Neutrophil gelatinase-associated lipocalin (NGAL) is a rapidly emerging biomarker for early detection of acute kidney injury (AKI). The aim of the study is to evaluate the impact of sepsis on serum NGAL in critically ill children and to investigate whether the presence of sepsis affects the ability of serum NGAL to predict AKI. Sixty-eight patients, who met criteria of sepsis and related syndromes, were classified into two groups (septic shock and severe sepsis). They were reclassified regarding who developed AKI into AKI and non- AKI groups. Twenty sex and age-matched healthy subjects served as a control group. Serum NGAL was assayed using Enzyme-linked Immunosorbent Assay (ELISA), and serum creatinine was measured using the kinetic spectrophotometric method. Serum NGAL levels were significantly high in critically ill septic patients compared to healthy controls (median: 100.4ng/ml vs. 47.1 ng/ml, P>0.0001), and were significantly higher in septic shock (median 105.1ng/ml) than in severe sepsis (median 96ng/ml) with P values of 0.005. However, there was no significant difference in the levels of serum NGAL between AKI patients and non- AKI patients (P=0.3). Receiver operating characteristic (ROC) curve analysis of serum NGAL for prediction of AKI in critically ill septic children showed an area under the curve (AUC) was 0.56 (95% C.I. =0.42-0.70) with an optimal cutoff value of 102.5ng/ml, sensitivity=56.2%, specificity 55.6%, PPV =52.9%, NPV =58.8%, and accuracy=55.9%. For serum creatinine, the AUC was 0.97 (95% C.I. =0.94- 1.00) with an optimal cutoff value of 0.6 mg/dl, sensitivity= 90.6%, specificity 91.7%, PPV=90.6%, NPV=91.7% and accuracy=91.2%. In conclusion, serum NGAL is raised in critically ill septic children and is a marker of bacterial infection and systemic inflammation. However, in AKI associated with sepsis, serum NGAL is not a specific biomarker for the prediction of AKI and it loses its early predictive property. In such patients, serum creatinine is more specific than serum NGAL.