Abstract:

Objective To understand the current situation and influencing factors of delirium in lung transplant patients in the early postoperative period，and to construct and verify a risk prediction model. Methods The convenience sampling method was used to select patients who were admitted to a tertiary general hospital in Jiangsu Province for lung transplantation from June 2023 to November 2024. Lasso regression was used to screen variables，and logistic regression analysis was used to explore the influencing factors of early postoperative delirium in lung transplantation patients，and the risk prediction model was constructed and nomogram was drawn. The area under the working curve（AUC） of the subjects and the Hosmer-Lemeshow test were used to evaluate the discrimination and calibration of the model. Internal validation of the model was performed by repeated sampling 1000 times using Bootstrap method. Results A total of 228 lung transplant patients were included in this study，of which 76 developed delirium in the early postoperative period，with an incidence of 33.33%. Age≥51.5 years，hypertension combined with diabetes mellitus，the Lung Allocation System（LAS） score≥75.63，the duration of mean arterial pressure less than 70 mmHg≥12.50 min，intraoperative red blood cell transfusion≥750 ml，and midazolam dosage≥102.50 mg were the independent risk factors for early postoperative delirium in lung transplantation（P<0.05）. The AUC of the model was 0.771；the optimal cut-off value was 0.334；the sensitivity and specificity were both 0.724. The results of Hosmer-Lemeshow test showed that χ²=5.677，P=0.683；the internal verification showed that the AUC of the model was 0.737，and the absolute error of the actual and predicted values of the calibration curve was 0.023，and the calibration curve was close to the ideal curve. Conclusion Age，hypertension combined with diabetes mellitus，LAS score，duration of mean arterial pressure less than 70 mmHg，amount of intraoperative red blood cell transfusion，and midazolam dosage are the influencing factors for early postoperative delirium in lung transplant patients. The risk prediction model constructed in this study has good predictive performance and can be used as a risk prediction tool for early postoperative delirium in lung transplant patients，helping to timely and accurately identify high-risk groups of delirium，helping to reduce the incidence of early postoperative delirium，and improve patient prognosis.

Key words: Lung Transplantation, Postoperative Delirium, Prediction Model, Nursing Care

摘要：

目的 了解肺移植患者术后早期谵妄发生现状及影响因素，构建风险预测模型并验证。 方法 采用便利抽样法，选取2023年6月至2024年11月于江苏省某三级甲等综合医院进行肺移植手术的患者作为研究对象。采用Lasso回归筛选变量，通过Logistic回归分析探索肺移植患者术后早期谵妄影响因素，构建风险预测模型并绘制列线图。采用受试者操作特征曲线下面积和Hosmer-Lemeshow检验评价模型的区分度和校准度。采用Bootstrap法重复抽样1 000次对模型进行内部验证。 结果 共纳入228例肺移植患者，其中76例术后早期发生谵妄，发生率为33.33%。年龄≥51.5岁、高血压合并糖尿病、肺源分配系统（the Lung Allocation System，LAS）评分≥75.63分、术中平均动脉压<70 mmHg（1 mmHg=0.133 kPa）的持续时间≥12.50 min、术中输注红细胞量≥750 ml、咪达唑仑用量≥102.50 mg是肺移植患者术后早期谵妄的独立危险因素（P<0.05）。该模型的受试者操作特征曲线下面积为0.771，最佳截断值为0.334，灵敏度和特异度均为0.724。Hosmer-Lemeshow检验结果显示χ²=5.677，P=0.683；内部验证显示，模型的受试者操作特征曲线下面积为0.737，校准曲线和实际曲线绝对误差为0.023，校准曲线接近理想曲线。 结论 年龄、高血压合并糖尿病、LAS评分、术中平均动脉压<70 mmHg的持续时间、术中输注红细胞量、咪达唑仑用量是肺移植患者术后早期谵妄的影响因素。构建的风险预测模型预测效能良好，可作为肺移植患者术后早期谵妄风险预测工具，帮助医护人员及时准确识别谵妄高危人群，改善患者预后。

关键词: 肺移植, 术后谵妄, 预测模型, 护理