We studied a group of obese hyperandrogenic amenorrheic women to determine the effects of weight loss on anthropometry, hormonal status, menstrual cycles, ovulation, and fertility. Fourteen women had polycystic ovaries, two the hyperandrogenism-insulin resistance-acanthosis nigricans syndrome, one hirsutism of adrenal origin, and three idiopathic chronic anovulation. The duration of amenorrhea before the study ranged from 3-17 months [mean, 8.6 +/- 4.5 (+/- SD)]. All women ate a hypocaloric diet for a period of 8.0 +/- 2.4 months. Weight loss ranged from 4.8 to 15.2 kg (mean, 9.7 +/- 3.1 kg; 1.35 +/- 0.56 kg/month) and the waist to hip ratio, which was used as a measurement of body fat distribution, decreased from 0.86 +/- 0.1 to 0.81 +/- 0.06 (P less than 0.0001). The women's mean plasma testosterone and LH concentrations decreased significantly (P less than 0.001 and P less than 0.005, respectively). A significant positive correlation was found between the decreases in plasma testosterone levels and the decreases in glucose-stimulated insulin levels. Moreover, the decreases in the waist to hip ratio correlated positively with the decreases in glucose-stimulated insulin levels and inversely with the decreases in plasma 17 beta-estradiol. No relationships were found between weight loss and the changes in plasma insulin, steroid, and gonadotropin concentrations. The responsiveness to the weight reduction program was evaluated by comparing the number of menstrual cycles during the study period with the number reported before it. Eight women had significantly improved menstrual cyclicity (responders), while 12 did not (nonresponders). The clinical characteristics and hormone values were similar in responder and nonresponder women. In the group as a whole, 33% of the menstrual cycles during the study were ovulatory, and 4 pregnancies occurred. Hirsutism improved significantly in more than half of the women, as did acanthosis nigricans when present. We conclude that weight loss is beneficial in all obese hyperandrogenic women regardless of the presence of polycystic ovaries, the degree of hyperandrogenism, and the degree and distribution of obesity.