Testosterone supplementation and sexual function: a meta-analysis study

• Published in: Journal of sexual medicine Vol. 11; no. 6; p. 1577
• Main Authors: Corona, Giovanni, Isidori, Andrea M, Buvat, Jaques, Aversa, Antonio, Rastrelli, Giulia, Hackett, Geoff, Rochira, Vincenzo, Sforza, Alessandra, Lenzi, Andrea, Mannucci, Edoardo, Maggi, Mario
• Format: Journal Article
• Language: English
• Published: Netherlands 01.06.2014
• Subjects: Adult; Androgens - therapeutic use; Combined Modality Therapy; Dietary Supplements; Humans; Libido - drug effects; Male; Penile Erection - drug effects; Phosphodiesterase 5 Inhibitors - therapeutic use; Randomized Controlled Trials as Topic; Sexual Behavior - drug effects; Sexual Dysfunction, Physiological - therapy; Testosterone - therapeutic use; Erectile Dysfunction; Testosterone; Libido; Orgasm; PDE5i
• ISSN: 1743-6109
• DOI: 10.1111/jsm.12536

The role of testosterone supplementation (TS) as a treatment for male sexual dysfunction remains questionable. The aim of this study was to attempt a meta-analysis on the effect of TS on male sexual function and its synergism with the use of phosphodiesterase type 5 inhibitor (PDE5i). An extensive Medline, Embase, and Cochrane search was performed. All randomized controlled trials (RCTs) comparing the effect of TS vs. placebo or the effect of TS as add on to PDE5is on sexual function were included. Data extraction was performed independently by two of the authors (A. M. Isidori and G. Corona), and conflicts resolved by the third investigator (M. Maggi). Out of 1,702 retrieved articles, 41 were included in the study. In particular, 29 compared TS vs. placebo, whereas 12 trials evaluated the effect of TS as add on to PDE5is. TS is able to significantly ameliorate erectile function and to improve other aspects of male sexual response in hypogonadal patients. However, the presence of possible publication bias was detected. After applying "trim and fill" method, the positive effect of TS on erectile function and libido components retained significance only in RCTs partially or completely supported by pharmaceutical companies (confidence interval [0.04-0.53] and [0.12; 0.52], respectively). In addition, we also report that TS could be associated with an improvement in PDE5i outcome. These results were not confirmed in placebo-controlled studies. The majority of studies, however, included mixed eugonadal/hypogonadal subjects, thus imparting uncertainty to the statistical analyses. TS plays positive effects on male sexual function in hypogonadal subjects. The role of TS is uncertain in men who are not clearly hypogonadal. The apparent difference between industry-supported and independent studies could depend on trial design more than on publication bias. New RCTs exploring the effect of TS in selected cases of PDE5i failure that persistently retain low testosterone levels are advisable.