Effects of Alcohol Abuse on Reproductive Function in Women


 Alcohol abuse and alcoholism are associated with disorders of reproductive function in both men and women. Amenorrhea, anovulation, and luteal phase dysfunction may occur in alcohol-dependent women and alcohol abusers. Yet there has been relatively little research on the consequences of alcohol abuse for female reproductive function. Recent clinical and survey studies of alcohol effects on pituitary gonadotropins and gonadal steroid hormones in women are reviewed. Experimental studies of the acute and chronic effects of alcohol on the hypothalamic-pituitary-gonadal axis in normal women and in animal models are also described. Recent studies of the acute effects of alcohol on opioid antagonist and synthetic LHRH-stimulated pituitary gonadotropins are summarized. Possible mechanisms underlying alcohol-induced disruptions of menstrual cycle regularity are discussed.

The adverse effects of alcohol on reproductive function in men are well documented. Impotence and diminished sexual interest are common clinical complaints among alcohol-dependent men. Testicular atrophy, low testosterone levels, and gynecomastia are often associated with chronic alcohol abuse (). There is now considerable evidence that alcohol inhibits testosterone biosynthesis by direct toxic effects on the testes. The extent to which hypothalamic and pituitary factors contribute to alcohol-induced male reproductive system pathology remains controversial, but there is little question that alcohol is a gonadal toxin in the male.’

In view of the considerable evidence for severe alcohol-related reproductive system dysfunctions in men, it is surprising that there has been so little attention to alcohol’s effects in women. Recent reviews have consistently pointed out that almost nothing is known about the acute and chronic effects of alcohol on the hypothalamic-pituitary-gonadal axis in women or in animal models. Historically, objective information about alcohol problems in women has been rather sparse, and most behavioral and biological studies of alcoholism have been conducted in men. An emerging recognition that alcohol’s effects on female reproductive function is an important area of inquiry appears to reflect several interacting factors. A subtle but important change in attitudes toward women’s health issues is illustrated by the 1985 Public Health Service Report on Women’s Health Issues, which focuses on the special health problems of women throughout the life-cycle. There has also been an increased awareness that a substantial number of women are afflicted with alcoholism. A survey of major metropolitan areas sponsored by the National Institute on Mental Health reported that alcoholism and alcohol abuse were the fourth most frequent of all psychiatric disorders among young women aged 18-24. Moreover, a recent survey of 10,000 first admissions for alcoholism treatment to a proprietary hospital system indicated that 24% were women. Finally, accumulating evidence that alcohol and other drugs adversely affect fetal growth and development argues for the importance of examining alcohol’s effects on those reproductive hormones essential for healthy prenatal development.

This chapter will discuss clinical data concerning alcohol effects on reproductive function in women of childbearing age. Recent experimental studies of alcohol effects on pituitary and gonadal hormones in women and selected studies in animal models will also be described. The effects of alcohol on reproductive hormones in postmenopausal women have been reviewed by Gavaler and will not be repeated here. Interpretation of data derived from clinical, survey, and experimental studies is qualified by the limitations inherent in each approach. Yet, despite these limitations, several lines of evidence converge to indicate that alcohol abuse and alcohol dependence are associated with a wide range of reproductive system dysfunctions in women and in animal models. Moderate and heavy social drinkers also may have similar disorders of menstrual cycle function and fertility. The mechanisms underlying alcohol’s toxic effects on female reproductive function remain to be determined.

Clinical Studies of Reproductive System Dysfunctions in Alcoholic Women

Clinical Studies of Alcohol Effects on Female Reproductive Function in Social Drinkers

Alcohol-related luteal phase dysfunction and anovulation have also been observed in healthy, well-nourished women who were given an opportunity to self-administer alcohol for 21 days on a clinical research ward. Twenty-six social drinkers were admitted to a clinical research ward for 35 days. A 7-day alcohol-free baseline was followed by 21 days of alcohol availability and a postalcohol period of 7 days. Women could earn alcohol (beer, wine, or distilled spirits) or 50c for 30 min of performance on a second-order fixed-ratio-300, fixed-interval-1-sec schedule of reinforcement. Points earned for alcohol and for money were not interchangeable.

