Does Testosterone Therapy Increase Risk of Cardiovascular Events among Men? A Meta-analysis

Importance: There has been increasing interest in use of testosterone therapy (TT) beyond patients with hypogonadism to include younger men without documented hormone measurements for the purpose of improving libido, sexual function, bone density, and body mass. However, there is no conclusive data about safety of TT due to lack of adequately powered randomized clinical trials (RCTs) specifically designed for this purpose. Objective: To examine the overall risk of cardiovascular events associated with TT via meta-analysis of published randomized and observational studies. Data Sources: We searched MEDLINE, EMBASE, CINAHL, the Cochrane Controlled Trials Register and the National Institute of Health Clinical Trials.gov database from 1966 to 2014. Study Selection: Out of the initial 2,800 studies identified, we obtained a total of 34 studies for detailed analysis after applying our inclusion/exclusion criteria. Two reviewers used eligibility criteria to assess all titles, abstracts, and full texts and resolved disagreements by discussion. Data Extraction and Synthesis: One reviewer did data abstractions and quality assessments, which were confirmed by a second reviewer. Data were then collected and analyzed using random and fixed effect model, as appropriate. Risk estimates were extracted as adjusted hazard ratios (HRs) from included studies. Main Outcome and Measures: Association of TT with cardiovascular events as a primary endpoint and association of TT with ischemic heart disease, all-cause mortality and cerebrovascular events as secondary endpoints. Results: TT was associated with increased incidence of cardiovascular events (adjusted hazard ratio (HR) = 1.41, 95% CI = 1.19-1.67, p = 0.0004), all-cause mortality (adjusted HR = 1.29, 95% CI = 1.03-1.62, p = 0.02), and ischemic heart disease (adjusted HR=1.51, 95% CI = 1.05-2.18, p = 0.02) but there was no clear association with cerebrovascular events (adjusted HR=0.91, 95% CI = 0.66-1.25, p=0.54).  Subgroup analyses of our primary endpoint by study type (randomized versus observational studies) did not change our results (adjusted HR=1.40, 95% CI = 1.05-1.87, p = 0.02 and adjusted HR=1.54, 95% CI = 1.09-2.17, p = 0.01 respectively). Additional analysis using meta-regression and sensitivity analyses to account for factors such as history of prior CV events, indication for TT and duration of follow up did not change our results. However, we did notice lack of association between CV events and Intramuscular testosterone. Conclusions and Relevance: TT may be associated with an increased risk of all-cause mortality, cardiovascular events, and ischemic heart disease. These findings support the need for an adequately powered randomized study.