Alcohol consumption and insulin concentrations. Role of insulin in associations of alcohol intake with high-density lipoprotein cholesterol and triglycerides.

BACKGROUND The relation between alcohol intake and insulin levels may explain, in part, the reported associations of alcohol with cardiovascular disease risk factors, including high-density lipoprotein (HDL) cholesterol, triglycerides, blood pressure, and glucose levels, each of which has been recognized as a component of the insulin resistance syndrome. METHODS AND RESULTS Subjects included nondiabetic participants of the Kaiser Permanente Women Twins Study (1989 through 1990). Usual alcohol intake was assessed as part of a food frequency questionnaire. For women from twin pairs in which both twins drank (n = 338), an increment of 12 g of alcohol per day (about one drink) was associated with an 8% lower 2-hour post-glucose-load insulin (P < .01) in a multiple regression analysis for twin data, adjusted for age, body mass index, waist-to-hip ratio, total caloric intake, and family history of diabetes. With genetic influences removed by matched analysis of the subset of 98 monozygotic twin pairs, an intrapair difference of 12 g of alcohol per day was associated with a 12.4% intrapair decrement in postload insulin (P < .01). Inverse associations were also seen for fasting insulin. Alcohol consumption was inversely associated with postload glucose but not with fasting glucose in unmatched (P = .05) and matched (P = .005) analyses. A significant positive association of alcohol intake with high-density lipoprotein cholesterol and an inverse relation of alcohol intake with triglycerides were each independent of insulin levels (P < or = .02 in the matched models). Neither systolic nor diastolic blood pressures were related to alcohol consumption in this sample, perhaps because of the rather low level of alcohol intake in the study population (median, 4 g/d). CONCLUSIONS Within the range of light to moderate drinking habits, alcohol consumption was inversely related to fasting and postload insulin levels. This relation did not explain associations of alcohol intake with lipid levels and may instead reflect an additional mechanism by which moderate alcohol consumption impacts cardiovascular disease risk.