Safety and efficacy of testosterone for women: a systematic review and meta-analysis of randomised controlled trial data

Background The benefits and risks of testosterone treatment for women with diminished sexual wellbeing remain controversial. We did a systematic review and meta-analysis to assess potential benefits and risks of testosterone for women. Methods We searched MEDLINE, Embase, the Cochrane Central Register of Controlled Trials, and Web of Science for blinded, randomised controlled trials of testosterone treatment of at least 12 weeks' duration completed between Jan 1, 1990, and Dec 10, 2018. We also searched drug registration applications to the European Medicine Agency and the US Food and Drug Administration to identify any unpublished data. Primary outcomes were the effects of testosterone on sexual function, cardiometabolic variables, cognitive measures, and musculoskeletal health. This study is registered with the International Prospective Register of Systematic Reviews (PROSPERO), number CRD42018104073. Findings Our search strategy retrieved 46 reports of 36 randomised controlled trials comprising 8480 participants. Our meta-analysis showed that, compared with placebo or a comparator (eg, oestrogen, with or without progestogen), testosterone significantly increased sexual function, including satisfactory sexual event frequency (mean difference 0·85, 95% CI 0·52 to 1·18), sexual desire (standardised mean difference 0·36, 95% CI 0·22 to 0·50), pleasure (mean difference 6·86, 95% CI 5·19 to 8·52), arousal (standardised mean difference 0·28, 95% CI 0·21 to 0·35), orgasm (standardised mean difference 0·25, 95% CI 0·18 to 0·32), responsiveness (standardised mean difference 0·28, 95% CI 0·21 to 0·35), and self-image (mean difference 5·64, 95% CI 4·03 to 7·26), and reduced sexual concerns (mean difference 8·99, 95% CI 6·90 to 11·08) and distress (standardised mean difference −0·27, 95% CI −0·36 to −0·17) in postmenopausal women. A significant rise in the amount of LDL-cholesterol, and reductions in the amounts of total cholesterol, HDL-cholesterol, and triglycerides, were seen with testosterone administered orally, but not when administered non-orally (eg, by transdermal patch or cream). An overall increase in weight was recorded with testosterone treatment. No effects of testosterone were reported for body composition, musculoskeletal variables, or cognitive measures, although the number of women who contributed data for these outcomes was small. Testosterone was associated with a significantly greater likelihood of reporting acne and hair growth, but no serious adverse events were recorded. Interpretation Testosterone is effective for postmenopausal women with low sexual desire causing distress, with administration via non-oral routes (eg, transdermal application) preferred because of a neutral lipid profile. The effects of testosterone on individual wellbeing and musculoskeletal and cognitive health, as well as long-term safety, warrant further investigation. Funding Australian National Health and Medical Research Council.