Five women were classified as heavy drinkers who consumed an average of 7.8 drinks per day. Twelve women were classified as moderate drinkers who consumed an average of 3.85 drinks per day, and nine women were classified as occasional drinkers who consumed an average of 1.22 drinks per day. The moderate and heavy drinkers achieved peak blood alcohol levels that averaged between 109 and 199 mg/dl. Peak blood levels of the occasional drinkers averaged between 87 and 148 mg/dl.

Forty percent of all women who were moderate (2.5-4.0 drinks per day) or heavy (over four drinks per day) alcohol users had significant abnormalities of the menstrual cycle, including anovulation, luteal phase deficiency, delayed ovulation, and recurrent hyperprolactinemia. Women who consumed less than 2.5 drinks per day during the inpatient research ward study did not show significant derangements in menstrual cycle function. Therefore, living on a research ward for 35 days did not cause significant deviations in menstrual cycle duration or alter normal hormonal profiles during the follicular, periovulatory, or luteal phases of the menstrual cycle. Since these women were otherwise healthy and well nourished, we concluded that alcohol and not extraneous factors accounted for the menstrual cycle derangements observed.

Survey Reports of Alcohol Effects on Reproductive Function

Episodic heavy drinking is also associated with reports of menstrual cycle disorders. In 1984, the first survey data on drinking and female reproductive dysfunction from a representative national sample were reported. A stratified household sample of 917 women showed a strong association between alcohol consumption and a variety of menstrual disorders, including dysmenorrhea, heavy menstrual flow, and premenstrual discomfort. The incidence of these disorders increased with usual drinking levels. Women who consumed six or more drinks per day at least five times a week had elevated rates of gynecological surgery (other than hysterectomy) and obstetrical disorders. These are the first data to provide evidence that drinking and reproductive dysfunction are related in the general nonclinical population.

Survey data also indicate an association between alcohol use and spontaneous abortion. A prospective study of 32,019 women at their first antenatal visit to a Kaiser Hospital Clinic was conducted between 1974 and 1977. There were 1503 spontaneous abortions, 714 in the first trimester (5-14 weeks) and 789 in the second trimester (15-27 weeks). Thus, the overall rate of abortion was 14.4%, and 2.6% of these occurred during the second trimester. An alcohol-use questionnaire revealed that 44.6% drank less than one drink daily, 2.4% had one to two drinks daily, 0.4% had 3.5 drinks daily, and 0.1% had more than six drinks daily. Women who had one or more drinks each day had more spontaneous abortions primarily during the second trimester than women who were abstinent or only occasional drinkers. The second-trimester spontaneous-abortion risk was 1.03 for occasional drinkers, 1.98 for regular drinkers who consumed one to two drinks per day and 3.53 for women who had more than three drinks per day.

There was no significant effect of alcohol on first-trimester miscarriages. The relative risk of first-trimester miscarriages was 1.12 in occasional drinkers, 1.15 in regular drinkers (one or two drinks per day), and 1.16 in women who had three drinks or more daily. Statistical analyses showed that the effects of alcohol were independent of the effects of cigarette smoking, and moreover, the effects of drinking were greater than the effects of smoking. Similar conclusions concerning an association between spontaneous abortion and moderate drinking have been reached by other investigators. A comparison of drinking frequency in 616 women who had spontaneous abortions and 632 women who had normal deliveries resulted in an estimate that 25% of women who drank twice a week were likely to abort and about 14% of women who drank less frequently were likely to abort. One ounce of absolute alcohol was estimated to be the minimum harmful alcohol dose.

Studies of Chronic Alcohol Effects in Animal Models

The development of valid animal models of alcoholism is essential for analysis of the derangements of reproductive function observed in alcoholic women. It has been difficult to determine the relative contribution of alcohol-associated liver disease, malnutrition, and the direct toxic effects of alcohol in alcoholic women. Alcohol-induced disruptions of reproductive function in an animal model can be attributed to alcohol without the confounding influence of liver disease, pancreatitis, malnutrition, or severe weight loss, any of which could disrupt menstrual cycle regularity. Moreover, animal models permit a systematic evaluation of the effects of alcohol dose and the duration of alcohol exposure on pituitary and gonadal hormones under controlled conditions.

