Windfeld-Mathiasen · Circulation 2025 · matched cohort study · n=60639

Cardiovascular Disease in Anabolic Androgenic Steroid Users.

Cited 55 times in the scientific literature.

Level 3 - non-randomized controlled study

Retrospective matched cohort study linked to nationwide registries

PubMed 39945117 · doi:10.1161/CIRCULATIONAHA.124.071117 · record verified 2026-08-31

What was done

Men sanctioned for anabolic androgenic steroid (AAS) use in Danish fitness center antidoping programs between 2006 and 2018 (n=1,189) were matched 1:50 by age and sex to general population controls (n=59,450). Nationwide Danish registries provided data on medical diagnoses, prescriptions, education, and occupational status through June 30, 2023 (mean follow-up: 11 years). The study evaluated adjusted hazard ratios (aHR) for acute myocardial infarction, coronary revascularization (PCI/CABG), venous thromboembolism, ischemic stroke, arrhythmia, cardiomyopathy, heart failure, and cardiac arrest.

What was found

AAS users had significantly higher incidence rates of major cardiovascular conditions compared with matched controls: - Acute myocardial infarction: aHR 3.00 (95% CI, 1.67–5.39) - PCI or CABG: aHR 2.95 (95% CI, 1.68–5.18) - Venous thromboembolism: aHR 2.42 (95% CI, 1.54–3.80) - Arrhythmia: aHR 2.26 (95% CI, 1.53–3.32) - Cardiomyopathy: aHR 8.90 (95% CI, 4.99–15.88) - Heart failure: aHR 3.63 (95% CI, 2.01–6.55) Ischemic stroke and cardiac arrest had too few cases among AAS users to be reported.

Why it matters

This study provides long-term, population-matched epidemiological evidence that illicit AAS use substantially increases the risk of severe cardiovascular disease, showing a nearly ninefold increase in cardiomyopathy and a threefold increase in myocardial infarction.

Limits

The cohort of AAS users was defined solely by antidoping sanctions in fitness centers, which may select for a specific subset of users and miss undetected users in the control group. Registry data lacked details on specific AAS compounds, dosages, cycles, duration of use, concurrent illicit substance use, smoking, and dietary habits, leaving potential for residual confounding. Case counts for cardiac arrest and ischemic stroke were insufficient for analysis.

Cited by