RESUMO
BACKGROUND: To assess the ability of urinary neutrophil gelatinase-associated lipocalin (UNGAL) to discriminate between culture-positive vs. culture-negative late-onset sepsis evaluations. METHODS: This is a prospective observational study of 136 neonates who underwent ≥1 sepsis evaluation at >72 h of age. Urine was obtained at the time of sepsis evaluation to measure UNGAL concentration. Using generalized estimating equations controlling for gender, gestational and postnatal age, acute kidney injury, and within-patient correlations, pair-wise contrasts between mean log UNGAL concentrations of infants with negative sepsis evaluations vs. culture-positive sepsis and presumed sepsis were assessed. Discrimination characteristics at several UNGAL cutoff concentrations were assessed using receiver-operating characteristic curves. RESULTS: The predicted mean log UNGAL values of culture-positive sepsis and presumed sepsis vs. negative sepsis evaluations differed significantly (P < 0.001 and P = 0.02, respectively). At a cutoff ≥ 50 ng/ml, UNGAL discriminated between culture-positive sepsis and culture-negative sepsis evaluations with sensitivity = 86%, specificity = 56%, positive predictive value = 41%, negative predictive value = 92%, and number needed to treat = 3. CONCLUSION: UNGAL is a noninvasive biomarker with high negative predictive value at the time of late-onset sepsis evaluation in neonates and could be a useful adjunct to traditional components of sepsis evaluations.