Predictors of Hospitalization for Heart Failure Decompensation in 18-months Follow-up After Index Hospitalization for Acute Heart Failure.

Introduction: Heart failure (HF) has very high rate of repeat hospitalizations due to HF decompensation (HHFD), sometimes very shortly after discharge for acute HF.

Aim: The aim of this paper is to investigate rate of HHFD and to identify their possible predictors.

Patients And Methods: Total amount of 222 patients hospitalized at Clinic for heart and vessel disease and rheumatism in acute HF were followed for next 18 months for occurrence of HHFD. During hospitalization were collected demographic data, risk factors, routine laboratory tests and admission BNP (brain natriuretic peptide), discharge BNP, percentage change of BNP during hospitalization, high sensitive troponin I, CA125 (cancer antigen125) and cystatin C.

Results: In next 18 months 129 patients (58.11%) reached end-point HHFD- mean time of its occurrence was 2.2 (95% CI=1.67-2.7) months. Patients with HHFD had more often arterial hypertension (HTA) (p=0.006), had higher BMI (p=0.035) and had higher values of bilirubin, admission BNP (p=0.031), discharge BNP (p <0.001), CA125 (p=0.023) and cystatin C (p=0.028). There was no difference in troponin values (p=0.095), while % reduction of BNP during hospitalization was lower (p<0.001) in group with HHFD. In univariate Cox hazard analysis HTA was positively and BMI negatively correlated with HHFD, while in multivariate Cox hazard analysis independent predictors were HTA (HR 1.6; 95% CI=1.1-2.2; p=0.018) and BMI<25 (HR 1.6; 95% CI=1.1-2.3; p=0.007). In univariate Cox hazard analysis admission BNP, discharge BNP, rise of BNP during hospitalization, CA125 and bilirubin were positively correlated, while sodium was negatively correlated with HHFD. In multivariate Cox hazard analysis there was only one independent predictor of HHFD - discharge BNP (HR 6.05; 95% CI=1.89-19.4; p=0.002). Conclusion: Arterial hypertension, BMI>25 and discharge BNP are independent predictors of HHFD. This could help us to identify high-risk patients for readmission who should be monitored more frequently and treated with sophisticated drug and device therapy.