Objective To explore the influencing factors for the occurrence of delirium in patients with post⁃traumatic acute kidney injury, and to construct a nomogram prediction model. Methods A retrospective analysis on clinical data from 106 patients with post⁃traumatic acute kidney injury was conducted. According to the occurrence status of delirium among patients, patients were divided into delirium group or non⁃delirium group. The clinical data of the two groups were compared, and the multivariate Logistic regression model was adopted to analyze the influencing factors for the occurrence of delirium in patients with post⁃traumatic acute kidney injury. A nomogram prediction model for the occurrence of delirium in patients with post⁃traumatic acute kidney injury was constructed based on its influencing factors and the nomogram prediction model was used for visualization. The receiver operating characteristic curve, Hosmer⁃Lemeshow test and calibration curve were employed to evaluate the diagnostic efficiency, discrimination, and accuracy of the model, respectively. Results Age, Acute Physiology and Chronic Health Evaluation Ⅱ (APACHEⅡ) score, duration of mechanical ventilation, proportion of benzodiazepine use, duration of renal replacement therapy, blood urea nitrogen, and prothrombin time (PT) of patients in the delirium group were longer or higher than those in the non⁃delirium group (P<0.05). Age, APACHEⅡ score, duration of mechanical ventilation, use of benzodiazepins, duration of renal replacement therapy, blood urea nitrogen, and PT were the influencing factors for the occurrence of delirium in patients with post⁃traumatic acute kidney injury (P<0.05). The sensitivity of the nomogram prediction model constructed based on the aforementioned influencing factors for predicting the occurrence of delirium in patients with post⁃traumatic acute kidney injury was 0.89 (95% CI: 0.73-0.97), specificity was 0.71 (95% CI: 0.60-0.83), and an area under the curve was 0.85 (95% CI: 0.76-0.94). The results of goodness of fit test revealed that the discrimination of the model was favorably high (χ2=0.822, P=0.365). The calibration curve indicated that the calibration curve was basically consistent with the ideal curve, exerting a favorably high accuracy. Conclusion Age, APACHEⅡ score, duration of mechanical ventilation, benzodiazepine use, duration of renal replacement therapy, blood urea nitrogen, and PT are associated with the occurrence of delirium in patients with post⁃traumatic acute kidney injury. The construction of a nomogram prediction model helps to early identify the occurrence of delirium.

无