Objectives: To evaluate the possible effects of Tribulus terrestris herbal medicine in the erectile dysfunction treatment and to quantify its potential impact on serum testosterone levels. Design and methods: A prospective, randomized, double-blind and placebo-controlled study including 30 healthy men selected from 100 patients who presented themselves spontaneously complaining of erectile dysfunction, β₯40 years of age, nonsmokers, not undergoing treatment for prostate cancer or erectile dysfunction, no dyslipidemia, no phosphodiesterase inhibitor use, and no hormonal manipulation, and if present, hypertension and/or diabetes mellitus should be controlled. International Index of Erectile Function (IIEF-5) and serum testosterone were obtained before randomization and after 30 days of study. The patients were randomized into two groups of fifteen subjects each. The study group received 800 mg of Tribulus terrestris, divided into two doses per day for 30 days, and the control group received placebo administered in the same way. Results: The groups were statistically equivalent in all aspects evaluated. The mean (SD) age was 60 (9.4) and 62.9 (7.9) years, p = 0.36, for intervention and placebo groups, respectively. Before treatment, the intervention group showed mean IIEF-5 of 13.2 (5---21) and mean total testosterone 417.1 ng/dl (270.7---548.4 ng/dl); the placebo group showed mean IIEF-5 of 11.6 (6---21) and mean total testosterone 442.7 ng/dl (301---609.1 ng/dl). After treatment, the intervention group showed mean IIEF-5 of 15.3 (5---21) and mean total testosterone 409.3 ng/dl (216.9---760.8 ng/dl); the placebo group showed mean IIEF-5 of 13.7 (6---21) and mean total testosterone 466.3 ng/dl (264.3---934.3 ng/dl). The time factor caused statistically significant changes in both the groups for IIEF-5 only (p = 0.0004); however, there was no difference between the two groups (p = 0.7914).