Increased qrs voltages could be a promising factor to differentiate between CMR-matched elite athletes and dilated cardiomyopathy gene carriers

Abstract Background Differentiating exercise-induced cardiac remodelling (EICR) from mild dilated cardiomyopathy (DCM) is challenging when indexed left ventricular end-diastolic volumes (LVEDVi) and ejection fractions (LVEF) fall within the ‘grey zone’. When cardiac magnetic resonance imaging (CMR) is insufficient for differentiation, a multimodality approach combining ECG findings with CMR parameters could help. However, the discriminative value of ECG parameters in grey zone athletes and DCM patients remains underexplored. Purpose To determine whether ECG findings can improve differentiation between grey zone elite athletes and mild phenotype DCM-gene carriers (DCM-GC) with similar LVEDVi and LVEF. Methods We conducted a cross-sectional study of 1:1 CMR-matched elite athletes (ELITE cohort, (16 years old; 10 training hours; competing at national, international, and Olympic levels; no para-athletes)) and DCM-GC (from the Cardiogenetic Biobank). Matching was based on sex, increased LVEDVi (108 ml/m2 in males; 96 ml/m2 in females) and/or slightly reduced LVEF (40-50% in males; 40-51% in females) (1), using the smallest propensity score distance. ECGs were analysed according to the 2018 international recommendations for ECG interpretation in athletes (2). We compared abnormal and EICR (sinus bradycardia/1st-degree AV block/Mobitz Type I AV block/early repolarisation/incomplete right bundle branch block) ECG findings between groups. A two-tier decision tree (abnormal vs. normal or borderline ECG; presence of EICR ECG findings) was employed to assess sensitivity, specificity, and area under the curve (AUC). Logistic regression analysed associations between QRS voltage and DCM, adjusted for age, sex, and body mass index, and assessed sensitivity, specificity, and AUC. The cutoff value corresponding to the optimal predicted probability was analysed. QRS voltages were calculated with the Modular ECG Analysis System and not corrected for body surface area. Results After matching, 33 elite athletes and 33 DCM-GC were included (42.4% women), with similar LVEDVi (111.75 ml/m2 ± 16.37 vs. 110.08 ml ± 24.00, p=.741) and LVEF (52.10% ± 4.81 vs. 52.07% ± 5.74, p=.979). Elite athletes showed more EICR-related ECG findings (78.8% vs. 45.5%, p=.011). Pathological ECG findings showed a trend toward DCM-GC being more common (6.1% vs. 18.2%, p=.258). The two-tier decision tree had an AUC of 0.621 (95% CI: 0.507–0.735), sensitivity of 48%, and specificity of 76%. Elite athletes had higher QRS voltages (20.29 ± 5.96 mV vs. 14.53 ± 5.14 mV, p=.005). Logistic regression revealed that QRS voltage was independently associated with group status (p=.008), with an AUC of 0.842 (95% CI: 0.724-0.960), sensitivity of 62%, and specificity of 95%. The cutoff value was 13.52 mV. Conclusions Lower QRS voltages, in the diagnostic ‘grey zone’ between EICR and mild DCM, may indicate pathological cardiac remodelling, as DCM-GC generally showed lower QRS voltages.ROC curve for the Two-Tier Decision TreeFor image description, please refer to the figure legend and surrounding text. ROC curve for the Logistic RegressionFor image description, please refer to the figure legend and surrounding text.