Worse long-term outcomes in new-onset HFpEF vs HFrEF and HFmrEF: findings from the Stockholm PREFERS study

BACKGROUND AND AIMS: Prospective outcome data in new-onset heart failure (HF) by ejection fraction (EF) category treated in HF clinics are limited. The Stockholm PREFERS study compared long-term outcomes in new-onset HF with preserved EF (HFpEF), mildly reduced EF (HFmrEF), and reduced EF (HFrEF). METHODS AND RESULTS: Between 2015 and 2019, 547 patients were enrolled: HFpEF n = 135 (25%), HFmrEF n = 61 (11%), and HFrEF n = 351 (64%). The mean age was 76, 71, and 67 years for HFpEF, HFmrEF, and HFrEF; the median baseline EF was 55%, 45%, and 30% (all P < .001 across groups). The primary outcome was time to cardiovascular (CV) mortality or first HF hospitalization (HFH). Secondary outcomes were all-cause mortality, CV mortality, and HFH. NT-proBNP and EF were reassessed at 1 year. The median follow-up was 3.8 years (IQR 3.0-4.7). The risk for the primary outcome was higher in HFpEF than in HFmrEF/HFrEF (unadjusted HR 2.2; 95% CI 1.5-3.1; P < .001; adjusted HR 1.7; 95% CI 1.1-2.9; P < .05). Overall event rates were: all-cause mortality 12.8%, CV mortality 9.1%, HFH 12.3%. At 1 year, NT-proBNP decreased in HFmrEF by 42% (P < .05) and in HFrEF by 55% (P < .001), with EF increases of 2% points (pp) (P < .05) and 16 pp (P < .001), respectively. In HFpEF, EF decreased by 5 pp (P < .001) with no change in NT-proBNP. CONCLUSIONS: In new-onset HF managed in university hospital-based HF clinics, HFpEF had worse long-term outcomes than HFmrEF and HFrEF combined. The findings highlight the severity of HFpEF and the need for more effective treatment strategies. CLINICAL TRIAL: NCT03671122.