The female rhesus monkey offers the particular advantage that its reproductive physiology is very similar to that of human females and the neuroendocrine regulation of primate reproductive function has been studied extensively (). Rodents have an estrus cycle rather than a menstrual cycle, and there appear to be major differences in hypothalamic control of the rodent estrus cycle and the primate menstrual cycle. Consequently, some aspects of the neural control of reproductive function in rodents are not completely generalizable to humans. However, rodents provide a useful model that has been widely used for the analysis of alcohol’s effects on pituitary and gonadal hormones.

Both pharmacological (forced-administration) models and behavioral (self-administration) models of alcoholism are available. In pharmacological models, the dose of alcohol administered and the duration of exposure are determined by the investigator. In behavioral models, the animal works at a behavioral task to obtain alcohol or freely ingests alcohol solutions, usually in the context of working at a behavioral task for food. Alcohol has been administered orally in a liquid diet or by restricting fluid availability to alcohol, nasogastrically by intubation, intragastrically by infusion, intravenously by infusion, and via inhalation. Examination of the advantages and limitations of each type of model is beyond the scope of this chapter. The relative value of each model is dependent on the types of questions that are subsequently asked. However, the development of animal models of alcohol addiction in several species is one of the most significant advances in research on alcoholism in recent years ().

Alcohol Effects on Reproductive Function in Primate Models

Amenorrhea, atrophy of the uterus, and decreased ovarian mass were observed in female macaque monkeys trained to self-administer alcohol using operant behavioral procedures. After at least 10 months of normal ovulatory menstrual cycles, monkeys were trained to work for food (1 g banana pellets) and then implanted with an intravenous catheter using aseptic procedures. An average of 64 responses was required for each food pellet or i.v. alcohol injection (0.12 g/kg per injection) under a second-order schedule of reinforcement [FR 4 (VR 16:S)]. Food and alcohol each were available during four 1-hr sessions each day, and the nutritionally fortified banana pellet diet was supplemented with fresh fruit, vegetables, biscuits, and multiple vitamins each day. The onset and duration of menstrual bleeding were determined with daily vaginal swabs. Venous blood samples were collected periodically for radioimmunoassay of neuroendocrine hormones.

Three monkeys began to self-administer relatively high doses of alcohol immediately and averaged between 2.3 and 3.2 g/kg per day during the first month of alcohol availability. Self-administration of high doses of alcohol (2.9-4.4 g/kg per day) for 3-6.5 months was accompanied by amenorrhea that persisted for 84 to over 200 days. Amenorrheic monkeys developed post-session blood alcohol levels ranging from 266 to 438 mg/dl, comparable to levels observed in alcoholic men during intoxication. After the 11-hr interval between the midnight and noon alcohol sessions, each monkey showed signs of physical dependence evidenced by gross tremor of the extremities and nystagmus. Monkeys that self-administered relatively low doses of alcohol (1.3 and 1.6 g/kg per day) for 119 and 173 days, respectively, continued to have stable ovulatory menstrual cycles.

Figure “Alcohol and food self-administration by three alcohol-dependent amenor-rheic monkeys. The average daily dose of alcohol self-administered over 10-day periods in an operant paradigm is shown by the bars and the average number of 1-g banana pellets self-administered during the same period is shown by the circles. Each data point is the mean (± standard error) of 10 values, except for the final data point for monkey 10-80, which died of an alcohol overdose (the average food and alcohol intake during the final 3 days is shown). Each monkey was first given access to alcohol (0.12 g/kg per injection) during menstruation or during the late luteal phase of the control menstrual cycle and did not menstruate subsequently” shows patterns of alcohol and food self-administration over successive 10-day intervals by the amenorrheic monkeys for 84, 93, and 160 consecutive days. Despite daily intoxication, two monkeys showed no change in operant food self-administration from baseline during the first 50-60 days of alcohol exposure, and one monkey worked for significantly more food after 70 days of alcohol self-administration than during the prealcohol baseline. No monkey had evidence of liver disease, and laboratory tests of lipid and carbohydrate metabolism, electrolyte homeostasis, and hematological function were normal. These data indicate that alcohol, rather than malnutrition or liver disease, accounted for the observed disruption of menstrual cycle regularity. It is also unlikely that seasonal factors contributed to these findings since amenorrhea was observed between the fall and spring, the usual breeding period in the Northern Hemisphere.

