[The prognostic value of serum albumin for in-hospital mortality in patients with heart failure basing on propensity score matching]

Abstract

Objective: To analyze the prognostic value of serum albumin for in-hospital mortality in patients with heart failure.

Methods: A total of 2 430 consecutive heart failure patients aged at (68±14) years were enrolled in the study. Patients were divided into 2 groups according to serum albumin concentration on admission: the normoalbuminemia group (≥35 g/L) and the hypoalbuminemia group (<35 g/L). Propensity score matching was conducted to reduce confounding bias between the groups. Cox proportional-hazards regression models were used to evaluate the prognostic value of serum albumin for in-hospital mortality in patients with heart failure before and after matching.

Results: Compared with those in the normoalbuminemia group, subjects in the hypoalbuminemia group were older, and had higher NYHA functional status and higher in-hospital mortality. More patients were diagnosed with dilated cardiomyopathy and valvular heart disease, and fewer were with hypertension in the hypoalbuminemia group than those patients in the normoalbuminemia group. Moreover, patients in this group presented with faster heart rate and lower systolic blood pressure than those in the normoalbuminemia group. They had higher levels of direct bilirubin, alkaline phosphatase, glutamyltranspetidase, creatinine, uric acid, urea, and B-type natriuretic peptide (BNP) and lower levels of hemoglobin, total cholesterol (TC), and serum sodium compared with subjects in the normoalbuminemia group. Left ventricular ejection fractions (LVEF) of patients were lower in the hypoalbuminemia group than those of patients in the normoalbuminemia group. More patients were of long-duration and treated with spirolactone. With protensity score matching, 631 pairs of patients were successfully matched. Before matching, the in-hospital mortality in normoalbuminemia group was 1.2% and that in hypoalbuminemia group was 5.7%. The multivariate Cox regression analysis indicated that the risk for in-hospital death in patients with heart failure was 12.0% greater for each 1 g/L decrement in albumin level after adjusted for all clinical factors (HR 1.120, 95% CI 1.057-1.186; P<0.001). The same held after matching. The in-hospital mortality in normoalbuminemia group was 2.9%, and that in hypoalbuminemia group was 5.7%. The multivariate Cox regression analysis showed that the risk for in-hospital death in patients with heart failure was 11.0% greater for each 1 g/L decrement in albumin level after adjusted for all clinical factors (HR 1.110, 95% CI 1.043-1.181; P=0.001).

Conclusions: Serum albuminis is an independent risk factor for in-hospital mortality in patients with heart failure. Treatment of hypoalbuminemia may lower the in-hospital mortality in patients with heart failure.