Background: The aim of this study is to test the hypothesis whether the combined use of a cardio-specific biomarker, the brain natriuretic peptide (BNP) and a marker of early renal damage, the assay of urinary neutrophil gelatinase-associated lipocalin (uNGAL), may improve risk stratification in pediatric cardiac surgery.
Methods: We prospectively enrolled 135 children [median age 7 (interquartile range 1-49) months] undergoing to cardiac surgery for congenital heart disease. All biomarkers were evaluated pre- and post-operatively at different times after cardiopulmonary-bypass (CPB): uNGAL at 2, 6 and 12 h; BNP at 12 and 36 h; serum creatinine at 2, 6, 12, and 36 h. Primary endpoints were development of acute kidney injury (AKI) (defined as 1.5 serum creatinine increase) and intubation time.
Results: AKI occurred in 39% of patients (65% neonates and 32% older children, p=0.004). The peak of uNGAL values occurred more frequently at 2 h. uNGAL values at 2 h [median 28.2 (interquartile range 7.0-124.6) ng/L] had a good diagnostic accuracy for early diagnosis of AKI with an AUC (area under the curve) ROC (receiver operating characteristic) curve of 0.85 (SE 0.034). Using multivariable logistic regression analysis, development of AKI was significantly associated with uNGAL values at 2 h after CPB [OR=1.88 (1.30-2.72, p=0.001)], together with the CPB time and Aristotle score, as an index of complexity of the surgical procedure, while pre-operative BNP values were not. Furthermore, uNGAL and pre-operative BNP values (together with Aristotle score) were significantly associated with adverse outcome (longer intubation time and mortality).
Conclusions: Pre-operative BNP and uNGAL values after surgery (together with the Aristotle score) were independently associated with a more severe course and worse outcome in children undergoing cardiac surgery for congenital heart disease.