Enhanced 5 alpha-reductase activity has been found in the skin of the majority of women with hirsutism. Finasteride is a specific competitive inhibitor of 5 alpha-reductase, preferentially inhibiting the type 2 isoenzyme. Therefore, we randomly assigned 14 hirsute women in a 2:1 ratio to 1 of 2 treatment arms: 1) finasteride (F) treatment (n = 9; 5 mg, orally, daily), or 2) spironolactone (S) treatment (n = 5; 100 mg, orally, daily). Each group was treated for 6 months. Patients were evaluated at baseline and after 3 and 6 months of treatment. The 2 groups were similar in age, weight, hip/waist ratio, baseline Ferriman-Gallwey score (F, 19 +/- 2; S, 19 +/- 2), and baseline androgen levels. Finasteride treatment resulted in a significant increase in testosterone (T; P < 0.01) and the T/dihydrotestosterone ratio (P < 0.01). Finasteride caused a significant decrease in 5 alpha-androstane-3 alpha,17 beta-diol glucuronide (3 alpha-diolG; P < 0.05), the 3 alpha-diolG/T ratio (P < 0.01), and the 3 alpha-diolG/androstenedione ratio (P < 0.05). All changes were consistent with 5 alpha-reductase inhibition. In contrast, spironolactone treatment did not result in significant changes in serum hormone levels. Both treatments produced a significant decrease in anagen hair diameters [F, -14.0 +/- 6.7% (P < 0.05); S, -13.4 +/- 3.8% (P < 0.05)] and Ferriman-Gallwey scores [F, -2.1 +/- 0.4 (P < 0.05); S, -2.5 +/- 0.7 (P < 0.05)]. In conclusion, despite significantly different effects on androgen levels, finasteride and spironolactone treatment resulted in a similar clinical effect on hirsutism. Both caused significant, but limited, improvement in hirsutism. Although promising, further studies with finasteride are needed to verify its effectiveness as a treatment for hirsutism. Such studies will also provide a better understanding of the relative contribution of 5 alpha-reductase isoenzymes to hirsutism.