Association of Abnormal Left Ventricular Functional Reserve With Outcome in Heart Failure With Preserved Ejection Fraction

OBJECTIVES This study sought to determine the prognostic value of abnormal diastolic and systolic responses to exercise (on the basis of exertional E/e' and global longitudinal strain rate [GSR]) in a well-characterized population of patients with heart failure with preserved ejection fraction (HFpEF). BACKGROUND Impaired cardiovascular functional reserve is believed to contribute to adverse outcomes in HFpEF. However, the exact characteristics of pathophysiological profiles associated with increased clinical risk are still poorly defined. METHODS A complete echocardiogram (including assessment of myocardial deformation) was performed at rest in 205 patients (64 +/- 8 years of age) with symptomatic HFpEF. Echocardiography following maximal exercise was undertaken to assess abnormal diastolic reserve (AbnDR) (exertional E/e' > 14) and exercise GSR. Patients were followed over 26 +/- 5 months for death and cardiovascular or heart failure (HF) hospitalization. RESULTS Cardiovascular hospitalization or death occurred in 64 patients (31%), including 51 (25%) with HF hospitalization. The composite endpoint was associated with AbnDR (hazard ratio: 2.69; 95% confidence interval: 1.44 to 5.04; p = 0.002) and reduced exercise GSR (hazard ratio: 0.14; 95% confidence interval: 0.04 to 0.49; p = 0.002). Both exercise parameters showed prognostic value, independent from and incremental to clinical data and B-type natriuretic peptide. The ability of E/e' and GSR measurements to predict outcomes on exertion exceeded their prognostic value at rest, and the presence of reduced exertional GSR in patients with AbnDR was associated with worse prognosis (p = 0.03 for the composite endpoint and p = 0.01 for HF hospitalization). CONCLUSIONS Both left ventricular systolic and diastolic reserves contribute to risk prediction in HFpEF. The inclusion of the exertional assessment of left ventricular function to diagnostic algorithms may improve the prognostication process in this disease condition. (C) 2018 by the American College of Cardiology Foundation.