alcohol consumption was made possible, because they grouped together a series of rather aspecific birth defects into a well-defined teratologic syndrome which they termed the FAS. Listed in Table 9 are the criteria for recognizing a new teratogenic agent in man; the first criteria is an abrupt increase in the incidence (or recognition) of a particular defect or syndrome. There are now well over 200 cases of FAS reported in the literature. Since almost all the cases have been reported in offspring of chronic alcoholic women, it is clear that heavy and prolonged alcohol beverage consumption by the mother fulfills the first criterion for recognition of teratogenicity in the human. However, this criterion has not veh been met for moderate alcoholic beverage consumption, e.g., that not associated with medically recognizable signs of maternal chronic alcoholism. Although there are still some (120) who question ethanol as the etiologic agent causing growth and behavior Aisonders in the offspring of female alcohol abusers, there is no doubt that ethanol remains the only common consumption factor in all reported cases of FAS (121). Thus, according 112 to Clarren and Smith (4), whatever the basic biochemical mechanism of FAS teratogenesis is, alcohol beverage consumption plays a key role. Although criterion 1 of Table 9 is clearly met, criterion 2 does not really apply to ethanol, and the available clinical evidence is insufficient to draw any conclusions for criteria 3 and 4. TABLE 9. CRITERIA FOR RECOGNIZING A NEW TERATOGENIC AGENT IN MAN (122). 1. An abrupt increase in the incidence of a particular defect or association of defects (syndrome). 2. Coincidence of this increase with a known environ- mental change, e.g., widespread use of a new drug. 3. Known exposure to the environmental change early in pregnancies yielding characteristically defective infants. 4. Absence of other factors common to all pregnancies yielding infants with the characteristic defect (s). Thus, it is not known whether maternal exposure to alcohol during early pregnancy yields characteristically defective infants. It has been suggested that "binge" drinking in early pregnancy may be responsible for alcohol tera- togenicity, but as far as I know, there are no human data supporting such an hypothesis. Further, it is not known whether pre-pregnancy chronic ethanol exposure with or without early (or late) pregnancy exposure is an important factor in the teratogenicity of alcohol in the human. 113 As far as criterion 4 is concerned, all other factors known to be associated with alcoholic women (e.g., mal- nutrition, drug use, cigarette smoking, etc.) must be eliminated as possible etiologic and/or contributing factors in the FAS. As will be discussed later in the epidemiology section (Section B-3), the relative importance of these associated factors is unclear, although active investigations are under way to elucidate this difficult area. The task of evaluating the potential teratogenicity of "low" (or "moderate") doses of alcohol during pregnancy is particularly difficult. As mentioned before, the full- blown FAS is characteristic only of infants whose mothers have been alcoholic for a prolonged period (usually years) before gestation. Therefore, attempts have been made to relate alcohol consumption by the mother to other less specific end-points related to alcohol embryotoxicity including stillbirth, decreased birth weight, prematurity, neonatal survival, and congenital malformations. No clear conclusions have as yet emerged. An intrinsic difficulty of these studies is that all the commonly measured signs of poor pregnancy outcome are aspecific and known to have a multifactorial etiology (112). As far as FAS is concerned, Clarren and Smith caution that any specific listing of individual features of the FAS which are thought to be essential to the FAS diagnosis could be arbitrary and mis- leading. Thus, they have been reluctant to positively identify an individual as affected by ethanol without some alteration in brain function, growth, and facial appearance. They stress that until more knowledge has been accumulated, less than complete expression of FAS can only be referred to as "suspected" fetal alcohol effects (4). In Section A-3, the reproductive toxicity of ethanol in laboratory animals has been reviewed. Since it is important to consider the potential teratogenicity of alcohol in the human in the light of experimental evidence accumulated in laboratory animals, a systematic discussion of these data in relation to well established teratologic principles is in order. According to Wilson (123), there are six accepted general principles of chemical teratogenesis that have evolved from animal studies. Principle 1. Susceptibility to Teratogenesis Depends on the Genotype of the Conceptus and the Manner in Which This Genotype Interacts with Environmental Pactors. It is well known in experimental teratology that embryo- toxic effects are species and strain specific. For example, mouse embryos are usually susceptible to cleft palate induction by glucocorticoids, whereas most other mammalian embryos are not. This finding has been interpreted to mean that mice possess inborn chemical or anatomical features which make themmore vulnerable (less resistant) to these agents than are other animals. This susceptibility could be related to such factors as the rate at which the hormone is absorbed, eliminated, or transformed by the maternal animal; its rate of passage across the placenta; or the nature of 115 its interactions within the cells and tissues of the embryo. Regardless of the mechanism by which the mouse is rendered more susceptible to the teratogenicity of glucocorticoids, it is clear that this sensitivity is, at least to some extent, genetically determined. The thalidomide episode exemplifies the differences which can exist among species with regard to susceptibility to a teratogen. Man and other higher primates are extremely sensitive to thalidomide teratogenicity, showing a characteristic series of reduction defects of limbs and face; some rabbit and a few mouse strains react in a less characteristic way to much larger doses, whereas most other mammalian forms are quite resistant. Other intra- as well as inter-strain differences in sensitivity to teratogenic agents (not to mention inter- species differences) are so well known as to require no further discussion here. It only need be emphasized that most such differences probably have an hereditary basis, although the precise nature may not be known because the metabolic and structural sites of action of teratogenic agents are often not known. These considerations must be taken into account when attempting to extrapolate results from laboratory animals to the human situation. It is clear then, that results on the embryotoxicity of ethanol in laboratory animals cannot necessarily be used to establish the risk that this substance poses to the human fetus. 