• Zhi-dong Zang, Ying-zi Huang, Yi Yang, Feng-mei Guo, and Hai-bo Qiu.
• Department of Critical Care Medicine, Zhong-Da Hospital, Southeast University, Nanjing 210009, China.
• Zhonghua Nei Ke Za Zhi. 2010 May 1;49(5):396-9.

ObjectiveTo determine whether urinary neutrophil gelatinase-associated lipocalin (uNGAL) and urinary interleukin-18 (uIL-18) are early markers of acute kidney injury (AKI) in critically ill patients.MethodsNinety-two critically ill patients were studied for one week after their enrollment into our hospital. During the study, 46 patients who met the RIFLE criteria were selected as AKI group and the remaining 46 patients without AKI taken as a control group. The two groups were matched for age, gender and illness severity. Urine samples were collected daily for one week. The receiver operating characteristic curve was used to evaluate the early diagnostic value of uNGAL, uIL-18 and serum creatinine (SCr).ResultsAs compared with the levels obtained 3 days before the diagnosis of AKI, the uNGAL levels in the AKI group increased significantly (P < 0.05), while uIL-18 and SCr levels did not change 2 days prior to the diagnosis of AKI (all P > 0.05). uNGAL and uIL-18 levels increased significantly (all P < 0.05), while SCr levels did not change 1 day prior to the diagnosis of AKI in the AKI group (P > 0.05). The levels of uNGAL, uIL-18 and SCr did not change significantly in the control group during the study period (all P > 0.05). Three days before the diagnosis of AKI, concentrations of uNGAL, uIL-18 and SCr were not the predictive of AKI. Two days before the diagnosis of AKI, the area under the curve (AUC) of uNGAL was 0.840 (95%CI 0.672 - 1.009, P < 0.05), which indicated that uNGAL was the predictive of AKI while uIL-18 and SCr were not. One day before the diagnosis of AKI, the AUC of uNGAL and uIL-18 were 0.830 (95%CI 0.711 - 0.950, P < 0.05) and 0.818 (95%CI 0.697 - 0.938, P < 0.05), indicating that uNGAL and uIL-18 were the predictive of AKI while SCr was not.ConclusionuNGAL and uIL-18 may be the early predictive markers of AKI in critically ill patients.

18 keywords...

hide…