LH levels were significantly lower than baseline LH levels measured at menstruation in each monkey that developed amenorrhea. However, in monkeys that self-administered low doses of alcohol, LH levels did not differ significantly from control levels. Necropsy revealed pathological changes in the reproductive systems of two monkeys, including marked atrophy of the uterus with a paucity of glands and dense endometrial stroma. Decreased ovarian mass with an absence of corpora lutea suggests that ovulation did not occur. The pathological picture was consistent with that seen in early menopause. These data indicate that chronic alcohol intoxication produces similar disruptions of reproductive function in alcoholic women and macaque monkeys. The validity of this model is especially compelling since each monkey controlled her alcohol dose and self-administered alcohol to the point of daily intoxication and physical dependence. Our ongoing studies with the primate alcohol self-administration model have shown that chronic alcohol administration also suppresses ovulation and results in luteal phase dysfunction in otherwise healthy animals. Studies to determine the critical alcohol dose necessary to disrupt menstrual regularity and the relative vulnerability of the follicular and luteal phases to alcohol toxicity are in progress in our laboratory.

Alcohol Effects on Reproductive Function in Rodent Models

Chronic alcohol exposure also disrupts estrus cycle regularity and results in longer estrus cycles in adult and immature rodents. Five weeks’ exposure to a liquid diet containing 5% alcohol w/v significantly decreased the frequency of estrus and proestrus phases and also increased the duration of the diestrus and metaestrus phases in comparison to pair-fed or chow-fed controls. LH levels were lower and prolactin levels were higher in the alcohol-fed rats than in either control group. The alcohol-exposed rats consumed between 10.5 and 11.2 g/kg per day. Alcohol-exposed animals had variable cycles with diestrus phases ranging from 4 to over 30 days, whereas the duration of the estrus cycle in control animals was 4-5 days. On average, prolactin levels were 3-4 times higher in alcohol-exposed rats than in either control group.

Rats chronically exposed to alcohol also failed to show a significant increase in LH after acute administration of estradiol benzoate (50 (xg), whereas a significant LH increase occurred at 31 and 55 hr after estradiol administration in control animals. However, 5 weeks of chronic alcohol exposure (11 ± 2 g/kg per day) did not prevent a significant increase in prolactin following estradiol benzoate administration. These data were concordant with previous observations of increased prolactin levels during chronic alcohol exposure, and an alcohol-related hyperprolactinemia was suggested as one possible factor disrupting estrus cycles in rats.

Twelve days’ exposure to ethanol (4 or 8 g/kg) via intragastric intubation reduced the frequency of proestrus and estrus stages and increased diestrus in 60-100% of adult female rats studied. Single samples collected for LH analysis on days 9-12 of alcohol intubation revealed no alcohol-related suppression during proestrus and no differences from control animals. All animals returned to normal estrus cycles within 20 days. Similarly, after 16 weeks of exposure to a 5% ethanol liquid diet, rats with irregular estrus cycles were mated and subsequently delivered viable offspring. These data attest to the resiliency of the reproductive system.