116 Principle 2. Susceptibility to Teratogenic Agents Varies with the Developmental Stage at the Time of Exposure. It is a basic precept of biology that immature or developing organisms are more susceptible to change than are mature or fully developed ones. The "critical mements" concept of Stockard (124), has today become more generally known as "critical periods" in organogenesis. This critical period has been shown to be the time when the embryo is most susceptible to the induction of gross anatomical defects by environmental influences. Thus, the early events in organ formation have generally been found to be more sensitive to interference from extraneous influences than later ones, but "critical periods" are known to occur even during relatively late stages in the formation of some organs and systems. As the more basic organizational patterns of organ formation are laid down, the likelihood of major structural deviation naturally diminishes, and when definitive form and relationships are achieved, maldevelopment in the sense of gross defects is no longer of concern. The gross modeling that occurs during organogenesis, however, does not equip the organ to assume its ultimate functional role. The processes of cellular and tissue differentiation continue on more refined scales through histogenesis. Histogenesis generally begins before organogenesis is completed and continues well into the subsequent growth phase of most organs. Harmful influences during histogenesis would not result in gross malformations but could certainly 117 result in finer structural defects at the microscopic level. In close sequence with the progress of histogenesis is the acquisition of function. In fact a level of functional activity begins in many organs while histogenesis is in progress, but definite function is usually not achieved until histological differentiation is completed. It is thus not always possible to determine whether an agent causing functional abnormality has acted by interfering with ongoing histogenesis or has prevented the final step in functional maturation. As far as ethanol teratogenicity is concerned, experiments in animals have shown that ethanol is teratogenic, embryolethal and exerts growth retarding effects when given to pregnant laboratory animals during organogenesis. We do not know, however, whether if the human the spectrum of adverse effects (e.g., CNS deficiences, growth retard- ation, facial abnormalities and other malformations) a able to alcohol ingestion are produced during one specific "sensitive period" or whether each individual abnormality has its own period of sensitivity at a specific stage of gestation. Principle 3. Teratogenic Agents Act in Specific Ways (Mechanisms) on Developing Cells and Tissues to Initiate Abnormal Embryogenesis (Pathogenesis). Early studies in experimental teratogenesis were often interpreted as showing that particular agents produced char- acteristic types of malformations. Such associations of patterns of defects with causative factors were sometimes referred to as "agent specificity", implying that each agent had the capability to elicit specific defects. As more and more causative factors were described, however, it became apparent that the composition of the patterns of defects overlapped, i.e., that different agents often produced some of the same types of defects, or the same range of defects in different proportions, or sometimes even the same defects in similar incidences. If each causative factor did not produce a distinctive array of defects, there must be some commonality in the developmental pathways between the induction, by whatever means, and the final expression of abnormality (123). Evidence has now been assembled from the available teratological literature to support the existence of some nine ways or mechanisms by which teratogenic agents impinge on developing systems to initiate abnormal development. These proposed mechanisms of initial changes are listed in Table 10. TABLE 10. INITIAL TYPES OF CHANGES IN DEVELOPING CELLS OR TISSUES AFTER TERATOGENIC INSULT 1. Mutation (Gene) 2. Chromosomal breaks, nondisjunction 3. Mitotic interference 4. Altered nucleic acid integrity or function 5. Lack of normal precursors, substrates, etc. 6. Altered energy sources 7. Changed membrane characteristics 8. Osmolar unbalance 9. Enzyme inhibition 119 At the present time, the mechanism(s) of alcohol embryotoxicity is not known, and it would be inappropriate to advance any hypothesis in this regard. However, it is important to realize that alcohol embryotoxicity may produce developmental abnormalities which are not characteristic of ethanol alone. Thus, different chemicals acting by different mechanisms may induce the same types of abnormalities. As mentioned above, all the abnormalities of the FAS, taken singly, are rather aspecific since most of them are known to occur in various unrelated pathological conditions. For example, mental retardation, an important feature of FAS, can be caused by numerous other genetic and environmental conditions including chromosomal defects, inborn errors of metabolism and obstetrical and neonatal complications ( 25). To date, the