Testosterone (T) treatment suppresses serum gonadotropins and reduces sperm output sufficiently for contraceptive efficacy in ~70% of Caucasian men. In the remaining 30% of men, an increase in 5α-reductase activity may maintain testicular androgen activity, thus accounting for the failure of sperm suppression. The form of T therapy is a major consideration in the safety and acceptability of T-based contraception. As compared to T ester injections, T implants provide a more physiological serum T profile and fewer side effects, but have not yet been used in contraceptive efficacy studies. We have used T implants (800-1200 mg every 3 months) in 29 normal men for 3-16 months. T implants produced long-term suppression of sperm densities below 1 million/mL in ~70% of men without significant androgenic side effects. No pregnancies occurred in 214 months of exposure. In 16 men failing to suppress within 3 months of T 800 mg, no evidence of enhanced spermatogenic suppression was seen with the co-administration of the type 2 5α-reductase inhibitor, finasteride, for 3 months when compared to placebo. We conclude that: 1) T implants provide adequate spermatogenic suppression in ~70% of Caucasian men, a rate comparable to intramuscular T injections but with minimal side effects; and, 2) the inclusion of a type 2 5α-reductase inhibitor does not enhance spermatogenic suppression. Copyright (C) 2000 Elsevier Science Inc.