AIMS Left atrial (LA) dysfunction can be associated with left ventricular (LV) disorders; however, its clinical significance has not been well-studied in patients with acute heart failure (AHF). We evaluated prognostic power of peak atrial longitudinal strain (PALS) of the left atrium according to heart failure (HF) phenotypes and atrial fibrillation (AF).

METHODS AND RESULTS From an AHF registry with 4312 patients, we analysed PALS in 3818 patients. Patients were categorized into PALS tertiles. We also divided the patients according to HF phenotypes [HF with reduced ejection fraction (HFrEF), HF with mid-range ejection fraction (HFmrEF), or HF with preserved ejection fraction (HFpEF)] and presence of AF. The primary outcomes were all-cause mortality and HF hospitalization. PALS was weakly but significantly correlated with LA volume index (r = -0.310, P < 0.001), E/e' (r = -0.245, P < 0.001), and LV ejection fraction (r = 0.371, P < 0.001). A total of 2016 patients (52.8%) experienced adverse clinical events during median follow-up duration of 30.6 months (interquartile ranges 11.6-54.4 months). In the multivariate analysis, PALS was a significant predictor of events [hazard ratio (HR) 0.984, 95% confidence interval (CI) 0.971-0.996; P = 0.012]. Patients with the lowest tertile (HR 1.576, 95% CI 1.219-2.038; P < 0.001) had a higher number of events than those with the highest tertile in the multivariate analysis. In the subgroup analysis, however, PALS was not a prognosticator (HR 0.987, 95% CI 0.974-1.000; P = 0.056) in AF patients. The prognostic power of PALS was not different between HFrEF (HR 0.977, 95% CI 0.969-0.974; P < 0.001), HFmrEF (HR 0.984, 95% CI 0.972-0.996; P = 0.008), and HFpEF (HR 0.980, 95% CI 0.973-0.987; P < 0.001, P for interaction = 0.433).

CONCLUSION PALS was a significant prognostic marker in AHF patients. The prognostic power was similar regardless of HF phenotypes, but PALS was not associated with clinical events in AF patients.