AIMS: In contemporary heart failure (HF) practice, prognostic value for combinations of cardiac and non-cardiac predictors remains poorly understood. We analysed the combinatorial predictors of outcomes in acute HF patients. METHODS AND RESULTS: This longitudinal cohort study included consecutive patients admitted for acute decompensated HF between April 2015 and March 2018 in an urban hospital. The main outcomes are HF readmission within 6 months after discharge or all-cause death. A total of 451 patients with 662 admissions were enrolled and the data including frailty and echocardiographic parameters were analysed by multivariate and matched cohort analyses. The mean age of the patients was 76.8 years. We constructed a multi-frailty index (MFI) ranging from 0 to 3 points as a composite of non-cardiac comorbidities and biopsychosocial frailty. In matched cohort of patients with ejection fraction ≧50% (HFpEF), MFI ≧1, pulmonary hypertension (PH; peak flow velocity of tricuspid regurgitation ≧2.9 m/s by echocardiography), and pancytopenia at discharge were strong predictors of HF readmission [odds ratios (ORs), 4.33, 2.5, and 2.86; P = 0.02, 0.05, and 0.02, respectively], and MFI ≧2 was the only predictor for all-cause death. For ejection fraction <40%, age, BNP ≧800 pg/mL, increase in estimated glomerular filtration rate during hospitalization, and lymphocytopenia plus anaemia predicted HF readmission (ORs, 1.77, 2.72, 0.73, and 2.89; P = 0.001, 0.05, 0.04, and 0.03, respectively). In contrast, diabetes mellitus was the only specific predictor found in patients over 80 years old. CONCLUSION: These data identified multi-frailty and PH or mild pancytopenia as synergistic predictors of HF readmission in HFpEF patients. Published on behalf of the European Society of Cardiology. All rights reserved.
AIMS: In contemporary heart failure (HF) practice, prognostic value for combinations of cardiac and non-cardiac predictors remains poorly understood. We analysed the combinatorial predictors of outcomes in acute HF patients. METHODS AND RESULTS: This longitudinal cohort study included consecutive patients admitted for acute decompensated HF between April 2015 and March 2018 in an urban hospital. The main outcomes are HF readmission within 6 months after discharge or all-cause death. A total of 451 patients with 662 admissions were enrolled and the data including frailty and echocardiographic parameters were analysed by multivariate and matched cohort analyses. The mean age of the patients was 76.8 years. We constructed a multi-frailty index (MFI) ranging from 0 to 3 points as a composite of non-cardiac comorbidities and biopsychosocial frailty. In matched cohort of patients with ejection fraction ≧50% (HFpEF), MFI ≧1, pulmonary hypertension (PH; peak flow velocity of tricuspid regurgitation ≧2.9 m/s by echocardiography), and pancytopenia at discharge were strong predictors of HF readmission [odds ratios (ORs), 4.33, 2.5, and 2.86; P = 0.02, 0.05, and 0.02, respectively], and MFI ≧2 was the only predictor for all-cause death. For ejection fraction <40%, age, BNP ≧800 pg/mL, increase in estimated glomerular filtration rate during hospitalization, and lymphocytopenia plus anaemia predicted HF readmission (ORs, 1.77, 2.72, 0.73, and 2.89; P = 0.001, 0.05, 0.04, and 0.03, respectively). In contrast, diabetes mellitus was the only specific predictor found in patients over 80 years old. CONCLUSION: These data identified multi-frailty and PH or mild pancytopenia as synergistic predictors of HF readmission in HFpEF patients. Published on behalf of the European Society of Cardiology. All rights reserved.