OBJECTIVETo explore the predict value of plasma soluble ST2 (sST2) on one-year mortality for hospitalized patients with chronic heart failure (HF).

METHODSA total of 1 244 consecutive hospitalized patients admitted to Heart Failure Center Fuwai Hospital between March 2009 and July 2012 and with HF as their primary diagnosis were included. Plasma sST2 was measured in all patients and patients were followed up for 1 year, and the primary endpoint was defined as all-cause death.

RESULTSThere were 193 deaths during follow up. sST2 concentrations at admission were positively correlated with NT-proBNP, NYHA functional class and heart rate, and negatively correlated with left ventricular ejection fraction, blood sodium, total cholesterol and glomerular filtration rate at admission. sST2 concentrations were significantly higher in non-survivors compared with survivors (P < 0.001). Multivariable Cox regression analyses showed that sST2 independently predicted 1-year mortality (per 1 log unit, hazard ratio 1.87, 95% confidence interval: 1.56 to 2.25, P < 0.001). In receiver operating characteristic analyses, the area under the curve for ST2 was 0.776 which was similar to that of N-terminal pro-B-type natriuretic peptide (NT-proBNP) (AUC = 0.775). The prognostic value was improved when combining these two biomarkers together (AUC = 0.813).

CONCLUSIONSsST2 concentration at admission is correlated with clinical and biochemical indexes and associated with 1-year mortality for hospitalized patients with HF.