Objective: Left ventricular hypertrophy (LVH) is a strong risk factor for cardiovascular events. ECG is the most widely used method for LVH detection. Despite the abundance of ECG LVH criteria, their prognostic values have been compared in only a few studies, and little has been known about how sex modifies the prognostic value of LVH. We assessed the relationship between ECG LVH and incident cardiovascular events in the general population.
Methods: Several ECG LVH criteria were measured in 3059 women and 2456 men participating in the Health 2000 Study - a national general population survey. Association between ECG LVH and cardiovascular events were analyzed with Cox proportional-hazards models.
Results: ECG LVH was more prevalent in women than in men when measured with Cornell-based criteria, but less prevalent or nondifferent when measured with other criteria. The association between ECG LVH and events showed higher hazard ratios for women than in men. Sex × LVH interaction terms were statistically significant in part of the LVH criteria. In adjusted Cox models, Sokolow-Lyon voltage performed the best. The composite of Sokolow-Lyon voltage and Cornell voltage was statistically significantly associated with events in both sexes.
Conclusion: Sex affects both the prevalence rates and prognostic values of ECG LVH criteria in the general population, while showing higher prognostic value of ECG LVH in women than in men. For clinical use, the composite of the Sokolow-Lyon voltage and the Cornell voltage seems to be a good option.