Abstract
BACKGROUND: The prognostic importance of left atrial (LA) dysfunction is increasingly recognized. Magnetic resonance imaging can provide excellent visualization of the LA wall. We aimed to study the association of LA dysfunction measured using feature-tracking magnetic resonance imaging with incident adverse cardiovascular events among subjects with or without heart failure (HF) at baseline. METHODS AND RESULTS: We prospectively studied 640 adults without HF (n=419), HF with preserved ejection fraction (n=101), or HF with reduced ejection fraction (n=120). We measured phasic LA function by volumetric and feature-tracking methods to derive longitudinal strain. The composite outcome of incident HF hospitalization or death was adjudicated during a median follow-up of 37.1 months. Measures of LA phasic function were more prominently impaired in subjects with HF with reduced ejection fraction than among subjects with HF with preserved ejection fraction. In unadjusted Cox proportional hazards models, all measures of phasic LA function and volumes (maximum, minimum, and diastatic) were associated with the composite outcome. However, in analyses that adjusted for clinical risk factors, HF status, maximum LA volume, left ventricular mass, and left ventricular ejection fraction, measures of conduit and reservoir LA function, but not booster-pump function, were associated with the composite outcome. The strongest associations were observed for conduit longitudinal strain (standardized hazard ratio, 0.66; 95% CI, 0.49-0.88; P=0.004), conduit strain rate (standardized hazard ratio, 1.59; 95% CI, 1.16-2.16; P=0.0035), and reservoir strain (standardized hazard ratio, 0.68; 95% CI, 0.52-0.89; P=0.0055). CONCLUSIONS: Phasic LA function measured using magnetic resonance imaging feature tracking is independently predictive of the risk of incident HF admission or death, even after adjusting for LA volume and left ventricular remodeling.