Reduced Ejection Fraction in Elite Endurance Athletes: Clinical and Genetic Overlap With Dilated Cardiomyopathy

Abstract

Background: Exercise-induced cardiac remodeling can be profound, resulting in clinical overlap with dilated cardiomyopathy, yet the significance of reduced ejection fraction (EF) in athletes is unclear. The aim is to assess the prevalence, clinical consequences, and genetic predisposition of reduced EF in athletes.

Methods: Young endurance athletes were recruited from elite training programs and underwent comprehensive cardiac phenotyping and genetic testing. Those with reduced EF using cardiac magnetic resonance imaging (defined as left ventricular EF <50%, or right ventricular EF <45%, or both) were compared with athletes with normal EF. A validated polygenic risk score for indexed left ventricular end-systolic volume (LVESVi-PRS), previously associated with dilated cardiomyopathy, was assessed. Clinical events were recorded over a mean of 4.4 years.

Results: Of the 281 elite endurance athletes (22±8 years, 79.7% male) undergoing comprehensive assessment, 44 of 281 (15.7%) had reduced left ventricular EF (N=12; 4.3%), right ventricular EF (N=14; 5.0%), or both (N=18; 6.4%). Reduced EF was associated with a higher burden of ventricular premature beats (13.6% versus 3.8% with >100 ventricular premature beats/24 h; P=0.008) and lower left ventricular global longitudinal strain (-17%±2% versus -19%±2%; P<0.001). Athletes with reduced EF had a higher mean LVESVi-PRS (0.57±0.13 versus 0.51±0.14; P=0.009) with athletes in the top decile of LVESVi-PRS having an 11-fold increase in the likelihood of reduced EF compared with those in the bottom decile (P=0.034). Male sex and higher LVESVi-PRS were the only significant predictors of reduced EF in a multivariate analysis that included age and fitness. During follow-up, no athletes developed symptomatic heart failure or arrhythmias. Two athletes died, 1 from trauma and 1 from sudden cardiac death, the latter having a reduced right ventricular EF and a LVESVi-PRS >95%.

Conclusions: Reduced EF occurs in approximately 1 in 6 elite endurance athletes and is related to genetic predisposition in addition to exercise training. Genetic and imaging markers may help identify endurance athletes in whom scrutiny about long-term clinical outcomes may be appropriate.

Registration: URL: https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=374976&isReview=true; Unique identifier: ACTRN12618000716268.

Keywords: arrhythmias, cardiac; cardiomegaly; cardiomegaly, exercise-induced; cardiomyopathies; fibrosis; genetics; genome.

Figure 1.

Population distribution on the basis of ejection fraction and the relationship between LVEF and RVEF. A, Doughnut chart of the population distribution on the basis of ejection fraction. A reduced EF is defined as an LVEF <50% or an RVEF <45%. B, The relationship between LVEF and RVEF in endurance athletes. The dashed vertical and horizontal lines represent the cutoffs for a reduced LVEF (50%) and RVEF (45%), respectively. The dotted vertical and horizontal lines represent the cutoffs for a borderline LVEF (55%) and RVEF (50%). EF indicates ejection fraction; LV, left ventricular; and RV, right ventricular.

Figure 2.

The correlation between ejection fraction, end-diastolic volumes, and end-systolic volumes. Scatter plot and correlation analysis between (A) LVEF and LVEDVi, (B) LVEF and LVESVi, (C) RVEF and RVEDVi, and (D) RVEF and RVESVi. The vertical dotted lines represent the cutoff values of reduced LVEF at 50% and RVEF at 45%. The horizontal dotted lines represent the upper limit of the 99% CI of LVEDVi (122 mL/m²), LVESVi (55 mL/m²), RVEDVi (130 mL/m²), and RVESVi (63 mL/m²) as reported by D’Ascenzi et al. EF indicates ejection fraction; LV, left ventricular; LVEDVi, indexed left ventricular end-diastolic volume; LVESVi, indexed left ventricular end-systolic volume; RV, right ventricular; RVEDVi, indexed right ventricular end-diastolic volume; and RVESVi, indexed right ventricular end-systolic volume.

Figure 3.

Correlation between LVEF and LV GLS. For athletes with reduced EF (orange) and preserved EF (blue). EF indicates ejection fraction; GLS, global longitudinal strain; and LV, left ventricular.

Figure 4.

Changes in ejection fraction during exercise. Changes in LVEF (A) and RVEF (B) from rest to peak exercise in athletes with a reduced (orange dots) and preserved EF (blue triangles). Box-and-whiskers plots of delta LVEF (C) and delta RVEF (D) between rest and peak exercise in athletes with a reduced (orange) and preserved EF (blue). EF indicates ejection fraction; LV, left ventricular; and RV, right ventricular.

Figure 5.

Polygenic risk score for LVESVi: violin plots showing the distribution of LVESVi-PRS in athletes with a reduced or normal EF, in comparison with a healthy elderly population from the ASPREE study and patients with established DCM. Mean LVESVi-PRS in athletes with reduced EF is higher than in control subjects and athletes with normal EF, but similar to patients with DCM. Data for mean ± interquartile range in each group are shown. ASPREE indicates Aspirin in Reducing Events in the Elderly; DCM, dilated cardiomyopathy; EF, ejection fraction; LVESVi, indexed left ventricular end-systolic volume; and PRS, polygenic risk score.