The comparison of clinical significance between atrial and ventricular structural remodeling in patients with heart failure with preserved ejection fraction – insights from PURSUIT-HFpEF registry

Abstract

Abstract Introduction Structural remodeling is an important aspect of pathophysiology of heart failure (HF). The compensatory mechanism against diastolic dysfunction would be more obvious on the left atrium rather than left ventricle in HF with preserved ejection fraction (HFpEF). Purpose To investigate the impact of left atrial (LA) and left ventricular (LV) remodeling on clinical outcomes in HFpEF patients. Methods Analysis was performed from a prospective multicenter observational registry for HFpEF (PURSUIT-HFpEF Registry) conducted in the Osaka region of Japan. A total of 757 patients hospitalized for acute HF (diagnosed by using Framingham criteria) met the inclusion criteria: a LV ejection fraction ≥50% and brain natriuretic peptide ≥100pg/ml. We included 381 patients (age, 82±9 years; men, 45%; atrial fibrillation (AF), 34%) whose follow-up data after survival discharge were available and whose LA volume index (LAVI) and left ventricular end-diastolic volume index (LVEDVI) at discharge were measured by the biplane Simpson's method using echocardiography. The primary endpoint was defined as the composite of death, re-hospitalization for HF, and cerebrovascular accident. Results The LAVI and LVEDVI at discharge was 54±25 ml/m2 and 55±21 ml/m2, respectively (r2=0.014, p=0.021). When patients were classified into the LAVI tertiles groups (T1: <40ml/m2 (n=124), T2: 41–59ml/m2 (n=127), T3: >60ml/m2 (n=130)), there was a positive correlation between the prevalence of diastolic dysfunction and the LAVI tertiles (T1, 21.0%; T2, 48.8%; T3, 51.5%; p<0.001). On the other hand, significant association was not observed between the prevalence of diastolic dysfunction and the LVEDVI tertile groups (p=0.42). During the median follow-up period of 396 days (IQR, 345–698), the composite endpoint was observed in 131 patients and there was a positive correlation between the endpoint and the LAVI tertiles (T1, 24.2%; T2, 38.6%; T3, 40.3%; p=0.011). On the other hand, no correlation was found between the endpoint and the LVEDVI tertiles (p=0.13). After adjustment for age, gender, and presence of diastolic dysfunction, Cox regression analysis revealed that not LVEDVI but LAVI at discharge was a significant predictor of the composite endpoint in the entire cohort (hazard ratio per 10 ml/m2 increase; 1.09; 95% confidence interval, 1.02–1.16; p=0.009) and in the sinus rhythm subgroup (1.10; 1.01–1.20; p=0.034). However, not LAVI but LVEDVI was a significant predictor for the adverse outcomes in the AF subgroup (1.23; 1.04–1.46; p=0.016). Conclusions Increased LAVI at discharge, which was associated with diastolic dysfunction, related to unfavorable prognosis in patients with HFpEF. However, LAVI was no more a predictor for the adverse outcomes but LVEDVI was in the AF subgroup. The clinical significance of atrial and ventricular remodeling may differ between sinus rhythm and AF rhythm in HFpEF population. Funding Acknowledgement Type of funding source: Private company. Main funding source(s): Roche Diagnostics K.K. and Fuji Film Toyama Chemical Co. Ltd.