It is known that levels of brain natriuretic peptide predict outcomes of treatment for adults with decompensated heart failure. We hypothesized that it could predict outcomes in children with this condition.

We divided retrospectively 82 patients with serial measurements of brain natriuretic peptide into 3 groups: those who survived and did not need readmission within less than 60 days; those who survived but needed readmission within less than 60 days; and those who died in hospital or within less than 60 days. Initial and final levels of the peptide correlated with adverse outcomes.

The percent change in level of the peptide was minus 78 percent, minus 38 percent, and 138 percent in the readmission-free group, the readmitted, and nonsurviving groups, respectively. Final levels were significantly lower in the readmission-free group than in the readmitted and nonsurviving groups (p equals 0.013 and p is less than 0.00001, respectively) and in the readmitted group than in the nonsurvivors (p equals 0.013). On univariate analysis, the final level, the change in level, and the percentage change in level significantly predicted outcomes (p equals 0.0002, 0.0072 and 0.0005, respectively). On multivariate analysis, only the final level of the peptide significantly predicted outcomes (p equals 0.01).

A final level of brain natriuretic peptide of greater than or equal to 760 picograms per millilitre strongly predicted an adverse outcome. Patients with higher final levels may be at higher risk of death and readmission, suggesting that this variable effectively predicts the response to treatment and prognosis in children with heart failure.