Clinical Endocrinology

Androgens improve cavernous vasodilation and response to sildenafil in patients with erectile dysfunction


Antonio Aversa, Associazione Fatebenefratelli per la Ricerca biomedica e sanitaria, Lungotevere degli Anguillara 12–00186 Rome, Italy. Tel: +39 066837300; Fax: +39 064455345. E-mail:


objectives We have recently shown that, in men with erectile dysfunction (ED), free testosterone (FT) directly correlates with penile arterial inflow. This led us to further investigate the effect(s) of androgen administration on cavernous arteries in patients failing sildenafil treatment.

design Prospective randomized placebo-controlled pilot study.

patients Twenty patients with arteriogenic ED as evaluated by dynamic colour duplex ultrasound (D-CDU) studies, normal sexual desire but testosterone (T) and FT in the lower quartile of normal range (low-normal), not responding to sildenafil treatment (100 mg) on six consecutive attempts.

measurements All patients had D-CDU, hormonal [LH, prostate-specific antigen (PSA), total and free testosterone, sex hormone-binding protein (SHBG), oestradiol], biochemical [haematocrit, low-density lipoprotein (LDL) and HDL cholesterol, triglycerides], and sexual evaluations [International Index of Erectile Function (IIEF)] before and after 1 month of therapy with transdermal testosterone (5 mg/day, n = 10) or placebo along with sildenafil treatment on demand. Measurement of flow parameters by D-CDU on cavernous arteries was the primary endpoint of the study. Improvement of erectile function was assessed using the IIEF questionnaire and the Global Assessment Question (GAQ).

results One month treatment with transdermal testosterone led to a significant increase in T and FT levels (23·7 ± 3·3 SD vs. 12·8 ± 2·1 nmol/l and 473 ± 40·2 vs. 260 ± 18·1 pmol/l, P < 0·01, respectively). In addition testosterone administration induced a significant increase in arterial inflow to cavernous arteries measured by D-CDU (32 ± 3·6 vs. 25·2 ± 4 cm/s, P < 0·05), with no adverse effects. Also, a significant improvement in erectile function domain score at IIEF was found in the androgen but not in the placebo-treated patients (21·8 ± 2·1 vs. 14·4 ± 1·4, P < 0·05) which was associated with significant changes in the GAQ score (80%vs. 10%, P < 0·01).

conclusions In patients with arteriogenic ED and low-normal androgen levels, short-term testosterone administration increases T and FT levels and improves the erectile response to sildenafil likely by increasing arterial inflow to the penis during sexual stimulation.