Background: This study tested various neurohormones for prediction of heart failure death (death owing to progressive deterioration of ventricular function; HFD). Moreover, B-type natriuretic peptide (BNP) as a predictor of sudden death (SD; as reported previously) and the best predictor of HFD were combined for a simple risk stratification model.
Design: BNP, the N-terminal fragment of BNP (N-BNP), and of the atrial natriuretic peptide (N-ANP) and big endothelin levels were obtained from 452 patients with a left ventricular ejection fraction </= 35%. Outcome was documented during an observation period up to 3 years.
Results: Two hundred and ninety-eight patients survived without heart transplantation, 65 patients underwent heart transplantation and 89 patients died (SD 44 patients, HFD 31 patients, other causes 14 patients). The only independent predictor of SD was BNP (as reported), and the best independent predictor of HFD was N-ANP (P = 0.0001). Analyzing 293 survivors and 44 patients with SD, fewer patients with BNP < 130 pg mL(-1) (Group A, n = 110) died [1%] as compared with patients with BNP > 130 pg mL(-1) and N-ANP < 6300 fmol mL(-1) (Group B, n = 177; 18%; P = 0.0001) and patients with BNP > 130 pg mL(-1) and N-ANP > 6300 fmol mL(-1) (Group C, n = 50; 19%; P = 0.0001). Analyzing 293 survivors and 31 patients with HFD, fewer patients died in Group A (n = 109; 0%; P = 0.0001) and Group B (n = 153; 6%; P = 0.0001) as compared with patients of Group C (n = 62; 34%).
Conclusion: Prognostic power of neurohormones depends on the mode of death. The combined determination of BNP and N-ANP identifies patients with minimal risk of death, elevated SD but low HFD risk as well as elevated SD and HFD risk.