Sex-specific performance of electrocardiographic criteria for left ventricular hypertrophy in elite athletes

Background: In athletes, left ventricular hypertrophy (LVH) criteria based on electrocardiograms (ECGs) have been validated almost exclusively in men using echocardiography. Sex-specific cardiac magnetic resonance (CMR) validation is lacking. Objective: To evaluate ECG-LVH criteria against contrast-enhanced CMR in male and female elite athletes. Methods: Cross-sectional study in healthy elite athletes. Eight ECG-LVH criteria of voltages and products and QRS duration were quantified using automated ECG analysis, and compared to CMR-derived LVH indicators (indexed left ventricular mass (LVM) and maximum left ventricular wall thickness (maxLVWT). Primary metrics of interest were sex-specific correlations (ρ) between ECG-LVH criteria and LVM/maxLVWT. Secondary metrics included discriminative performance (AUROC) and sensitivity at a 95%-specificity level for detecting increased LVM or maxLVWT. Results: Among 209 elite athletes (median age 25, 45% women), men more frequently met one or more voltage criteria than women (64% vs 45%, P = .010). In men, no ECG-LVH criteria showed meaningful correlations or discriminative performance for LVM or maxLVWT. Only QRS duration demonstrated discriminative performance (AUROC: LVM 0.67; maxLVWT 0.74). In women, all voltage and product criteria correlated with LVM (ρ = 0.25–0.45) with acceptable discrimination (AUROC: 0.63–0.73). Peguero-Lo Presti, Cornell-, and Modified Sum of 12-Lead Product showed moderate performance (sensitivities: 24%–29%) at 95% specificity in female athletes. Overall, increased precordial QRS voltages were independently correlated with lower extracellular volume (β = −0.3 mV per %ECV, P < .001). Conclusion: In elite athletes, ECG-LVH criteria, except for QRS duration, lack diagnostic value for LVH in men. Voltage criteria may have diagnostic potential in women.