ABSTRACT Objective: This study aimed to explore the risk factors and prediction of postoperative delirium (POD) in aortic arch replacement. Methods: The preoperative, intraoperative, and postoperative clinical data of patients undergoing aortic arch replacement in our hospital from January 1, 2018, to March 31, 2019, were retrospectively analyzed. According to whether POD occurred, cases were divided into delirium group and control group. The clinical data of the patients were firstly screened by univariate analysis, and then the meaningful variables were analyzed by multivariate logistic regression analysis. The receiver operating characteristic curve was used to analyze the meaningful factors in predicting the occurrence of POD after aortic arch replacement. Results: A total of 162 patients were included in the study. The prevalence of POD was 46.9% (n = 76). The results of univariate and multivariate analyses showed that the intraoperative minimum mean arterial pressure (MAP) (odds ratio [OR] = 0.635, 95% confidence interval [CI] = 0.5510.731, P < 0.001), postoperative C-reactive protein (OR = 1.010, 95% CI = 1.0031.018, P = 0.005), and postoperative infection (OR = 3.148, 95% CI = 1.1188.867, P = 0.030) were the independent risk factors for POD in aortic arch replacement. Among them, the intraoperative minimum MAP has the best prediction effect. When its cutoff value reaches ≤ 49 mmHg, the specificity for predicting POD is 84.88% and the sensitivity is 81.58%. Conclusions: A lower intraoperative minimum MAP is an independent risk factor and has the predictive value for POD in aortic arch replacement.