group of abnormalities which constitute the FAS, taken together, has been shown epidemiologically to be characteristic only of offspring of mothers who are abusers of alcohol. There- fore, any study assessing the possible teratogenicity of alcohol by measuring the incidence of individual abnormalities rather than the occurrence of the FAS must demonstrate that other possible confounding factor(s) beside alcohol are not responsible for the observed developmental toxicity. Principle 4. The Final Manifestations of Abnormal Development are Death, Malformation, Growth Retardation, and Functional Disorder The final manifestations of abnormal development (intrauterine and neonatal death, congenital malformations, growth retardation, and functional disorders, e.g., mental 120 retardation) have all been reported to occur in offspring of animals treated with ethanol during pregnancy, as well as in progeny of heavily drinking mothers. Although any one or all could probably be induced by embryotoxic agents in sufficient dosage during critical periods of development such as organogenesis, certain manifestations are most likely to follow exposure at particular stages of pregnancy. Before differentation begins, the early embryo is refractory to most chemical insults; however, if dosage is sufficiently high, death of the conceptus may be induced before maternal toxicity intervenes. After organogenesis begins, malforma- tion of specific organs or systems is produced with relative ease, reflecting the special sensitivities and needs of rapidly differentiating and growing tissues. A generalized cells necrosis insufficient to cause embryonic death could slow the overall rate of growth proportionate to the time required to replenish the lost cells; hence some general growth retardation might also occur. Functional deficit would not ordinarily be expected following damage inflicted during organogenesis per se because histogenesis and func- tional maturation are not conspicuous aspects of development at this time. Whatever the difference and interrelations among death, malformation, growth retardation and functional deficit, they can all result in developmental aberrations. There is no logical basis for giving emphasis to one manifestation over another in evaluating adverse influences on developmental processes. Traditionally, congenital malformations have 121 been used used as the main criteria in estimating adverse developmental effects, probably because structural defects are more conspicuous; however, this does not justify ignoring the changes in mortality, growth rate, or functional capacity. Whether various ethanol-induced developmental abnormalities in humans and in laboratory animals are the result of different levels of ethanol exposure during the period of organogenesis, or are due to a toxic effect of ethanol at different stages of gestation remains to be assessed. Animal experiments may be instrumental in clarifying this problem. Principle 5. The Access of Adverse Environmental Influences to Developing Tissues Depends on the Nature of the Influence (Agent). Basically, there are two routes of access by which environmental influences may reach developing tissues in utero: 1) by directly traversing the maternal organism, or 2) or being indirectly transmitted through it. Examples of agents that pass directly through maternal tissues without modification, except dosage reduction, are ionizing radiations, certain microwaves, and ultrasound. Other physical agents such as extremes of heat and cold are not directly transmitted to the conceptus because the homeostatic processes of the maternal body are, at least initially, able to counteract the effects. Mechanical impact, short of major trauma, also is to a large extent absorbed by the maternal body and the hydrostatic cushions provided by the chorion and amnion. The indirect transmission of substances from the 122 environment (maternal organism) to the conceptus takes place via the placenta. Chemical agents or their degradation products usually reach the embryo or fetus in some fraction of their concentration in maternal blood when this is more than negligible. Whether or nor they reach effective concentrations in the con- ceptus depends upon many sets of variables. Several pharmaco- kinetic mechanisms in the maternal organism may promptly reduce the blood concentration of any given chemical. Thus, the plasma half-life of compounds is dependent upon the processes of excretion, metabolism, plasma protein binding and distribution in the extravascular compartments (126). Ethanol, as has been described in Section B-1, rapidly distri- butes into the total body water, and the blood concentration of ethanol is mainly related to its metabolism to acealdehyde by alcohol dehydrogenase rather than via renal excretion. It was once believed that the placenta served as a barrier which protected the embryo or fetus from foreign chemical. Actually, this is not the case, since virtually all unbound chemicals in maternal plasma appear to gain access to the con- ceptus across the placenta. Many molecules of small size (less than 600 molecular weight) and low ionic charge cross by simple diffusion, others by facilitated diffusion, active transport, pinosytosis,or perhaps by leakage. Lipophilic chemicals are known to cross the placenta and other membranes more readily than other compounds. Ethanol is a lipid soluble, low molecular weight compound. It freely crosses the placenta 123 and readily reaches concentrations in the fetus approximately equal to those of maternal plasma. Although most experiments have been carried out during late gestation, there is reason to believe that ethanol freely reaches the conceptus during the early stages of pregnancy as well. The total dose of a chemical reaching the conceptus is a product of the interaction of many variables, some relating to maternal functional capacity, others dependent on the nature of the chemical itself, and yet others undoubt- edly reflecting little understood characteristics of the placenta. No equation has been yet devised