Association between exogenous testosterone and cardiovascular events: an overview of systematic reviews

Given the conflicting evidence regarding the association between exogenous testosterone and cardiovascular events, we systematically assessed published systematic reviews for evidence of the association between exogenous testosterone and cardiovascular events. We searched PubMed, MEDLINE, Embase, Cochrane Collaboration Clinical Trials, ClinicalTrials.gov, and the US Food and Drug Administration website for systematic reviews of randomised controlled trials published up to July 19, 2016. Two independent reviewers screened 954 full texts from 29 335 abstracts to identify systematic reviews of randomised controlled trials in which the cardiovascular effects of exogenous testosterone on men aged 18 years or older were examined. We extracted data for study characteristics, analytic methods, and key findings, and applied the AMSTAR (A Measurement Tool to Assess Systematic Reviews) checklist to assess methodological quality of each review. Our primary outcome measure was the direction and magnitude of association between exogenous testosterone and cardiovascular events. We identified seven reviews and metaanalyses, which had substantial clinical heterogeneity, differing statistical methods, and variable methodological quality and quality of data abstraction. AMSTAR scores ranged from 3 to 9 out of 11. Six systematic reviews that each included a meta-analysis showed no significant association between exogenous testosterone and cardiovascular events, with summary estimates ranging from 1.07 to 1.82 and imprecise confidence intervals. Two of these six metaanalyses showed increased risk in subgroup analyses of oral testosterone and men aged 65 years or older during their first treatment year. One meta-analysis showed a significant association between exogenous testosterone and cardiovascular events, in men aged 18 years or older generally, with a summary estimate of 1.54 (95% CI 1.09-2.18). Our optimal information size analysis showed that any randomised controlled trial aiming to detect a true difference in cardiovascular risk between treatment groups receiving exogenous testosterone and their controls (with a two-sided p value of 0.05 and a power of 80%) would require at least 17 664 participants in each trial group. Therefore, given the challenge of adequately powering clinical trials for rare outcomes, rigorous observational studies are needed to clarify the association between testosterone-replacement therapy and major adverse cardiovascular outcomes.