老年泌尿外科患者术后谵妄风险列线图模型的构建

目的建立预测老年泌尿外科患者术后谵妄发生风险的列线图模型,为术后谵妄早期识别和治疗提供理论依据。方法回顾性手术治疗的1346例老年患者的临床资料。采用单因素分析和多因素logistic回归模型来分析术后谵妄的预测因素,并根据相关预测因素构建术后谵妄的列线图模型。运用ROC曲线并通过绘制校准图来评估模型的区分度和一致性,使用1000次bootstrapp重采样对模型进行内部验证。结果多因素logistic回归表明,年龄、糖尿病史、Charlson并发症指数、手术时间和美国麻醉医师协会(ASA)分级是老年泌尿外科患者发生术后谵妄的预测因素,并依据上述5个预测因素构建了术后谵妄的风险列线图模型。预测模型的初始曲线下面积(AUC)为0.91,采用重复1000次Bootstrap自抽样方法对模型进行内部验证其AUC为0.90,提示模型区分度良好。校正曲线以及Hosmer-Lemeshow拟合优度检验提示模型预测值与实际观察值之间差异无统计学意义(χ²=7.75,P>0.05),表明该列线图模型校准度良好。结论根据年龄、糖尿病病史、Charlson并发症指数、手术时间以及ASA分级等5个预测因素建立的列线图模型在预测泌尿外科手术老年患者术后谵妄方面具有潜在的临床应用价值,但该模型在常规使用前需要进行外部验证。

Establishment of a nomogram model for predicting the risk of postoperative delirium in elderly patients with urological surgery

Objective To establish a nomogram model for predicting the risk of postoperative delirium in elderly urological patients,and to provide a theoretical basis for early identification and treatment of postoperative delirium.Methods The clinical data of 1346 elderly patients who underwent urological surgery were retrospectively analyzed.Univariate analysis and multivariate logistic regression model were used to analyze the predictors of postoperative delirium.ROC curve and calibration plot were used to evaluate the discrimination and consistency of the model,and 1000 bootstrap resamplings were used to verify the model.Results Multivariate logistic regression showed that age,diabetes history,Charlson complication index,operation time and American Society of anesthesiologists(ASA)classification were predictors of postoperative delirium in elderly patients with urology.A nomogram model of postoperative delirium risk was constructed based on the above five predictors.The area under the curve(AUC)of the prediction model was 0.91.After the 1000 repetitions of bootstrap resampling,the AUC was 0.90.It is suggested that the model has good discrimination.The calibration curve and Hosmer Lemeshow goodness of fit test showed that there was no significant difference between the predicted value and the observed value(χ²=7.75,P>0.05),which indicating that the calibration of the nomogram model was good.Conclusion The nomogram model established based on age,diabetes history,Charlson complication index,operation time and ASA classification has potential clinical application value in predicting postoperative delirium in elderly patients undergoing urological surgery,but the model needs external verification before routine use.