to express these complex interactions, in part because of all of the variables are not known, much less readily measured. However, it is well established that in the case of ethanol, the developing conceptus is exposed to approximately the same concentration of ethanol as is the mother. However, what constitutes an effective dose for the embryo is uncertain. According to Wilson (123), the embryo probably has a threshold for most chemical agents, that is, a dose below which no effects occur and above which peristent changes may be induced. As we previously discussed (Section A-3, comments) the important question pertains to how a dose of ethanol must be delivered to the embryo in order that it be toxic. Asked another way, which is more important in producing embryotoxicity -- peak concentration or duration of concentration above a given level? Translated into ethanol usage, is "binge" drinking more important in the teratogenicity of this chemical 124 than constant persistent alcohol abuse as in chronic alcoholic women? Principle 6. Manifestations of Deviant Development Increase in Degree as Dosage Increases from the No-Effect to the Totally Lethal Level. This principle concerns dosage effects generally, but is of critical importance to the current arguments about the existance of thresholds for various toxicologic effects. In the safety evaluation of new drugs and chemicals a major policy decision hinges on whether a range of dosage exists below which no adverse effect occurs. If it does, it permits compounds which may be deleterious at high dosage to be used with relative impunity at subthreshold dosage. If it does not, there is no entirely safe level, for as dosage decreases the probability of adverse effect also decreases, but theoretically does not reach zero until dosage is nil. In typical teratological studies using laboratory animals, in which intrauterine death and malformation are the criteria for adverse effects, a no-effect level has usually been found when a suitable range of dosage has been used. The same seems to hold true as regards growth retardation, al- though here the matter has been less extensively explored. The only embryotoxic manifestation about which uncertainty exists is postnatal functional abnormalities; this uncertainty arises from lack of information rather than to evidence refuting the existence of a threshold. Thus, past experience and present indications are that all manifestations of 125 abnormal development begin to be expressed only when dosage exceeds a demonstrable threshold or, conversely, that a lower range of dosage exists at which no embryotoxic effects occur. Different thresholds have been shown to exist for different types of embryotoxity, even when caused by the same agent. In fact, a low threshold for one manifestation, e.g., embryolethality during early stages, would preclude the [aspects recognition of other of embryotoxicity (e.g. teratogenesis). From the above considerations, it is clear that threshold value for ethanol toxicity during pregnancy probably exists. However, this value has not been firmly established; there- fore, it is not known whether low doses of ethanol ingested by the mother are of toxicologic importance for the develop ing conceptus. Further investigations should clarify this important problem, but the results that will be obtained are likely to be complex and difficult to interpret. In fact, it is likely that a series of threshold values, rather than only one value, will be found depending upon the animal species used, the frequency of ethanol administration, the stage of gestation of exposure, and the end points measured (i.e., con- genital malformations, embryolethality, growth retarding effects or postnatal functional abnormalities). 126 SECTION B-3: EPIDEMIOLOGICAL PRINCIPLES Ideally, epidemiological studies relating consumption of alcoholic beverages to pregnancy outcome should clarify the following: 1. The spectrum of embryotoxic effects produced by alcohol consumption. 2. Degree, frequency, duration, and time of alcohol consumption in relation to onset of pregnancy. 3. The etiologic importance of associated factors (i.e., smoking, drugs, malnutrition, etc.). 4. A measure of the risk involved. In order to evaluate the problems involved and to define the areas in which our knowledge is incomplete, a brief discussion of each of the above items is in order. 1. THE SPECTRUM OF EMBRYOTOXIC EFFECTS PRODUCED BY ALCOHOL CONSUMPTION The number of cases of FAS reported in the literature in the last few years are well over 200, and almost every affected child was born to a mother who was a medically recognized chronic alcoholic. This evidence leaves little doubt about an epidemiologic relationship between heavy and prolonged alcoholic beverage consumption by the mother and the birth of an offspring with the characteristic features of the FAS. This syndrome appears to have epidemiological specificity since as far as I know, the FAS has not been identified in the offspring of non-drinking mothers and it can readily be distinguished from other patterns of multiple malformations which share some common features, such as the syndrome of Cornelia de TOE... SA Trisomy 18, Familial Blepharophimosis, and Fetal Hydantoin Syndromes (18). Analysis of the dysmorphic features reveals that although no individual defect is pathognomonic, ocular and facial anomalies predominate (4). The presence of short palpebral fissures is a consistent and rather specific finding in the FAS (4). However, as correctly pointed out by Mulvihill et al. (18), there are three caveats in this respect: 1) palpebral fissure length is difficult to measure; 2) the normal standard was derived from an old survey of 243 white children (26,27),and 3) when standard deviations were calculated from the original data, in eight cases of FAS reviewed by these authors the palpebral fissure measurements were within two deviations of the mean, although all were below average. | Nonetheless, objective measurements are needed and criteria for measurement as well as norms and