In view of the role of both the de novo biosynthesis and receptor-mediated uptake of cholesterol for normal steroidogenesis, we evaluated whether extending the therapeutic dose of the hepatic hydroxymethyl glutaryl coenzyme A (HMG-CoA) reductase inhibitor, simvastatin, to 80 mg/d would affect adrenal and gonadal steroid synthesis in men with hypercholesterolemia. To evaluate this question, we enrolled men into a multicenter randomized, placebo-controlled study lasting 12 weeks. Men with serum low-density lipoprotein cholesterol (LDL-C) more than 145 mg/dL after 6 weeks of a lipid-lowering diet were randomized to 80 mg simvastatin or placebo. Half of the subjects were asked to undergo a 6-hour infusion of corticotropin (ACTH) to evaluate cortisol synthesis, and the entire cohort received a human chorionic gonadotropin (hCG) stimulation test to assess gonadal hormone secretion using pooled serum samples taken 15 minutes apart. A total of 81 men (age, 45 +/- 11 years; 93% Caucasian) with baseline serum LDL-C of 197 mg/dL (placebo, n = 39) and 184 mg/dL (simvastatin 80 mg, n = 42) completed the study. After 12 weeks, serum LDL-C, triglycerides, and high-density lipoprotein cholesterol (HDL-C) in the simvastatin group changed by -43%, -25%, and 8%, respectively (all P < .001). The basal cortisol level and the peak serum cortisol and area under the curve response to the 6-hour ACTH infusion were comparable between the two treatment groups at baseline and after 12 weeks. The pooled total testosterone level at baseline was 541 and 513 ng/dL in the placebo and simvastatin-treated groups, respectively, which declined to 536 +/- 20.5 ng/dL (-1.5%) and 474 +/- 30.4 ng/dL (-13.6%, P = .09) after treatment (mean +/- SD). The pooled free testosterone declined by 6.3% in the simvastatin group, versus a 4.9% increase in the placebo group (P = .588), while pooled bioavailable testosterone declined 10.2% in the simvastatin group and increased 1.4% in the placebo group (P = .035). There were no changes in serum gonadotropin levels or sex hormone-binding globulin (SHBG). After administration of hCG, there were no differences in the peak total pooled testosterone level before or after 12 weeks of treatment. Simvastatin 80 mg was well tolerated compared with placebo. In conclusion, basal and stimulated cortisol production was unaffected by the use of simvastatin 80 mg versus placebo. As reported with other statins and cholestyramine, there were small declines in the simvastatin-treated group for pooled total, free, and bioavailable testosterone after 12 weeks, although there was no compensatory increase in serum follicle-stimulating hormone (FSH) or luteinizing hormone (LH) levels.