In order to investigate the endocrine etiology of benign prostatic hyperplasia (BPH), we conducted a case-control study in Athens, Greece using 52 patients with histologically confirmed BPH and 52 healthy controls matched according to age and town of residence. Blood samples were collected from all participants and analyzed blindly in Boston, Mass. regarding testosterone (T), dihydrotestosterone, estradiol (E2), sex hormone-binding globulin, and dehydroepiandrosterone sulfate (DHEAS). Results from logistic regression models, adjusting for age, height, body mass index, years of schooling, and mutually among the measured hormones, indicate that DHEAS is significantly positively associated with the risk of BPH [odds ratio = 3.10 per standard deviation (60 micrograms/dl), 95% confidence interval (1.28,7.50)]. T and E2 were not significantly related to the risk of BPH.