variability of palpebral fissure length should be obtained. Knowledge should be gathered as to the factors which may alter the palpebral fissure length, including gestational age, maternal race, and severe intrauterine growth retardation not associated with a chronic alcoholic mother. As for other possible embryotoxic effects of maternal alcoholism (e.g., abortions, stillbirths, prematurity), 128 the data are difficult to evaluate. Many differences exist between drinking and non-drinking mothers which have not been accounted for in a number of presently available epidemiological studies (45, 85). Furthermore, experience with other environmental causes of altered embryogenesis would lead one to anticipate variable severity of FAS in infants born to alcoholic mothers. In fact, Hanson et al. (21 ), have attempted to correlate the occurrence of children with partially expressed FAS and "moderate" amounts of alcohol consumption by the mothers. Although their paper is intriguing, the results obtained so far must be inter- preted with caution. thas, the FAS is characterized by a pattern of develop- mental abnormalities including mental retardation, micro- cephaly, intra- and extrauterine growth retardation and other associated major and minor abnormalities (4). Some of these abnormalities are aspecific in that they are known to have multiple etiologies (127). As Clarren and Smith (4) cautioned (see Section B-2), a specific listing of individual features of the FAS which are thought to be essential to the FAS diagnosis could be arbitrary and misleading. At present, an epidemiological relationship has been established only between chronic alcoholic mothers and the occurrence of offspring who possess alterations in brain function, growth and facial appearance (i.e., the full blown FAS syndrome). Until more knowledge is accumulated, the identification of cases in which the FAS is less 129 completely expressed, as was attempted by Hanson et al. with moderate doses of alcohol, can only be considered © tentative (4). At the present time, there is no clear epidemiologic evidence of a relationship between partially expressed FAS and alcoholic beverage consumption by the mother. Future research should clarify this important hypothesis. 2. AMOUNT, FREQUENCY, AND TIME OF ALCOHOL CONSUMPTION IN RELATION TO PREGNANCY Detecting and quantifying alcoholic beverage consumption in any population is not an easy task, particularly when another dimension (i.e., pregnancy) is introduced as a variable. In order to better understand the progress made and the problems still remaining, the methodology for deter- mining alcohol consumption during Pregnancy used by two ongoing American studies of alcohol and pregnancy outcome is described. The Boston University School of Medicine Program (34,38). In May, 1974, a prospective study was begun at the Boston City Hospital. After informed consent had been obtained, women registering for prenatal care were interviewed with a structured interview questionnaire. Nutritional status was evaluated on the basis of replies to the question, "What did you eat yesterday and were yesterday's meals typical?" Responses were analyzed according to the recommended dietary allowances of the National Research Council. Adequacy of nine nutrients was assessed on the 139 basis of the recommended diet for women 20 to 30 years old. Specific inquiries were made about the present and past use of alcohol, tobacco, narcotics, sedatives, ampheta- mines, hallucinogens and marihuana. No data were available on fathers, nor could blood samples on mothers be obtained for vitamin levels and other chemical determinations, Beginning in July, 1975, women were interviewed again after delivery to record changes in drinking and nutritional patterns during the pregnancy. The volume and frequency of maternal alcohol intake was determined, with separate inquiries made as to the use of wine, beer and distilled spirits. Drinking patterns were classified using the Volume-Variability Index of Cahalan, Cisin and Crossley (128). The total monthly volume of alcohol was calculated by multiplying frequency of use of each beverage by the various quantities usually consumed. Division by 30 yielded the daily volume. Variability of alcohol intake was established as either high maximum (five or more drinks on an occasion) or low maximum (never consuming five drinks). Women who drank less than once a month were classified as abstinent or rare drinkers and placed in Group 1. Heavy drinkers constituted Group 3. They consumed five or more drinks on occasion and also had a consistent daily average of more than 45 ml. of absolute alcohol. All women who drank more than once a month but did not meet the criteria for heavy drinkers were classified as moderate drinkers and placed 131 in Group 2. Pregnancy and Health Study Project (University of Seattle) (21, 42, 49, 119) - The subjects were 1,529 pregnant women participating in a larger study of ingestions during pregnancy and subsequent health and development of the off- spring. All women receiving prenatal care by the £ifth month of pregnancy at two large Seattle hospitals were eligible for the study if they lived within 20 miles of Seattle and did not anticipate moving away. Only subjects agreeing to participate in the entire longitudinal study were interviewed. An acceptance rate of 85% was obtained from those considered eligible for study. Interviews were conducted over a one-year period in 1974 and 1975. Whenever possible the interview was con- ducted in private, in the absence of children or husbands. Subjects were assured that information from the interview was for research purposes only and would not be incorporated into their medical record unless they so desired. Women were given the 35-minute interview in their own homes or at a place of their choice by one of the seven trained interviewers using standardized procedures. Both questions and answers were given verbally, with all recording done by the interviewer. The entire interview included questions on pregnancy history and personal