Background. The exact association between urinary neutrophil gelatinase-associated lipocalin (uNGAL) and acute kidney injury (AKI) is unknown in a critical care setting, in which the population is heterogeneous and the aetiology of AKI is unclear. Aim of this study is to clarify if uNGAL level is an early diagnostic marker for AKI in patients with sepsis. Materials and methods. The current study was conducted on 86 sepsis patients. The prevalence of AKI was identified among them. The role of uNGAL in predicting AKI development, mortality rate and length of the intensive care unit (ICU) stay were analyzed. Sensitivity and specificity were calculated, and the area under the receiver operating characteristic curve was considered as the optimal uNGAL cut-off level for detecting all classifications of AKI. Results. Most patients belonged to the age group of 51–60 years and their mean age was 54.6 years. Most patients (65.11 %) were males. 26.75 % had both type 2 diabetes mellitus and hypertension. AKI was detected in 89 % of subjects in the current study, as per KDIGO definition. 15.12 % of patients had stage 1 CKD, 15.12 % had stage 2 CKD, and stage 3 CKD was diagnosed in 4.65 % of cases. Mortality rate was 11 %, and 89 % of patients were discharged. The mean ICU length of stay among patients with AKI is 8.9 days. There is significant association between the mean ICU length of stay and AKI presence (p = 0.03). 17.4 % (n = 15) of patients required renal replacement therapy. There is a very significant difference in mean baseline uNGAL in patients with and without AKI: 149.9 and 73.2 ng/ml, respectively (p = 0.0006). This indicated that baseline uNGAL levels predict AKI. The mean uNGAL in people with AKI was 356 ng/ml and in those without AKI, it was 95 ng/ml. There is a very significant difference in mean uNGAL 48 hours after in patients with and without AKI (p < 0.0001). At a cut-off value of 120, there were 69 true positive cases, 9 true negative cases, 0 false positive cases, and 8 false negative cases. Based on these, the sensitivity of uNGAL at baseline in detecting AKI is 89.61 %, specificity is 100 %, and accuracy is 90.70 %. At a cut-off point of 120, there were 77 true positive cases, 8 true negative cases, 1 false positive case, and 0 false negative cases. Based on these, the sensitivity of uNGAL 48 hours after was 100 %, specificity 88.89 %, and accuracy was 98.84 %. There is a significant association between uNGAL levels and the ICU length of stay (p = 0.00). Conclusions. Sensitivity analysis was done in cut-off value of 120 for urinary NGAL in predicting AKI. From these results we conclude that urinary NGAL at the time of ICU admission is a reliable marker of renal function in sepsis patients. There is a significant correlation between AKI presence and urinary NGAL, and the ICU length of stay. We recommend not to use uNGAL alone in predicting AKI. It should be combined with glomerular filtration rate to reliably detect AKI development. Study findings indicate that sepsis patients with elevated uNGAL require proper management with close monitoring of blood pressure, urine output and appropriate doses of diuretics to avoid the development of AKI.