Relations Between Cardiac Magnetic Resonance Imaging-Derived Left Ventricular Mass, Early Repolarization, and Cardiovascular Events (from the Dallas Heart Study)

Abstract

Early repolarization pattern (ERP) is associated with increased mortality in case-control studies, but the mechanism and role of left ventricular mass (LVM) remain unclear. Our objectives were to understand (1) whether ERP associates with adverse outcomes in a multiethnic population and (2) to explore the role of LVM in these associations. Participants from the Dallas Heart Study with an electrocardiogram interpretable for ERP, defined as Jpoint elevation ≥1mm in 2 contiguous leads, were included. Combined all-cause mortality and nonfatal cardiovascular disease (CVD) events and individual components were assessed using Cox proportional hazards modeling after adjustment for demographics, traditional CVD risk factors, electrocardiogram intervals, and cardiac magnetic resonance imaging-derived factors. Cardiac magnetic resonance imaging-defined LVM was then added to the most fully adjusted model. Of the 2,686 participants, 240 (8.9%) demonstrated ERP. Participants with ERP were more likely to be male and Black, with lower body mass index, greater left ventricular end-diastolic volumes, and LVM. Over a median follow-up of 11years, the combined end point occurred in 326 patients. Multivariable modeling demonstrated ERP was associated with the combined end point (HR [95% CI] 1.61 [1.14 to 2.26]), all-cause mortality (1.67 [1.00 to 2.80]). However, further adjusting for LVM attenuated the associations of ERP with the primary end point (HR [95% CI] 1.22 [0.85 to 1.77]) and secondary end points of mortality (1.39 [0.80 to 2.41]) and nonfatal CVD (1.05 [0.68 to 1.64]). ERP was associated with increased mortality and nonfatal CVD events, which was attenuated after adjusting for LVM, a previously under-recognized clinical phenotype. Previous associations of ERP with adverse cardiovascular outcomes may be partially explained by greater LVM in those with ERP.