statistics, smoking habits, beverage consumption habits during and prior to pregnancy, changes in habits during pregnancy, reasons for drinking and problems due to drinking, drug and medication ingestion during pregnancy, and diet. 132 The bewverage consumption interview was similar to that used by Jessor et al. (129) and Cahalan et al. (128). For each of the beverages (wine, beer, and liquor) women were asked how frequently they drank (with forced-choice answers ranging from "never" to "three or more times a day") and how often they drank each of the quantities "five or more," "three or more," or "one or two" glasses. Forced choice answers ranged from "never" to "nearly ever time." These quantity-frequency-variability questions for alcohol were preceded by identical questions regarding consumption of coffee and tea and followed by identical questions regarding soft-drink consumption. Each interview involved two time periods: the five months since the onset of pregnancy, referred to as "during pregnancy," and the month prior to pregnancy, referred to as "pre-pregnancy". Four alcohol scores were derived from this part of the interview; adequate test-retest reliability of these measures over a one-week interval was demonstrated (130). The four alcohol scores obtained from this part of the interview were: 1) The AA score (Jessor et al., 1968) (129), is a continuous variable representing average ounces of absolute alcohol consumed per day. It is a composite score of all types of alcohol consumed and is calculated according to . the amount of absolute alcohol consumed in each drink. 2) The VV Index (Cahalan et al., 1969) (128), is a 133 volume variability score that categorizes total alcohol consumption into 11 non-ordered categories according to subjects' average daily volume and daily variability. Thus this score gives a measure of the massing and spacing of total drinking behavior. ~~ 3) The QFV Index, quantity-frequency-variability (Cahalan et al.,1968) (128), yields five ordered categories of drinking behavior. Rather than summating across types of beverages drunk, this score reflects the quantity and variability of consumption of the most frequently consumed beverage, combined with the frequency of drinking any alcoholic beverage. 4) Beverage-specific QFV scores were also calculated, yielding a separate QFV for wine, beer and liquor. In addition to the above four scores which were all obtained from the quantity-frequency-variability part of the interview, four other alcohol scores were utilized, each deriving from a separate part of the interview: 5) The "Reasons for Drinking Soale” used in this study is dential to that utilized by Jessor et al.(129), and derives originally from Mulford and Miller (131, 132). This scale includes 25 statements presumably reflecting reasons why persons drink. Of these, 14 statements are recorded as a "personal-effect" score, presumably reflecting the degree to which a subject admits to drinking for personal reasons rather than social reasons. Previous research has shown some relationship between drinking for personal 134 reasons and alcohol problems (119). 6) The Alcohol Problems Scale, which derives from Rimmer, Pitts, Reich and Winekur (133), consists of four questions regarding serious alcohol-related problems in the areas of marriage and family, job, trouble with the law and/or hospitalization. The score reflects the number of alcohol-related problems ascribed to by a subject (range 0-4). 7) A Number of Intoxications Scale was constructed for the present study. Each subject was asked, "Have you been drunk or intoxicated during pregnancy?" Subjects responding affirmatively were asked to recall each occasion of intoxication during pregnancy and to report the amount and type of alcohol consumed, the duration of drinking on that occasion and when during pregnancy the intoxica- tion occurred. 8) An additional alcohol scale, called "Supplemental Questions," was used for subjects reporting that they drank the maximum amount (five drinks or more at a time) on ¢he guantity-frequency-variability interview. Using a procedure developed by Room (134) for assessing problem drinkers, subjects were asked how frequently they drank each of the following amounts: 5-7, 8-11 drinks, or 12 drinks or more. This set of questions differs in one important way from the quantity-frequency-variability questions. While the latter inquires about each kind of beverage separately, and the total alcohol consumption is 135 calculated later, the supplemental questions ask the subject to combine across categories of alcoholic beverages in order to determine the total number of drinks (all beverages combined) that are consumed on one occasion (119). LIMITATIONS OF THESE EPIDEMIOLOGIC METHODOLOGIES - In their elegant paper, Streissguth et al. (119) point out that B there are some limitations in estimating alcohol consumption by the various scoring systems. Thus, although at a cut-off consumption level of 1 ounce of absolute alcohol per day, the AA scale and the QFV Index each identify the same number of women as "heavy drinkers," the actual women identified by each scale are considerably different. Only a little over half of these "heavy drinkers" are simultaneously identified by both scales. The AA scale is weak in picking up binge drinkers (subjects drinking five or more drinks at a time), parti- cularly when the binging is infrequent. In addition, the AA misses 80% of the women who report being intoxicated at least once a month during pregnancy, as well as all the women who report serious alcohol-related problems. The AA scale is particularly good at picking up heavy wine drinkers, poorer at picking up heavy liquor drinkers and misses over half of the heavy beer drinkers. The main advantage of the AA score is its continuous nature, thus discriminating among subjects with very heavy alcohol ingestion. 