BACKGROUND: The H2FPEF score is recognized as a simple method to diagnose heart failure (HF) with preserved left ventricular ejection fraction (HFpEF). We investigated the value of the H2FPEF score in predicting subsequent cardiovascular events in HFpEF patients. METHODS: This study was a retrospective, single-center, observational study. We calculated the H2FPEF scores for 404 consecutive HFpEF patients. Subjects were subdivided into low- (0-3), intermediate- (4-6), and high-score (7-9) groups and followed for 50 months. The primary and secondary endpoints were composite cardiovascular/cerebrovascular events (cardiovascular death, nonfatal myocardial infarction, unstable angina pectoris, hospitalization for HF decompensation, and nonfatal stroke) occurrence and HF-related events (hospitalization for HF decompensation) occurrence at 50 months, respectively. RESULTS: Kaplan-Meier analyses demonstrated a significantly higher incidence of cardiovascular/cerebrovascular events among those with a higher H2FPEF score (log-rank test, P = 0.005). The HF-related event rate was higher in proportion to the H2FPEF score (log-rank test, P < 0.001). Multivariate Cox hazard analyses identified the H2FPEF score (per 1 point) as an independent predictor of cardiovascular and HF-related events (hazard ratio [HR], 1.179; 95% confidence interval [CI], 1.066-1.305; P = 0.001 and HR, 1.288; 95% CI, 1.134-1.463; P = 0.001, respectively). Receiver operating characteristic analysis showed that the H2FPEF significantly predicted cardiovascular events (area under the curve [AUC], 0.626; 95% CI, 0.557-0.693; P < 0.001) and HF-related events (AUC, 0.680; 95% CI, 0.600-0.759; P < 0.001). The cutoff H2FPEF score was 5.5 for the identification of cardiovascular and HF-related events. CONCLUSION: The H2FPEF score might be a potentially useful marker for the prediction of cardiovascular and HF-related events in HFpEF patients. CLINICAL TRAILS REGISTRATION: Trail Number UMIN000029600.