EFFECTS OF ALCOHOL-CONSUMPTION ON PLASMA AND URINARY HORMONE CONCENTRATIONS IN PREMENOPAUSAL WOMEN

Background: Most epidemiologic studies of the relationship between alc ohol consumption and breast cancer risk over the past decade have show n that persons who consume a moderate amount of alcohol are at 40%-100 % greater risk of breast cancer than those who do not consume alcohol. Dose-response effects have been observed, but no causal relationship has been established. Purpose: This study examines the hypothesis that alcohol consumption affects levels of reproductive hormones. Methods: A controlled-diet study lasting for six consecutive menstrual cycles was conducted. Participants were randomly assigned to two groups, and a crossover design was used. During the last three menstrual cycles, a lcohol consumption of the two groups was reversed. Thirty-four premeno pausal women, aged 21-40 years, with a history of regular menstrual cy cles, consumed 30 g of ethanol (equivalent to approximately two averag e drinks) per day for three menstrual cycles and no alcohol for the ot her three. All food and alcohol consumed were provided by the study. C aloric intake was monitored to ensure that each woman would maintain b ody weight at approximately the baseline level. Hormone assays were pe rformed on pooled plasma or 24-hour urine specimens collected during t he follicular (days 5-7), peri-ovulatory (days 12-15), and mid-luteal (days 21-23) phases of the third menstrual cycle for subjects on each diet. Results: Alcohol consumption was associated with statistically s ignificant increases in levels of several hormones. Plasma dehydroepia ndrosterone sulfate levels were 7.0% higher in the follicular phase (P = .05). In the peri-ovulatory phase, there were increases of 21.2% (P = .01) in plasma estrone levels, 27.5% (P = .01) in plasma estradiol levels, and 31.9% (P = .009) in urinary estradiol levels. In the lutea l phase, urinary estrone levels rose 15.2% (P = .05), estradiol levels increased 21.6% (P = .02), and estriol levels rose 29.1 % (P = .03). No changes were found in the percent of bioavailable estradiol, define d by the sum of percent free estradiol and percent albumin-bound estra diol. However, increased total estradiol levels in the peri-ovulatory phase suggest elevated absolute amounts of bioavailable estradiol. Con clusion: This study has shown increases in total estrogen levels and a mount of bioavailable estrogens in association with alcohol consumptio n in premenopausal women. Implication: This possible explanatory mecha nism for a positive association between alcohol consumption and breast cancer risk merits further investigation.