156 The relationship of VV to QFV is also much poorer for pregnant women compared to data presented by Cahalan et al. (128) for men and women drinkers combined. The VV Index picks up a higher proportion of the heavy wine drinkers, but misses one-fourth of the heavy liquor and heavy beer drinkers. Compared to the AA, the VV picks up more of the women reporting at least one intoxication during pregnancy and more of the women reporting serious alcohol-related problems. However, half of these are still missed by the VV. The number of intoxications does show some relationship to heavy drinking in that most subjects reporting higher numbers of intoxications are also heavy drinkers by both AA and VV. The predictive value of the intoxication scale used alone would have few false positives but a very large number of false negatives. Some very heavy drinkers report no intoxications and would thus be missed. As an example reported by Streissguth et al.(119), one woman with an AA score of 25.76 ounces/day, reported no intoxications. The number of alcohol-related problems is also a poor predictor of current drinking practices. Some pregnant women reporting problems had stopped drinking and were abstainers; some were apparently only infrequent but severe binge drinkers. The vast majority of the heavily drinking women in the Seattle study reported no alcohol- related problems. As pointed out by Streissguth et al.(119), this may be due to the fact that the items in the scale they used included only four of the most severe alcohol- 137 related problems. While the above discrepancies have involved primarily scales deriving from the same interview, two basic problems with the QFV interview are the low ceiling for maximum number of drinks per occasion and the inability of the interview to accomodate a summation of drinks per occa- sion across different types of alcohol. The addition of the "Supplemental Questions" to the usual QFV interview is recommended for identifying the potentially high risk drinkers. In addition, the "number of intoxications" scale has been recommended to help identify the timing of periods of high alcohol intake. Further questions regarding specific incidents of intoxica- tion can also be asked to clarify the duration of intoxica- tion, and both the types and amounts of beverages consumed. The elusiveness of the high risk drinker is best described by one woman who scarcely drank at all during pregnancy (119). She was categorized an "Infrequent" drinker on the VV Index during pregnancy, a "Moderate" drinker on the QFV, and her AA score was so low (0.06 ounces of absolute alcohol per day) as to be inconsequential. If only the AA and/or VV scales had been used she could easily have been placed in the control group as an infrequent drinker. However, the question on intoxications during pregnancy picked up two binges, one at six weeks gestation in which she drank five bottles of tequila and three beers in a four-hour period and another at 12 weeks in which she drank a six-pack of beer and two bottles of table wine in 138 a four-hour period. The QFV and VV indexes had no mechanism for dealing with the combined effect of the two types of liquor while the AA, which does summate across beverages, averages the quantity out over so many days that the daily rate becomes insignificant. In addition, neither scale could accurately reflect the amount in excess of five drinks that she consumed on these occasions, although this information was picked up by "Supplemental Questions." Outcome studies in which such drinking patterns are of interest will miss subjects like this if only the QFV inter- view is used. The poor inter-relationship of the numerous alcohol scores suggests that the validity of this type of alcohol consumption data may not be high. Part of this problem may be due to the attempt to oversimplify relatively complex drinking activity by reducing it to a single scale or a small number of categories. It is therefore probably advisable to use multiple alcohol assessments in attempting to correlate alcohol consumption by women and pregnancy outcome. In addition to uncertainties about the amount and frequency of alcohol consumption, the methodology of these studies does not give us an accurate measure of alcohol consumption during early embryogenesis when most drugs are known to exert teratogenic effects (123). In the Pregnancy and Health Study Project in Seattle, 139 an attempt has been made in this regard, but their methodology is not above criticism, As reported above, they estimated by interview at 5 months the alcohol consumption by the mother during two time periods: the five months since the onset of pregnancy, referred to as "during pregnancy,' and the month prior to pregnancy, referred to as ''pre-pregnancy.'" However, in a later publication from the same Project, the terminology of the latter group was changed to "alcohol intake prior to recognition of pregnancy" (21). Thus, it is not clear whether the alcohol consumption was before pregnancy or during early pregnancy. This issue may be of critical importance for understanding the etiology and pathogenesis of FAS . Thus, at least in this country women generally recognize that they are pregnant at about 6 weeks from their last normal period. At this time, the human conceptus is four weeks old and undergoing active organogenesis. Therefore, alcoholic consumption when pregnant women are "waiting for their period" and do not know they are preg- nant may be one of the most critical factors in establishing alcohol teratogenicity. In summary, present available epidemiologic data on embryotoxicity of ethanol during pregnancy have been obtained using techniques that do not measure alcoholic consumption precisely. Subjects identified as "heavy drinkers" by the AAA, and QFV, and VV scores are frequently not the same subjects (119). In addition, information obtained from retrospective interviews about drinking habits in early pregnancy tend to be inaccurate since the possibility of 140 NG memory bias exists (135), particularly in the occasional binge drinker. 