Electrocardiographic left ventricular hypertrophy and sudden cardiac death

Electrocardiographic (ECG) left ventricular hypertrophy (LVH) has been associated with sudden cardiac death (SCD), but comparative data on different ECG LVH criteria in high-risk populations are limited. We compared the association between multiple ECG LVH criteria and SCD events in two coronary angiography cohorts spanning acute and chronic coronary syndromes, including patients without angiographic evidence of coronary artery disease.

This retrospective study included 19,220 consecutive patients from two cohorts. ECG LVH criteria included Sokolow-Lyon, Cornell voltage, Cornell product, Peguero-Lo Presti, MESA, RaVL, and Romhilt-Estes criteria. Associations with SCD events were assessed using Fine-Gray subdistribution hazard models accounting for the competing risk of non-SCD death. For the most robust criteria, additional time-varying Cox proportional hazards models were conducted to evaluate longitudinal associations based on repeated ECG measurements (>280,000 ECGs). Additionally, the 3-year discriminative ability of each criterion for SCD event was assessed.

Across the combined cohorts, the median follow-up was 8.4 years, with a total of 858 SCD events. In multivariable Fine-Gray analyses, Cornell product showed the strongest association with SCD events (SHR 1.29, 95% CI 1.09-1.52, p 0.003). This association remained consistent in time-varying Cox models and after excluding patients with left ventricular ejection fraction ≤35%. Discriminative performance was modest (AUC 0.67, 95% CI 0.64-0.70).

After adjustment for established SCD risk factors, Cornell product showed the strongest association with SCD events, with consistent finding whether ECG LVH criterion was present at baseline or developed during the follow-up. However, both the strength of association and discriminative performance were modest.