In 1973, Jones and Smith (1) coined the term "fetal alcohol syndrome" (FAS) to describe a pattern of abnormalities observed in children born to alcoholic mothers. It was originally postulated that malnutrition might be responsible for these defects.

However, the pattern of malformation associated with FAS is not seen in children born to malnourished women, and alcohol has been found to be acutely toxic to the fetus independently of the effects of malnutrition (2,3).

Criteria for defining FAS were standardized by the Fetal Alcohol Study Group of the Research Society on Alcoholism in 1980 (4), and modifications were proposed in 1989 by Sokol and Clarren (5). The proposed criteria are 1) prenatal and/or postnatal growth retardation (weight and/or length below the 10th percentile); 2) central nervous system involvement, including neurological abnormalities, developmental delays, behavioral dysfunction, intellectual impairment, and skull or brain malformations; and 3) a characteristic face with short palpebral fissures (eye openings), a thin upper lip, and an elongated, flattened midface and philtrum (the groove in the middle of the upper lip).

Sokol and Clarren (5) suggested the term "alcohol-related birth defects" (ARBD) to describe anatomic or functional abnormalities attributed to prenatal alcohol exposure. The term "possible fetal alcohol effect(s)" (FAE) indicates that alcohol is being considered as one of the possible causes of a patient's birth defects. In the view of Sokol and Clarren, the frequent use of this term to indicate a birth defect judged milder than FAS is incorrect, although others continue to use it that way (5).

Mental Handicaps

Mental handicaps and hyperactivity are probably the most debilitating aspects of FAS (6), and prenatal alcohol exposure is one of the leading known causes of mental retardation in the Western World (7). Problems with learning, attention, memory, and problem solving are common, along with incoordination, impulsiveness, and speech and hearing impairment (8,6). Deficits in learning skills persist even into adolescence and adulthood (6,9).

It is generally accepted that the adverse effects of prenatal alcohol exposure exist along a continuum, with the complete FAS syndrome at one end of the spectrum and incomplete features of FAS, including more subtle cognitive-behavioral deficits, on the other. Thus, infants with suboptimal neurobehavioral responses may later exhibit subtle deficits in such aspects of daily life as judgment, problem solving, and memory (6).

Studies of the incidence of FAS are complicated by methodological problems. Data have been collected in various ways: 1) in the catchment approach, birth defects are monitored at the time of birth only; 2) in retrospective studies, children are identified as having FAS at some time after birth; and 3) in prospective studies, children are followed over time and assessed at various intervals from birth onward. Catchment data tend to underestimate FAS incidence because the neonatal period is a difficult time to detect FAS.

Not only are facial features associated with the syndrome difficult to recognize, but the central nervous system dysfunction, including mental retardation, may not be identified until several years after birth (5,10,11). On the other hand, retrospective and prospective studies may overestimate FAS incidence by oversampling populations where FAS incidence is unusually high (10). Analyses are further complicated by the unreliability of self-reports of maternal drinking (12).

Reported Cases

Catchment data on the incidence of FAS are derived from the Birth Defects Monitoring Program of the Centers for Disease Control (CDC) (13). Based on data from 1,500 hospitals, CDC reported the nationwide incidence of FAS to be 0.3-0.9 per 10,000 births (excluding Native Americans). In contrast, Abel and Sokol (10) surveyed 19 published epidemiologic studies worldwide. The overall rate from all studies was 1.9 cases per 1,000 live births. The average for retrospective studies surveyed by Abel and Sokol was 2.9 per 1,000, compared with 1.1 per 1,000 for prospective studies.

Most reported cases in the United States came from study sites where the mothers were black or Native American and of low socioeconomic status. The estimated rate at these sites was 2.6 per 1,000 compared with 0.6 per 1,000 from other study sites, where the mothers were predominantly white and of middle socioeconomic status (10).

According to the CDC catchment study, incidences of FAS per 10,000 total births for different ethnic groups were as follows: Asians 0.3, Hispanics 0.8, whites 0.9, blacks 6.0, and Native Americans 29.9 (13). Because of differences in study design, the ratios among the various ethnic groups derived from the CDC catchment data cannot be used to estimate FAS incidence for different ethnic groups as obtained from prospective and retrospective studies.

Among Native Americans, the incidence of FAS varies among different cultures. Health units serving principally Navajo and Pueblo tribes report an FAS prevalence similar to that for the overall U.S. population, while for Southwest Plains Indians, a much higher prevalence was reported (1 case per 102 live births) (14). Several factors, such as cultural influences, patterns of alcohol consumption, nutrition, and metabolic differences have been suggested to play a role in this difference (15).

Genetic Susceptibility?

In the case of blacks, the risk of FAS remains about sevenfold higher than for whites, even after adjustment for the frequency of maternal alcohol intake, occurrence of chronic alcohol problems, and parity (number of children borne) (16). This raises the question of some kind of genetic susceptibility, the nature of which is unknown.