Chronic alcohol exposure has been shown to produce ovarian failure in both weanling and adult female rats. Reduced ovarian weight reflected an absence of corpus lutea and corpora hemorrhagica, indicating a failure to ovulate. Reduced tissue mass of the uterus and fallopian tubes was consistent with estrogen deficiency. After 6 weeks of exposure to a 5% liquid diet, estradiol levels were significantly lower in alcohol-fed weanlings than in controls. LH levels were increased in both alcohol-fed weanlings and isocaloric controls relative to ad libitumied controls. Despite histological evidence of ovarian failure, LH and FSH levels in alcohol-exposed animals were significantly lower than in oophorectomized ad libitumied controls. These findings in weanling rats were confirmed in adult (60-day-old) female rats. Ovarian pathology following 4 months of exposure to a liquid diet containing 5% alcohol revealed an absence of corpora lutea and corpora hemorrhagica and only a few immature follicles. Again, the histological appearance of the uterus and of the fallopian tubes of alcohol-fed rats was consistent with estrogen deprivation.

These alterations of ovarian structure appeared to be due to alcohol toxicity and not to caloric deprivation or to liver disease. However, unlike the weanling rats, estradiol and LH were not significantly different from control levels, whereas FSH levels were significantly increased in adult females. The differential effect of chronic alcohol exposure on LH and FSH was interpreted to suggest that the absence of mature follicles in the ovary may have been associated with lower levels of the ovarian peptide inhibin, which down-regulates FSH.

Studies of Acute Alcohol Effects in Women and Animal Models

Directions for Future Research

It is apparent from the foregoing that much remains to be learned about alcohol’s effects on female reproductive function. Although disorders of menstrual cycle regularity observed in alcoholic women have also been demonstrated in otherwise healthy animal models, the clinical database is relatively small. It would be useful to have gynecological history data and complete endocrine profiles on a representative sample of alcoholic women to determine the prevalence and hormonal concomitants of amenorrhea, anovulation, and luteal phase dysfunction. Very little information is currently available about the time course and prospects for recovery of normal menses in abstinent former alcoholic women. The extent to which the reproductive system develops tolerance to the disruptive effects of chronic alcohol intoxication is also unknown. Yet alcohol-dependent women do conceive, and pregnancy is not always terminated by spontaneous abortion. Recent studies showing that alcohol intoxication may increase rather than decrease LH and estradiol are intriguing, and the implications for fertility and normal pregnancy remain to be determined.

Interpretation of data from clinical studies of alcohol-dependent women has always been limited by the fact that most have concurrent medical disorders (liver disease, pancreatitis, malnutrition) which could account for the derangements of the reproductive system. However, recent studies showing alcohol-related menstrual cycle disorders in normal social drinkers and in experimental animals argue strongly that alcohol is a reproductive system toxin and its deleterious effects may be augmented by concurrent alcohol-related medical disorders. Animal models of alcoholism are essential for analysis of alcohol’s effects on the hypothalamic-pituitary-gonadal axis under controlled conditions. Once the primary site or sites of alcohol’s toxic effects on the female hypothalamic-pituitary-gonadal axis have been clarified, analysis of the mechanisms by which alcohol intoxication disrupts this system should be greatly facilitated. Rapid advances in basic endocrinology during the past decade have resulted in an improved understanding of the neuroendocrine regulation of the menstrual cycle and shown the importance of the pulsatile pattern of gonadotropin secretory activity. The availability of sophisticated techniques for computerized analysis of hormone pulse frequency offers an important new dimension in endocrine research. Advances in basic pharmacology have resulted in the development of opioid antagonist drugs and synthesis of LHRH. Provocative tests of hypothalamic, pituitary, and gonadal function provide an opportunity to examine alcohol’s specific effects on each component of this complex interrelated system. Many questions are now unanswered, but systematic studies of alcohol’s effects should eventually clarify how alcohol, the most common intoxicant in contemporary society, can so profoundly disrupt reproductive function. As the mechanisms of alcohol’s effects on neuroendocrine regulation of the menstrual cycle are elucidated, this should also have important implications for understanding how alcohol intoxication increases risk for abortion and influences fetal growth and development.


Selections from the book: “Recent Developments in Alcoholism. Volume 6: Posttraumatic Stress Disorder. The Workplace. Consequences in Women. Markers for Risk.” Edited by Marc Galanter. An Official Publication of the American Medical Society on Alcoholism, the Research Society on Alcoholism, and the National Council on Alcoholism. 1986.