OBJECTIVE: To analyze the risk factors of renal replacement therapy (RRT) in acute kidney injury (AKI) patients after liver transplantation, and to investigate the prognosis effect of initial RRT treatment time. METHODS: Clinical data of 132 recipients undergoing organ donation for cardiac death (DCD) allograft orthotopic liver transplantation admitted to Ningbo Medical Center Lihuili Hospital and Ningbo Medical Center Lihuili Eastern Hospital from July 2014 to July 2018 was retrospectively analyzed. AKI was defined and staged by the criteria of Kidney Disease Improving Global Outcomes (KDIGO) guideline in the first 7 days. According to the implementation of RRT, the patients were divided into non-RRT group and RRT group. The differences in gender, age, body mass index (BMI), model for end-stage liver disease with serum sodium (MELD-Na) score, serum creatinine (SCr), and intraoperative norepinephrine (NE) dose, blood loss, fluid infusion, anhepatic phase time, duration of operation between two groups were compared. The statistically significant risk factors of AKI found by univariate analysis were selected and analyzed to find independent risk factors of RRT in AKI patients after liver transplantation with multivariate Logistic regression analysis. The receiver operating characteristic (ROC) curve was drawn to evaluate the test efficiency of all risk factors of RRT implementation. According to the implementation of RRT on KDIGO stage-2, all the patients on KDIGO stage-2 and stage-3 were divided into early group (initial RRT on KDIGO stage-2) and delayed group (including self-improvement without RRT on KDIGO stage-2 and initial RRT on KDIGO stage-3). The duration of mechanical ventilation, the length of intensive care unit (ICU) stay, AKI duration, incidence of catheter related bloodstream infection (CRBSI) and 28-day mortality were compared between the two groups. RESULTS: All 132 receptors were enrolled in the final analysis, and 77 patients developed AKI, accounting for 58.3%, among which 52 cases were in RRT group (67.5%) and 25 were in non-RRT group (32.5%). As shown by univariate analysis, the MELD-Na score (21.6±4.4 vs. 18.0±4.3), intraoperative NE dose (μg×kg×h: 7.5±1.2 vs. 5.2±1.7), blood loss [mL: 3 000 (2 200, 4 000) vs. 2 600 (1 800, 3 200)], fluid infusion [mL: 6 400 (4 500, 7 800) vs. 5 600 (4 200, 6 800)], and anhepatic period (minutes: 65.6±4.5 vs. 63.0±5.0) were significantly increased in RRT group as compared with those in non-RRT group (all P < 0.05). There was no significant difference in gender, age, BMI, SCr before operation or the duration of operation. It was shown by multivariate Logistic regression analysis that MELD-Na score before operation [odds ratio (OR) = 1.398, 95% confidence interval (95%CI) = 1.062-1.841, P = 0.017], intraoperative NE dose (OR = 4.724, 95%CI = 2.036-10.961, P = 0.000) and fluid infusion (OR = 1.002, 95%CI = 1.001-1.004, P = 0.010) were independent risk factors of RRT implementation in AKI patients after liver transplantation. It was shown by ROC curve analysis that the area under the ROC curve (AUC) of MELD-Na score, NE dose and fluid infusion for predicting the implementation of RRT in AKI patients after liver transplantation was 0.719, 0.867, and 0.670, respectively, which suggesting that NE dose had moderate predictive value, but MELD-Na score and fluid infusion had low predicative value. When the optimal cut-off value of NE dose was 6.5 μg×kg×h, the sensitivity was 84.6% and the specificity was 80.0%. The 28-day mortality was both 0 in early group (n = 25) and delayed group (n = 39). Compared with the early group, the duration of mechanical ventilation (hours: 41.0±1.0 vs. 35.8±6.7) and the length of ICU stay (hours: 98.8±6.6 vs. 94.2±7.3) were significantly increased in delayed group (both P < 0.05), there was no significant difference in AKI duration (days: 11.8±4.2 vs. 10.6±4.9) or the incidence of CRBSI [5.1% (2/39) vs. 4.0% (1/25), both P > 0.05]. CONCLUSIONS: MELD-Na score, intraoperative NE dose and fluid infusion were the independent risk factors of RRT implementation in AKI patients after liver transplantation. NE dose had moderate predictive value, but MELD-Na score and fluid infusion had low predicative value. Initial RRT on KDIGO stage-2 could reduce the duration of mechanical ventilation and the length of ICU stay.