Combining the use of amino-terminal pro-B-type natriuretic peptide and B-type natriuretic peptide in the prognosis of hospitalized heart failure patients

Objective: To investigate whether the combination of measuring amino-terminal pro-B-type natriuretic peptide (NT-proBNP) and B-type natriuretic peptide (BNP) could provide additional prognostic value in hospitalized heart failure patients.

Methods: We measured both BNP and NT-proBNP simultaneously at baseline in 1464 hospitalized heart failure patients who were admitted to our heart failure center. All patients were followed-up with the median follow-up period of 533 days. The primary endpoint is a composite of all-cause death (non-transplantation patients) or heart transplantation.

Results: The median molar ratio of NT-proBNP/BNP was 2.37, but the range of the molar ratio varied from 1.57 to 3.75 (lower quartile to higher quartile). Using the cut-off value of 1790 pg/mL for NT-proBNP and 495 pg/mL for BNP from the ROC curve analysis, univariate Cox proportional regression analysis showed that the low/high group (NT-proBNP < 1790 pg/mL and BNP ≥ 495 pg/mL), high/low group (NT-proBNP ≥ 1790 pg/mL and BNP < 495 pg/mL) and high/high group (NT-proBNP ≥ 1790 pg/mL and BNP ≥ 495 pg/mL) had significant higher risk of all-cause death or heart transplantation [HR (hazard ratio): 2.87, 95% CI (confidence interval): 1.69-4.89, p < .001; HR: 2.68, 95% CI: 1.91-3.76, p < .001; HR:5.07, 95% CI: 3.85-6.67, p < .001] than low/low group (NT-proBNP < 1790 pg/mL and BNP < 495 pg/mL). In turn, the high/high group had higher risk of all-cause death or heart transplantation than low/high (HR: 1.70, 95% CI: 1.04-2.80, p = .035) and high/low groups (HR: 1.88, 95% CI: 1.42-2.49, p < .001). The low/high and high/low groups had a similar risk of all-cause death or heart transplantation. Further multivariable Cox regression analysis also showed that both BNP and NT-proBNP above the cut-off values independently predicted the worst prognosis, while either one of the two biomarkers above the cut-off value indicated the moderate poor prognosis and both below the cut-off values indicated the best prognosis (p for trend< 0.001).

Conclusion: The plasma levels of NT-proBNP and BNP do not always increase proportionally in heart failure patients. The combination of testing NT-proBNP and BNP may add prognostic value to predict adverse events in hospitalized heart failure patients.