Clinical relevance of time course of BNP levels in neonates with congenital heart diseases

Introduction: The aim of this study is to better characterize the time courses of BNP levels throughout the first days of life in larger populations of neonate and infant with or without congenital heart diseases (CHD) in order to increase the diagnostic accuracy of BNP assay in pediatric patients with CHD. Materials and methods: BNP was measured by an automated platform (Triage BNP reagents, ACCESS Immuno-assay Systems, Beckman Coulter, Inc., Fullerton, CA 92835) in 218 neonates and infants with different CHD; 222 healthy children, matched for age, served as controls. Results: BNP values were significantly higher (P < 0.001) in the whole group of CHD patients (median 1029.8 ng/L, range 25-20,152 ng/L) than in controls (median 149.5 ng/L, range 9-866 ng/L). A different trend between BNP values and age was observed in healthy subjects and CHD patients. After an initial increase within the first 4 days of life, BNP values in CHD patients tend to stabilize to high values in the following days. On the contrary, in control subjects a peak of BNP levels was observed in the second or third day, followed by a progressive decrease. Therefore, the diagnostic accuracy of BNP assay, calculated in the samples collected in the first four days of life (AUC of ROC analysis 0.86, 95% CI 0.83-0.90) was significantly lower (P<0.0001) compared to samples collected from 5 days to 30 days of life (AUC 0.97, 95% CI 0.95-0.99). Optimal cut-off values for BNP assay, as calculated by ROC analysis, were also age-dependent (cutoff for the first 4 days of life: 363.5 ng/L; cutoff values from 5 to 30 days of life: 109.5 ng/L). Conclusions: Our study demonstrates that differences in time-courses of BNP values between newborns with and without CHD throughout the first days of life clearly affect the diagnostic accuracy of BNP assay. Indeed, the diagnostic accuracy of BNP assay in discriminating between healthy newborns and CHD patients progressively increases after the 4th day of life. As a result, also cutoff values of BNP assay greatly change throughout the first days of life. However, decision values of BNP assay are strongly method-dependent, consequently clinicians should give great care to compare results obtained by different laboratories, especially when different methods are used. (C) 2011 Elsevier B.V. All rights reserved.