Apart from epidemiology, the key questions in FAS research include, How much alcohol is too much? and, When is the fetus at greatest risk? The major problem in addressing these questions is the lack of a specific physiological measure that accurately reflects alcohol consumption. There is no biological marker currently available to measure alcohol intake, and self-reports of alcohol consumption may be unreliable, perhaps especially so during pregnancy (17).

Morrow-Tlucak and colleagues (18) found that women with more-serious alcohol-related problems are those more likely to underreport their alcohol consumption when interviewed during pregnancy.

While it is apparent that children who meet the criteria for FAS are born only to those mothers who consume large amounts of alcohol during pregnancy, studies have reported neurobehavioral deficits and intrauterine growth retardation in infants born to mothers who reported themselves to be moderate alcohol consumers during pregnancy (19,20,21).

 

Minimum Threshold Not Defined

In a prospective study of 359 newborns, Ernhart and colleagues (22) found a trend toward increasing head and facial abnormalities with increasing embryonic alcohol exposure. An effect occurred at even the lowest reported levels of alcohol intake, so that a clear threshold (minimum amount of alcohol to produce an effect) could not be defined (22).

Given the range of defects that result from prenatal alcohol exposure, the search for an overall threshold for fetal risk may be unreasonable. Instead, each abnormal outcome in brain structure and function and growth might have its own dose-response relationship (23). Animal research has shown that different profiles of alcohol-related birth defects are related to critical periods for specific aspects of fetal development (3).

Thus, heavy alcohol consumption throughout pregnancy results in a wide variety of effects characteristic of FAS, while episodic binge drinking at high le vels results in partial expression of the syndrome, with the abnormalities being unique to the period of exposure (24). Vulnerability of individual organ systems may be greatest at the time of their most rapid cell division (25).

An important strategy for preventing alcohol-related birth defects is the development of better screening techniques to identify women at high risk for heavy alcohol consumption throughout their pregnancy. Currently available laboratory tests for detecting biochemical markers of heavy drinking are not as sensitive as self-report screening instruments, whereas the latter are complicated by denial (12).

Abstaining is the Key

A possible way to overcome denial might be to inquire about past, rather than present, drinking. This is suggested by the results of a study showing that self-reports of first trimester drinking made at the seventh month of pregnancy are often higher than those made at the fourth month (26). The researchers suggested that women may feel safer reporting higher levels of drinking farther away from the event.

Although this strategy may not reveal a drinking problem until relatively late in pregnancy, intervention at this time is still useful. While abstaining during the second trimester does not eliminate the risk of fetal abnormalities, it does seem to mitigate some of the behavioral effects that may occur shortly after birth (27,20).

Sokol and colleagues (12) developed a simple and brief questionnaire to help circumvent denial and underreporting of heavy drinking by pregnant women. The test instrument, referred to as T-ACE, correctly identified 69 percent of the "risk drinkers" (defined as those consuming 1 ounce of absolute alcohol per day, equivalent to two standard drinks per day) out of a cohort of 971 pregnant women.

T-ACE was found to be superior to other standard instruments used for detecting alcohol abuse, such as MAST and CAGE. The test is brief, and may be administered easily in prenatal clinics and obstetricians' offices. Its key feature is a tolerance ("T") question, "How many drinks does it take to make you feel 'high?' " (Tolerance is acquired by drinking.) Clinical experience suggests that questions about tolerance are less apt to be perceived by lay persons as an indication of drinking, and are therefore less likely to trigger denial (12). A more reliable indicator of heavy drinking awaits the development of objective biochemical markers.

Fetal Alcohol Syndrome--A Commentary by
NIAAA Director Enoch Gordis, M.D.

From a scientific perspective, the link between moderate drinking and alcohol-related birth defects has not been clearly established. Whether there is a threshold below which alcohol can be consumed without harming the fetus is not known: self-reported data showing a relationship between moderate use and alcohol-related birth defects may often underestimate the true level of drinking. Researchers are working on developing an objective marker for alcohol consumption that will help clarify these questions and assist clinicians in identifying alcohol-abusing patients as a part of routine prenatal care, using, for example, blood samples typically drawn during an initial examination.

Clinicians, however, must offer advice to their patients based upon the best available scientific evidence. Although some clinicians believe that recommending total abstention for pregnant women may subject them to unwarranted guilt about drinking small amounts of alcohol, most accept the need for clinical caution. Because we do not know at what point alcohol damage begins, it is prudent to recommend, as I do, that pregnant women abstain from alcohol use pending confirmation of alcohol's role vis-à-vis fetal development.

There is good news in recent evidence that the number of women who consume alcohol during pregnancy is declining. However, it also appears that the rates of alcohol consumption among high-risk populations (pregnant smokers, unmarried women, women under the age of 25, and women with the least amount of education) remain virtually unchanged (28). This points to a need to develop better targeted prevention and education efforts to reach high-risk populations and to identify women at high risk through primary health care and other systems traditionally used by high-risk individuals before and during pregnancy.