P320Heart failure with mid-range (HFmrEF) or recovered (HFrecEF) ejection fraction: differential determinants of transition
Abstract Background Heart failure with mid-range ejection fraction (HFmrEF) has been identified as a multi-faceted phenotype that may encompass both patients with mild disease or those who from previous HFrEF recover EF (HFrecEF) Purpose To describe clinical characteristics and factors associated with phenotype transition at follow-up. Methods From 2009 to 2016, 1194 patients with baseline EF<50% and a second echocardiographic determination during clinically stability at a median of 6 months were enrolled in the IN-CHF Registry. Based on EF at enrollment, 335 (28%) had HFmrEF and 859 (72%) had HFrEF. We compared baseline clinical characteristics and predictors associated with follow-up reclassification to HFmrEF or full EF recovery Results When compared to HFrEF patients, those with HFmrEF had less often an ischemic etiology, advanced symptoms and a HF admission in the previous year. No other differences were found in clinical characteristics and drug therapy (Table). At a median follow-up of 6 months, 30% of HFrEF patients improved EF by 14 (9) units: 21% showed partial EF recovery (transition to HFmrEF) and 9% had full EF recovery. Conversely among HFmrEF patients 22% improved EF, by 9 (5) units, to full recovery, and 18% deteriorated by 1.5 (5.5) units sloping to HFrEF. By multivariable logistic regression analysis, variables associated with EF recovery at 6-month follow-up differed between baseline phenotypes. Within HFrEF, ischemic etiology (OR 0.46, 95% CI 0.33–0.64) and NYHA class III-IV symptoms (OR 0.57, 95% CI 0.38–0.68) were associated with a lower likelihood of EF recovery, while a history of HF<6 month correlated with a higher likelihood of EF recovery (OR 2.44, 95% CI 1.76–3.39). Within HFmrEF, while ischemic etiology (OR 0.66, 95% CI 0.19–0.68) was also associated with a lower likelihood of EF recovery, a history of atrial fibrillation at enrollment correlated with higher likelihood of EF recovery (OR 2.66, 95% CI 1.37–5.17) by 6 month-follow-up. At a median follow-up of 36+28 months mortality was 4.6% vs 6.9% in HFrecEF vs non-recovered patients (log rank p=0.08). Baseline characteristics HFrEF vs HFmrEF Conclusions HFmrEF patients showed a less severe clinical picture than HFrEF patients, but had EF recovery less often. EF improvement is negatively associated with ischemic etiology in both phenotypes, and positively associated with atrial fibrillation in HFmrEF and a short history of HF in HFrEF.