Objectives BNP and NT-proBNP are important cardiac biomarkers in the diagnosis and prognosis of congestive heart failure (CHF). However, the prognosis value of their ratio in patients with CHF is not clear. The goal of this study was to examine the prediction value of BNP, NT-proBNP and their ratio for in-hospital outcomes in CHF.

Methods In a cross-sectional study, patients with acute onset of CHF and admitted to cardiac care unit in Juntendo Hospital were enrolled from Jan to Dec 2009. We measured the serum level of BNP and NT-proBNP at the same time after admission, and other biomarkers were also measured and collected. The results were statistically analysed by software JMP 7.

Results A total of 193 patients were enrolled, with a mean age of 71.3±12.8 years old. 17 patients died in hospital, with a mortality rate of 8.8%. Univariate analysis showed that in-hospital mortality was significantly related with BMI, BNP, NT-proBNP, the ratio of NT-proBNP/BNP, RDW, LDL-C and CRP. The mean ratio of NT-proBNP/BNP was 16.7±11.6 for in-hospital death group and 9.5±8.6 for in-hospital relieving group (p<0.05). Logistic and multiple regression analysis showed that the ratio of NT-proBNP/BNP was an independent predictor for both in-hospital mortality and duration in hospital.

Conclusions The ratio of NT-proBNP/BNP is better for predicting in-hospital outcomes than BNP or NT-proBNP in congestive heart failure and it might be useful to predict short term outcomes in patients with acute exacerbation of heart failure.