Abstract: OBJECTIVE To investigate the risk factors of infectious complications after hepatectomy for hepatocellular carcinoma and the construction of a predictive model. METHODS A total of 148 patients admitted to the First Affiliated Hospital of Xinjiang Medical University for hepatic resection of hepatocellular carcinoma between Jan.2015 and Oct.2021 were retrospectively selected as the study subjects, and were divided into infection group (n=26) and non-infection group (n=122) according to presence or absence of postoperative infectious complications.The pathogenic bacteria of patients in the infection group were analyzed, the clinical data including age, operation time, serum albumin (Alb) were compared between the two groups.The risk factors for the occurrence of infectious complications were analyzed by multivariate Logistic regression, the risk prediction model was established, and the efficiency of the model in predicting the occurrence of infectious complications was tested using the receiver operating characteristic (ROC) curve. RESULTS A total of 45 strains of pathogens were cultured and isolated from 26 patients with post-hepatectomy infections for hepatocellular carcinoma, of which 26 strains of Gram-negative bacteria accounted for 57.79%, mainly Escherichia coli (28.54%) and Klebsiella pneumoniae (24.44%); 18 strains of Gram-positive bacteria accounted for 40.00%, mainly Streptococcus pneumoniae; and 1 strains of fungi accounted for 2.22%.Logistic regression analysis showed that combined diabetes mellitus, Alb <35 g/L and indwelling time of abdominal drainage tube ≥7 d were risk factors for infectious complications in patients undergoing hepatectomy for hepatocellular carcinoma.The model expression of the risk prediction model was as follow: P=1/1+e^{(-4.222+1.062×(diabetes mellitus)+1.261×(Alb <35 g/L)+0.956×(indwelling time of abdominal drainage tube ≥7 d)}, Hosmer-Lemeshow test χ²=3.204, P=0.921.The area under the curve (AUC) of the model for predicting infectious complications was 0.767 (P＜0.001), showing good fit and predictive efficiency. CONCLUSION Combined diabetes mellitus, Alb<35 g/L and indwelling time of abdominal drainage tube ≥7 d were risk factors for infectious complications in patients undergoing hepatectomy for hepatocellular carcinoma.The risk prediction model established based on the Logistic regression model could better predict the risk of infectious complications in patients, and targeted interventions should be formulated clinically based on the above risk factors to prevent the occurrence of postoperative infection.