3. THE ETIOLOGIC IMPORTANCE OF ASSOCIATED FACTORS The effect of maternal alcoholism, particularly in relation to incidence of abortions, stillbirth and neonatal mortality, prematurity, weight-height deficiency, and mental retardation with peculiar facial features, is a complex problem. The similarity in the overall pattern of anomalies in the children of chronic alcoholic mothers suggests a single etiology, most likely environmentally determined by some as yet unknown alteration in maternal metabolism. Regarding direct toxicity to the developing fetus, the most obvious possibility is ethanol itself. Other possibilities involving direct toxicity include a metabolite of ethanol, such as acetaldehyde, or an unknown toxic agent in the alcoholic beverages consumed. The adverse effect on growth and embryogenesis of maternal chronic alcoholism could also be the indirect consequence of generalized maternal undernutrition or deficiency of a specific nutrient or vitamin (2). Thus, exposure of the developing conceptus to alcohol may not be the only toxic factor in determining the poor pregnancy outcome in alcoholic women, but other contributing factors may also be at work. It has been epidemiologically determined that the heavily drinking mothers are older and of high parity as compared to non-drinking or light-drinking mothers (40). They have a poor weight gain during pregnancy (2), are 141 more likely to experience bleeding during early pregnancy, and they have a history of more children with low birth weight (33). Furthermore, heavy alcohol consumption is highly correlated with cigarette smoking (21, 33, 40, 42) and to a lesser extent with caffeine use (35) and chronic alcoholic mothers frequently present clinical signs of malnutrition (136). All these factors with alcohol intake are known per se to represent an increased risk of poor pregnancy outcome (11z, 137). According to Mau and Netter (35), isolated alcohol consumption per se is not the only reason for the unfavorable outcome of pregnancy in chronic alcoholic mothers, but is / part of a multiplicity of unfavorable factors associated with this type of patient including possibly her constitu- tional and character type. Whether this hypothesis has any validity remains to be assessed, but it points out our present lack of knowledge about the importance of associated factors which may contribute to the occurrence of abnormal offspring in chronic alcoholic mothers, Therefore, any study design investigating the relation- ship hetween alcohol intake by the mother and pregnancy outcome should consider the possible confounding factors. Unfortunately, the confounding factors are numerous and not always well identified. Factors such as maternal age and socioeconomic status are clearly related to pregnancy outcome, thus making the selection of the population to be observed critical, But these are not the only factors. For example, prematurity and low birth weight are related (or possibly related) to innumerable high-risk factors 142 including socio-economic status, inadequate prenatal nutri- tion, maternal childhood factors, inadequate prenatal care, pregnancy spacing, maternal age, birth order, marital status, toxic substances ingested including legal and illegal drugs, maternal infections, toxemia of pregnancy, ethnic group, and genetic make-up (138). Although some studies have attempted to control for a number of the known confounding variables, our knowledge is at present too limited to evaluate the relative importance of each of these factors in determining the poor pregnancy outcome of chronic alcoholic women. Furthermore, if prospective epidemiologic attempts are made to evaluate pregnancy outcome in drinking, but not medically recognizable alcoholic mothers, the design complexity of such a study is greatly increased. Sample size and type of the population to be considered, accurate and objective measurement of alcohol intake in relation to stages of pregnancy, controls for the numerous medical and obstetrical confounding factors, and establish- ment of the validity of the end points measured, are problems that should be carefully considered. These pro- blems require resolution before initiation of any epidemiologic survey. We urgently need epidemiologic studies on alcohol toxicity during pregnancy which utilize approaches to the problem in its total complexity. It is only in this way that the results obtained will be relevant to entire 143 populations. 4. A MEASURE OF THE RISK INVOLVED The present available data are not sufficient to accurately, establish the incidence of FAS in the general population and the relative risk for a drinking mother of producing an affected child. However, some attempts have made in this respect. Jones and Smith (139) reported that 43% of the pregnancies of 23 chronic alcoholic mothers had been an adverse outcome. Four of the offspring died in the perinatal period and 6 FAS cases were diagnosed in the survivors. The same group of workers reported in another publication (140) that the risk of poor fetal outcome for chronic alcoholic mothers is in the range of 30-50%. Hanson et al. (21) speculate that the risk of having an affected newborn increases proportionately with the average daily alcohol intake of the mother during the month prior to recognition of her pregnancy. According to them, "if the maternal ingestion is less than, 6 one ounce of absolute alcohol per day, the apparent risk for abnormalities appears to be low. In the range of 1 to 2 ounces of absolute alcohol per day, the risk may approach 10%. Among women who drink an average of 2 or more ounces of alcohol daily, 197% had infants who were considered abnormal." Unfortunately, this attempt to establish a dose-effect relationship is as provocative as it is premature. ; Except for two infants born to mothers who were chronic alcoholics, they did not find any additional cases of FAS 144 using the criteria described by Clarren and Smith (4). They did, however, attempt to identify features compatible with fetal alcohol syndrome (i.e., partial expression of FAS) by examining 163 infants preselected by a research assistant in pairs: one new born from a drinking mother (AA>1.0/day) and the other from an abstainer or light drinking mother (AA