Higher Risk for Cardiovascular Events Found With Testosterone Injections Versus Gels

Introduction: Given recent concerns regarding a potentially increased risk of cardiovascular events among men using testosterone supplementation, this study was designed to compare the cardiovascular safety of 3 forms of testosterone: injections, patches, and gels.

Methods: This retrospective cohort study involved data from 544,115 men (age ≥18 years) who initiated use of testosterone injection (37.4%), patch (6.9%) and gel (55.8%) following 180 days of no testosterone use. The data were derived from commercially insured and Medicare populations in the United States (US) and from general practitioner records in the United Kingdom (UK). All participants were followed for up to 1 year.

Results: Pooled data from the various populations showed that adjusted hazard ratios for various cardiovascular outcomes were higher among users of testosterone injection versus gel (Table). While the increased risk for cardiovascular outcomes among injection users was consistently found among the three databases, the absolute incidence of the composite outcome of myocardial infarction, angina, and stroke was relatively small during the 1 year follow-up period, ranging from 23.1 to 36.6 events, per 1,000 person-year with gels and injections, respectively, in the Medicare cohort. The absolute incidence was lower in the other cohorts.

The testosterone patch was linked to a slightly increased hazard ratio for myocardial infarction compared with the testosterone gel (adjusted hazard ratio, 1.21), but all other outcomes were similar. The hazard ratio for venous thromboembolism was similar among the three dosage forms.

Conclusion: Use of testosterone injections was associated with a higher risk for cardiovascular events (myocardial infarction, unstable angina, and stroke), hospitalizations, and death compared with use of testosterone gels. Patches and gels had a similar risk profile.


Layton JB, Meier CR, Sharpless JL, et al. Comparative safety of testosterone dosage forms. JAMA Intern Med. 2015;175(7):1187-1196.

Commentary by Tamara L Wexler MD, PhD

Tamara L. Wexler, MD, PhD, is an endocrinologist specializing in neuroendocrinology and reproductive endocrinology. She is the Director of the NYU Langone Medical Center Pituitary Center in New York, NY, as well as an Attending in Medicine at Massachusetts General Hospital, Boston, MA.

In the past few years, there have been several reports of increased cardiovascular risk with testosterone use. We reviewed some of this data in the March 2014 EndoScan , from the TOM (Testosterone in Older Men with Mobility Limitations) trial, in which a higher rate of cardiovascular (CV) adverse events noted in the testosterone vs placebo group led to early termination,1 to two large retrospective observational studies,2,3 which found a higher rate of CV events in men on testosterone replacement, and underscored the need for larger and longer randomized controlled trials.

As always, it is essential to consider the particular cohort studied in determining generalizability of the results:
• The average age of men in the TOM study was 74 years;
• Angiography was among the inclusion criteria in one of the retrospective observational studies, a VA-based study by Vigen and colleagues;2 and
• Findings in the other observational study, a claims database-based study by Finkle and colleagues,3 were limited to men >65 years of age or with a history of CV disease.

Particularly relevant is whether the subjects met Endocrine Society4 or AACE5 guidelines for replacement—definitively low testosterone (T) measured with two morning values and, with consideration for BMI, in the presence of symptoms—or whether T treatment was initiated without establishing morning serum levels below the normal reference range.Layton and colleagues had previously reported on the use of testosterone in men in the US and UK without evidence of low testosterone or even of testing.6

In this 2016 publication, Layton and colleagues examine whether the preparation of testosterone used is related to increased cardiovascular risk in a retrospective observational study using US insurance claims records and UK General Practitioners' records. Information on the use of testosterone, type of testosterone preparation, and incidence of CV events over one year of testosterone treatment, was based solely on the data in these records. Cardiovascular events monitored included myocardial infarction, unstable angina, stroke, and composite of these 3 acute events; venous thromboembolism; mortality; and all-cause hospitalization.

Layton and colleagues found that all preparations of testosterone (injection, patch, or gel) were associated with some increase in the events studied. The relative risk was greatest in men using injected testosterone. This may be related to the pharmacokinetics of injections, with relative peaks after each injection, versus the more consistent absorption with gels and patches.

The study considered only the first year on testosterone, with a relatively small incidence of events in the population. The Medicare population had the largest number of events, but still found only 23.1 (with gels) to 36.6 events (with injections) per 1,000 person-years. The relatively higher incidence in the Medicare population may be related to the frequent use of injections versus other preparations.

Of note, this study did not include non-testosterone users as a control to establish an expected baseline rate of events in these populations. In addition, as with any database-based study, the information gathered is only as reliable as the input, and diagnostic codes may well be incomplete or inaccurate.7

Most importantly, as noted by the authors, there is no indication whether the subjects had testosterone deficiency at the outset. Thus, results should not be used to infer an increased rate of cardiovascular or other events in users of testosterone versus the general population, but rather offer information on the relative risk of the 3 preparations studied. It is also important to note that the different populations (commercially insured US; Medicare US, UK) tended toward different preparations and the results may therefore reflect both the underlying population as well as the preparation.

1. Basaria S, Coviello AD, Travison TG, et al. Adverse events associated with testosterone administration. N Engl J Med. 2010;363:109-122.

2. Vigen R, O'Donnell CI, Barón AE, et al. Association of testosterone therapy with mortality, myocardial infarction, and stroke in men with low testosterone levels. JAMA. 2013;310(17):1829-1836.

3. Finkle WD, Greenland S, Ridgeway GK, et al. Increased risk of non-fatal myocardial infarction following testosterone therapy prescription in men. PLoS One. 2014 Jan 29;9(1):e85805.

4. Bhasin S, Cunningham GR, Hayes FJ, et al. Testosterone therapy in men with androgen deficiency syndromes: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2010;95(6):2536-2559.

5. Goodman N, Guay A, Dandona P, Dhindsa S, Faiman C, Cunningham GR; AACE Reproductive Endocrinology Scientific Committee. American Association of Clinical Endocrinologists and American College of Endocrinology Position Statement on the association of testosterone and cardiovascular risk. Endocr Pract. 2015;21(9):1066-1073.

6. Layton JB, Li D, Meier CR, et al. Testosterone lab testing and initiation in the United Kingdom and the United States, 2000-2011. J Clin Endocrinol Metab. 2014 Jan 1:jc20133570 [Epub ahead of print].

7. Klabunde CN, Warren JL, Legler JM. Assessing comorbidity using claims data: An overview. Med Care. 2002;40(8)suppl:IV-26-35.