Objective: The aim of this study was to evaluate the diagnostic and prognostic accuracy of brain natriuretic peptide assay in neonates and children undergoing surgery for congenital heart disease.
Design: Prospective, observational study.
Setting: Single center.
Patients: We enrolled 336 consecutive children (median age, 6 mo [range, 0-37 mo]) undergoing cardiac surgery (87 neonates; age, 7 d [5-12]; median, 25th-75th percentile; 24 infants and children; age, 11 mo [4-60]) and 436 healthy controls.
Interventions: Brain natriuretic peptide was measured preoperatively, on every postoperative day in the ICU, and at discharge. Intubation time was the primary outcome.
Measurements and main results: Preoperative brain natriuretic peptide values in patients with congenital heart disease were higher than those in controls (p < 0.01). Brain natriuretic peptide had a good diagnostic accuracy in discriminating between patients with congenital heart disease and healthy controls with an area under the curve = 0.918 for neonates and area under the curve = 0.894 for older children. The best cutoff values, calculated by receiver operating characteristic analysis, were different for the two age subgroups with cutoff values of 363.5 ng/L for neonates and 23.5 ng/L for older children. At 24 hours after surgery, although brain natriuretic peptide decreased in neonates (baseline 2723 vs 1290 ng/L, p < 0.001), it increased in children (60 vs 365 ng/L at 24 hours, p < 0.001). Multivariable analysis identified the preoperative level of brain natriuretic peptide in infant/children and the difference in brain natriuretic peptide value (baseline 24 hours) in neonates, as independent predictors of intubation time. Furthermore, body surface area, Aristotle score, and cardiopulmonary bypass time had an independent significant effect on the endpoint in either group.
Conclusions: Baseline cardiac endocrine function and its response to surgical stress are dependent on age in neonates and children, undergoing cardiac surgery for congenital heart disease. Brain natriuretic peptide shows a good diagnostic and prognostic accuracy in this setting, with different features in either neonates or infants/children subsets.