摘要
目的探讨老年患者肺癌根治术后谵妄(POD)的危险因素,并在此基础上构建与验证预测POD发生风险的列线图模型。方法选择择期全麻下行肺癌根治术老年患者580例,男349例,女231例,年龄≥65岁,ASAⅠ—Ⅲ级。根据术后3 d内是否发生POD将患者分为两组:POD组与非POD组。通过Lasso回归筛选与POD发生相关的临床变量,并采用多因素Logistic回归分析确定独立危险因素,以此建立预测POD发生风险的列线图模型。分别通过C-index、校准曲线和受试者工作特征(ROC)曲线验证模型的区分度、一致性和准确性,并采用决策曲线分析(DCA)确定模型的临床实用性。结果有46例(7.93%)患者发生POD。多因素Logistic回归分析显示,年龄≥75岁、术前简易精神状态量表(MMSE)评分≤25分、术前预后营养指数(PNI)<45、查尔森合并症指数(CCI)评分≥2分、鳞癌、术中低血压和手术时间≥3 h为POD的独立危险因素。以此构建的列线图模型经内部验证,该模型的C-index为0.864(95%CI 0.811~0.917);校准曲线显示,该模型预测POD发生风险与实际POD发生风险平均绝对误差为0.038;ROC曲线显示,该模型预测POD发生风险的曲线下面积为0.866(95%CI 0.835~0.892),敏感性86.96%,特异性73.78%;DCA分析显示该模型具有较好的临床实用性。结论年龄≥75岁、术前MMSE评分≤25分、PNI<45、CCI评分≥2分、鳞癌、术中低血压和手术时间≥3 h是肺癌根治术老年患者POD的独立危险因素,依此构建的列线图模型对POD发生风险具有良好的预测效能。
Objective To explore the related risk factors of concurrent postoperative delirium(POD)in elderly patients undergoing radical resection of lung cancer,and a nomogram model to predict the risk of POD was established and validated based on the results.Methods A total of 580 elderly patients undergoing radical resection for lung cancer under general anesthesia,including 349 males and 231 females,aged≥65 years,falling into ASA physical statusⅠ-Ⅲ,were divided into the POD group and the non-POD group according to whether POD occurred within 3 days after operation.The Lasso model was used to screen the clinical variables related to POD,and the independent risk factors was identified by multivariate logistic regression analysis,so as to establish a nomogram model to predict the risk of POD based on the results.The discriminative ability,calibration ability and accuracy of the model were evaluated by C-index,calibration curve and receiver operating characteristic(ROC)curve,respectively.And the clinical practicability of the model was determined by decision curve analysis(DCA).Results POD occurred in 46(7.93%)patients.Multivariate logistic regression analysis showed that aged≥75 years,preoperative mini mental state examination(MMSE)score≤25 points,preoperative prognostic nutritional index(PNI)<45,Charlson complication index(CCI)score≥2 points,squamous cell carcinoma,intraoperative hypotension,and operating time≥3 h were independent risk factors of POD in elderly patients undergoing radical resection of lung cancer.The nomogram model establishment based on the results was internally validated and displayed good discrimination with a C-index of 0.864(95%CI 0.811-0.917).The calibration curve displayed a good consistency with a mean absolute error of 0.038 between predicted POD risk and the actual POD risk.The ROC curve showed a good accuracy with an area under the ROC curve of 0.866(95%CI 0.835-0.892),sensitivity of 86.96%and specificity of 73.78%.And the DCA curve demonstrated that the model had good clinical practicability.Conclusion The aged≥75 years,preoperative MMSE score≤25 points,PNI<45,CCI score≥2 points,squamous cell carcinoma,intraoperative hypotension and operation time≥3 h are independent risk factors of concurrent POD in elderly patients undergoing radical resection of lung cancer,and the nomogram model established based on such results shows good predictive performance for the POD.
作者
李繁
黎仕焕
谢爽
LI Fan;LI Shihuan;XIE Shuang(Depart-ment of Anesthesia,the Second Affiliated Hospital of Hainan Medical University,Haikou 570311,China)
出处
《临床麻醉学杂志》
CAS
CSCD
北大核心
2022年第10期1013-1019,共7页
Journal of Clinical Anesthesiology
关键词
老年
肺癌根治术
术后谵妄
危险因素
列线图
预测模型
Aged
Radical resection of lung cancer
Postoperative delirium
Risk factors
Nomo⁃graph
Predicting model
作者简介
通信作者:谢爽,Email:99945360@qq.com。
