Abstract Background Cross-sectional studies suggest an association between use of anabolic-androgenic steroids (AAS) and structural and functional myocardial changes. Whether drug discontinuation may attenuate these changes remains debated due to conflicting results and methodological limitations. Purpose To assess myocardial changes in AAS users, compared to strength-trained non-using athletes, during long-term follow-up. Methods In this prospective cohort study, male AAS users and strength-trained male non-users (controls) were included and examined at two time points. AAS use as well as non-use were verified by blood and urine analyses. Transthoracic echocardiography and laboratory tests were performed at both time points. Results A cohort of 32 AAS users (median age 33) and 13 controls (median age 34) were followed for a median of 16 and 13 years, respectively. At baseline, AAS users had a median AAS usage duration of 5-10 years. At follow-up, 15 had stopped AAS use (discontinued use), while 17 remained continued users. Among those who had discontinued AAS use, the median time since AAS cessation was 12 years. At baseline, AAS users displayed concentric left ventricular (LV) hypertrophy, a remodeling pattern that was absent among controls. LV systolic function was impaired in AAS users compared to controls, with LV ejection fraction (LVEF) at 49% (44, 53) vs. 53% (51, 56) (p=0.05), and global longitudinal strain (GLS) at -14.5% (-15.1, -11.5) vs. -18.6% (-19.5, -16.7) (p0.01). At follow-up, LV mass index (LVMi) was reduced compared to baseline values both in those with continued AAS use and those who had discontinued use, but the reduction was most pronounced in those who had discontinued AAS use (figure). Compared to controls, LV systolic function remained significantly impaired in AAS users who continued use: LVEF was 50% (44, 51) and 54% (52, 55) in controls (p0.01). In the group who discontinued AAS use, LVEF was 52% (51, 55) suggesting a slight improvement (figure). GLS was -18.3% (-19.1, -17.4) in controls, -13.9% (-15.1, -11.8) in those who continued AAS use, and -16.6% (-19.3, -15.6) in the group who discontinued use. The median change in LVEF over time (LVEFΔ) differed significantly between those who continued AAS use and those who discontinued: -2 (-6, 2) vs. 3 (1, 8), p0.01. Conclusions Long-term AAS use was associated with myocardial remodeling and LV dysfunction. AAS users who had discontinued use during follow-up showed partial recovery of LV systolic dysfunction, even after more than a decade of AAS use.
Fyksen et al. (Sat,) studied this question.