Left atrial stiffness and its association to left ventricular geometry in resistance trained athletes using anabolic-androgenic steroids

Place, F and Carpenter, H and Howard, LJ and Shardey, JKK and Morrison, BN and Chester, Neil and Cooper, R and Stansfield, B and George, Keith P and Angell, Peter J. (2025) Left atrial stiffness and its association to left ventricular geometry in resistance trained athletes using anabolic-androgenic steroids. European Heart Journal - Cardiovascular Imaging, 26 (1). ISSN 0195-668X

Text Full_draft_final.docx - Accepted Version Download (2MB)

Abstract

Background
Anabolic-Androgenic Steroids (AAS) are commonly used in resistance trained (RT) individuals despite the negative impact on left ventricular (LV) structure and function.

Purpose
Due to the inter-dependence of the left atrium (LA) and the LV, alongside the importance of the atria contributing to LV filling and cardiac output, this study aimed to assess the relationship between LV geometry and LA stiffness in young RT users of AAS.

Methods
Utilising a cross-sectional design, male (n=99) and female (n=19) RT individuals aged 29±6yrs were grouped based on self-reported AAS user status: current (CU; n=70), past (PU; n=22) and non-users (NU; n=27). All participants underwent transthoracic echocardiography with strain imaging. Left ventricular geometry was defined based on LV mass index and relative wall thickness as per European Association of Cardiovascular Imaging guidelines. Left atrial stiffness index (LASi) was calculated as the ratio of E/E’ to LA reservoir strain (LARes) and concentricity as the ratio of LV mass to end-diastolic volume^0.67. Ejection Fraction (EF) was calculated using Simpsons biplane method. All data was presented as mean ± standard deviation with group differences assessed using a one-way ANOVA with post-hoc Bonferroni adjustment and Kruskal-Wallis test for non-normally distributed indices. A Fishers exact test was used to compare geometry between groups. Relationships between LASi and concentricity were assessed using multi linear regression

Results
Current users presented greater LV remodelling compared to PU and NU (eccentric hypertrophy (45%, 9% and 0%), concentric remodelling (2%, 0% and 0%) and concentric hypertrophy (8%, 0% and 0% respectively; p<0.001). EF was lower in those with concentric (52±9, p=.006) and eccentric (53±5, p<.001) hypertrophy than with normal geometry (58±5), but no difference between concentric and eccentric hypertrophy. LARes was lower in CU (32.0±7.5) compared to PU (39.1±8.3, p=.001) and NU (40.2±7.7, p<.001). E/E’ and LASi were higher in CU than PU (p=.006, p=.001) and NU (p<.001, p<.001; E/E’: 6.7±1.8, 5.9±1.5, and 5.7±1.4; LA stiffness index: 0.21±0.08, 0.16±0.05, and 0.15±0.05). Concentricity was higher in CU than PU (p<.001) and NU (p<.001; 8.3±2.2; 6.4±1.4 and 5.7±2.2 respectively). Controlling for user status, concentricity predicted LASi (R²=0.23, F(2,109)=16.1, p<.001) with a 0.1 increase in LASi for each 10g.ml^-0.667 increase in concentricity.

Conclusions
Although current users of AAS presented with more eccentric hypertrophy, those with increased concentricity and concentric hypertrophy had higher LA stiffness index. These data highlight 1) the potential greater negative functional impact of concentric compared to eccentric hypertrophy in young RT individuals using AAS, 2) the added value of LASi above conventional indices such as EF and 3) the possible reversible nature of LV and LA remodelling following cessation of AAS use.

Item Type:	Article
Additional Information and Comments:	© The Author(s) 2025. Published by Oxford University Press on behalf of the European Society of Cardiology.
Faculty / Department:	Faculty of Human and Digital Sciences > School of Health and Sport Sciences
Depositing User:	Peter Angell
Date Deposited:	06 Mar 2026 16:28
Last Modified:	06 Mar 2026 16:28
URI:	https://hira.hope.ac.uk/id/eprint/4863

Actions (login required)

View Item