Older men with low testosterone randomly assigned to testosterone gel for 1 year saw modest decreases in cholesterol and insulin levels compared with men assigned placebo, but no changes were observed in markers of inflammation, fibrinolysis or myocardial damage, study data show.
“The effects of testosterone treatment on cardiovascular risk is uncertain,” Emile R. Mohler III, MD, of the division of cardiovascular disease at the University of Pennsylvania Perelman School of Medicine, and colleagues wrote. “Some retrospective studies using electronic medical records have reported more cardiovascular adverse events in men taking testosterone than in men not taking it, but others have not. ... Meta-analyses of clinical trials have generally not shown more adverse cardiovascular events in men taking testosterone than in men taking placebo, but none of the individual trials was designed prospectively to capture these events.”
Mohler and colleagues analyzed data from 788 men aged at least 65 years with low testosterone (average of two serum testosterone levels of 275 ng/dL or less) enrolled in the testosterone trials (T Trials), a group of seven double-blind, placebo-controlled trials designed to evaluate the efficacy of testosterone therapy in men with age-related low testosterone. Researchers assigned participants to 5 g daily of 1% testosterone gel (AndroGel, AbbVie) or placebo for 1 year. Participants provided blood samples at baseline and 3, 6, 9 and 12 months to assess lipid profile, glucose concentration, and levels of serum insulin, HbA1c, C-reactive protein, D-dimer and troponin I. Researchers used random-effects models for longitudinal data to assess the effects of testosterone, including visit time as a categorical variable and a single main effect for treatment. The main outcome was serum markers of CV risk.
After 12 months, lipid levels decreased in both the testosterone and placebo arms, but after adjusting for baseline levels, men in the testosterone arm had a mean 6.1 mg/dL greater reduction in total cholesterol vs. men in the placebo arm (P < .001). Men assigned to testosterone also saw greater reductions in HDL cholesterol (adjusted difference, –2 mg/dL; P < .001) and in non-HDL cholesterol (adjusted difference, –4.2; P = .005) vs. men assigned placebo. Changes in LDL cholesterol were marginally greater in the testosterone arm, and there were no between-group differences in triglyceride levels, according to researchers. Results persisted in a separate analysis, excluding 15 men who initiated lipid-lowering therapy after baseline.
Testosterone slightly decreased fasting insulin levels (adjusted mean difference, –1.7 IU/mL; P = .02) and homeostatic model assessment of insulin resistance (adjusted mean difference, –0.6; P = .03) when compared with placebo. There were no between-group changes observed for levels of D-dimer, C-reactive protein, interleukin-6, troponin, glucose or HbA1c.
The researchers noted that the significance of the decreases in cholesterol levels is uncertain because LDL and HDL fell slightly, as did insulin and HOMA-IR.
“In the 138 men in the T Trials who underwent CT angiography at baseline and month 12, testosterone treatment was associated with a greater increase in noncalcified coronary artery plaque volume than placebo treatment; yet in all T Trial participants, a similar number of men (seven) in each treatment arm experienced major adverse cardiovascular events,” the researchers wrote. “A trial of a much larger number of men treated for a much longer time would be necessary to determine if testosterone treatment of hypogonadal men affects clinical cardiovascular risk.” – by Regina Schaffer
Disclosures: The authors report no relevant financial disclosures.