P1638Predictors of in-hospital and late mortality in acutely decompensated chronic heart failure

Abstract

Background Heart failure (HF) is a global pandemic, and the cause of the greater health expenditure on hospitalization. Purpose To determine, in patients admitted due to decompensated heart failure (dHF), the predictors of in-hospital and late all-cause mortality. Methods Retrospective and longitudinal observational study of consecutive patients admitted with a diagnosis of dHF in a General Hospital between 2000–2015. Primary objective was in-hospital and late mortality. Results A total of 3705 patients were included (76.9 + 10.1 years, 54.1% women). Diabetes was present in 1611 (43.5%) patients, hypertension in 3210 (86.6%), active smoking in 221 (6%), COPD in 989 (26.7%), neoplasms in 415 (11.2%), chronic renal failure (CRF) in 628 (17%), previous myocardial infarction (pMI) in 449 (12.1%), stroke in 416 (11.2%), previous heart failure (pHF) in 1015 (27.4%), atrial fibrillation in 1672 (45.1%). A total of 239 (6.5%) lived in a nursing home (NH). The rate of 6-months readmission due to heart failure was 473 (12.8%). The in-hospital all-cause mortality rate was 9.3% (345). During a median follow-up of 930 days (RI 230–2611), 2165 (58.4%) patients died. The following variables were independent predictors of in-hospital mortality (Odds ratio, 95% CI): age 1.03 (1.02–1.05), basal creatinine 1.18 (1.01–1.39) and urea 1.014 (1.011–1.018) (model AUC 0.78 (0.75–0.81)). We identified the following predictors of late mortality (Hazard ratio, 95% CI): age 1,029 (1,022–1,035), diabetes mellitus 1,13 (1,02–1,25), COPD 1,15 (1,03–1,28), neoplasms 1,17 (1,008–1,361), pMI 1.19 (1.02–1.38), pHF 1.25 (1.12–1.39), NH 1.24 (1.03–1.49), readmission 1.73 (1.51–1.97), urea 1.003 (1.001–1.005) (model AUC 0.73 (0.71–0.75)). Conclusions In-hospital mortality was significantly associated with age and markers of renal function, while for late mortality the predictors were comorbidities, hospital readmissions and living in a nursing home.