Predictors of one year all-cause death in acute heart failure patients enrolled in the nationwide BLITZ-HF study

Abstract Background Outcomes of patients admitted for acute heart failure (AHF) are generally poor, then it is important to recognize outcome predictors in order to better identify and manage patients at higher risk. Aims To identify independent predictors of 1-year all-cause death in AHF patients enrolled by Italian cardiology sites participating to the BLITZ-HF study. Methods BLITZ-HF was a prospective nationwide study based on a web based recording system used during two enrollment periods (08/03/2017 – 04/09/2017 and 24/12/2017 – 09/04/2018). Overall, 7218 patients with acute and chronic HF were enrolled by 106 sites, 58 were lost to follow-up. Patients were followed for a median of 370 days [IQR 339–395]. Cox proportional hazards models were used to identify the independent predictors of all-cause death from hospital admission to 1-year follow-up. Demographic variables and baseline characteristics statistically significant at univariate analysis were included in the multivariable models. Different Cox proportional models were fitted: 1) adjusted for age and gender, 2) model 1 + clinical variables, 3) model 2 + laboratory and instrumental examinations, 4) model 3 + therapy. Results The present analysis refers to the 1470 patients admitted for AHF not lost to follow-up. Mean age was 73±12, with nearly 50% having more than 75 years of age. Female gender accounted for about one third of cases. More than half were de novo HF patients and 43% had an ischemic aetiology. Nearly 40% had a history of atrial fibrillation (AF) and chronic kidney disease (CKD) and one fifth had a history of COPD and peripheral obstructive artery disease. Furthermore, mean systolic blood pressure (SBP) on admission was 131±28 and, heart rate (HR) was 91±25. The majority of HF patients had reduced EF (HFrEF) (58%), followed by HF with preserved EF (HFpEF) (24%) and HF with mid range EF (HFmrEF) (18%). One year all-cause mortality was 16.6%. Independent predictors of all-cause mortality are shown in the Table. In the final model adjusted for age, gender, clinical variables, laboratory and instrumental examinations and medical treatments, the following variables resulted as independent predictors of one-year all-cause mortality: age (HR 1.02), Confuse/obnubilate mental status (HR 2.07), creatinine >1.5 mg/dl (HR 1.54), BUN >43 mg/dl, (HR 2.74), aortic stenosis (HR 1.70), inotrope use (HR 1.43), IV furosemide >125 mg (HR 1.74), SBP (HR 0.99). Conclusions In our study several independent predictors of one-year all-cause mortality have been identified, confirming and reinforcing previous findings. Interestingly, our independent predictors derived from different domains (demographic, clinical, biohumoral, instrumental and treatment variables). These findings further underline the importance of a comprehensive assessment in the prognostic evaluation of AHF patients. Funding Acknowledgement Type of funding sources: Private company. Main funding source(s): The study was funded by Heart Care Foundation with a partial unrestricted support from Abbott, Daiichi Sankyo, Medtronic, Servier, Vifor.