Exercise-induced cardiovascular remodelling in a large cohort of female, elite athletes: towards sex-specific CMR reference ranges

Abstract

Abstract Funding Acknowledgements Type of funding sources: Other. Main funding source(s): Amsterdam Movement Sciences (AMS); Dutch National Olympic Committee & National Sports Federation (NOC*NSF); Heart To Handle. Background Differentiating between exercise-induced cardiac remodelling (EICR) and pathology constitutes a central challenge in sports cardiology. To facilitate this differentiation, reference ranges for cardiovascular magnetic resonance imaging (CMR) are needed. However, female athletes, and especially, female elite athletes with potentially the most outspoken adaptation, are severely underrepresented. Moreover, no large studies including CMR data in female, elite athletes are currently available. Purpose To quantify EICR on CMR in a large cohort of female, elite athletes, compared with currently available reference values of the general female, and female athlete population. Methods We performed a cross-sectional CMR analysis in female elite athletes aged ≥16 years, included in the ELITE cohort. We excluded athletes with known cardiovascular disease. The primary outcome was EICR quantification as BSA indexed RV and LV end-diastolic volume (EDVi), LV wall mass (LVMi), LV remodelling index (LVMi/LVEDVi), and LV/RV ratio (LVEDi/RVEDVi). Second, we compared our parameters with reference ranges (1,2). CMR was performed according to a uniform protocol, and included cine-imaging and delayed hyperenhancement, preferentially on 1.5T. A dedicated core-lab analysed all CMRs in Circle Cardiovascular Imaging. Results We included 102 female elite athletes, 97% caucasian, mean age of 26.3 ±5.0, BSA 1.79 ±0.14 m2, and mean professional athlete years of 10.3 ±5. Main athlete disciplines (≥10 hours/week) were field hockey (15%), rowing (13%), road cycling (12%), and European style football (10%). Female elite athletes had marked EICR as compared with general population reference values, with higher LVEDVi (108 ±13.9 vs 69 ±12 ml/m2, p<0.05), RVEDVi (110 ±15.3 vs 76 ±14 ml/m2, p<0.05), and LVMi (49.9 ±11.2 vs 45 ±7 g/m2, p<0.05) (Figure 1). Compared with current female elite athlete references (n=33), our female elite athlete population demonstrated a markedly lower RVEDVi (118 ±17 vs 110 ±15 ml/m2), and a higher upper-limit (95th percentile) of LVMi (66 vs 72.3 g/m2) with comparable LVEDVi. LV remodelling showed a lower LVM/LVEDV ratio (0.46 ±0.08 vs 0.7 ±0.1 ml) compared to the general population, with balanced dilatation (LVEDV/RVEDV=0.98 ±0.05). In general, we observed EICR as increased cardiac volumes in 67% (n=68), increased cardiac volumes and mass in 21% (n=21) lone increase in cardiac mass in 1% (n=1), with 11% (n=11) demonstrating normal geometry (Figure 2). Conclusion EICR on CMR in female, elite athletes is mainly characterised by isolated increased volumes, with a considerable proportion (11%) demonstrating no EICR. Compared with the general population, female athletes have larger cardiac ventricular volumes and wall mass; compared with available elite athlete reference ranges comparable ventricular volumes, but a larger spread in cardiac wall mass. Our results constitute a first step towards sex-specific CMR reference ranges for female athletes.