Whole blood assessment of neutrophil gelatinase-associated lipocalin versus pediatricRIFLE for acute kidney injury diagnosis and prognosis after pediatric cardiac surgery: Cross-sectional study

Objective: To assess the ability of a single whole blood neutrophil gelatinase-associated lipocalin measurement in predicting acute kidney injury occurrence, its severity, and the need for postoperative renal replacement therapy after pediatric cardiac surgery. Design: Single-center prospective cross-sectional study. Setting: Tertiary care pediatric cardiac intensive care unit. Patients: Consecutive children <1 yr old with congenital heart diseases undergoing cardiac surgery with cardiopulmonary bypass. Interventions: None. Measurements and Main Results: Neutrophil gelatinase-associated lipocalin levels were measured after pediatric cardiac intensive care unit admission. Pediatric score indicating level of renal damage by Risk, Injury, Failure, Loss of function and End-stage kidney disease (pRIFLE) was used as the reference method. Acute kidney injury was diagnosed in 90 (56%) of the 160 enrolled patients. The number of abnormal neutrophil gelatinase-associated lipocalin samples (above the cutoff level of 150 ng/mL) was 12 over 90 (13%) in acute kidney injury population and 6 over 70 in non-acute kidney injury patients (8%) (odds ratio 1.6; 95% confidence interval 0.6-4.7; p = .31). Sensitivity of neutrophil gelatinase-associated lipocalin for acute kidney injury detection was 0.13 and specificity 0.91. The number of patients with abnormal neutrophil gelatinase-associated lipocalin samples was not significantly different within pediatric score indicating level of renal damage by pRIFLE (p = .69); furthermore, we found abnormal neutrophil gelatinase-associated lipocalin levels in 4 (30%) over 13 renal replacement therapy patients and in 14 (10%) over 133 children without renal replacement therapy need (odds ratio 4.2; 95% confidence interval 1.2-10.2; p = .02). Mean cross-clamp time (p = .28), inotropic score (p = .19), surgical risk score (p = .3), mean length of mechanical ventilation (p = .48), and pediatric cardiac intensive care unit stay (p = .57) did not significantly differ between children with abnormal and normal neutrophil gelatinase-associated lipocalin values. Conclusions: Neutrophil gelatinase-associated lipocalin measured at pediatric cardiac intensive care unit arrival does not accurately predict acute kidney injury diagnosis, according to pediatric score indicating level of renal damage by pRIFLE classification. In these patients, neutrophil gelatinase-associated lipocalin might be helpful for renal replacement therapy prediction. (Pediatr Crit Care Med 2012; 13:667-670)