老年喉癌患者术后谵妄列线图预测模型的建立与验证

Establishment and validation of a predictive nomogram model for predicting postoperative delirium in elderly laryngeal cancer patients

目的探讨老年喉癌患者发生术后谵妄(Postoperative delirium,POD)的危险因素,建立列线图模型并验证其效能。方法收集2017年6月~2023年1月福建省肿瘤医院收治的426例老年喉癌患者作为建模组,采用LASSO回归、十折交叉验证筛选POD的相关变量,Logistic回归分析确认独立危险因素,采用R软件建立列线图模型并进行内部验证。选取2020年3月~2022年12月福建医科大学附属协和医院收治的135例老年喉癌患者作为验证组,对模型进行外部验证。结果建模组和验证组的基线资料比较,差异无统计学意义(P>0.05)。建模组中,年龄、糖尿病、美国麻醉医师协会(American society of anesthesiologists,ASA)分级、术前血清白蛋白是POD的独立危险因素(P<0.05)。建模组和验证组的受试者工作特征曲线下面积(Area under the receiver operator characteristic curve)分别为0.803(95%CI:0.746~0.959)和0.790(95%CI:0.701~0.879)。两组的校正曲线表明预测结果和实际结果的一致性良好。结论本研究建立的老年喉癌患者POD列线图模型具有较好的预测效能,有助于临床医护人员早期识别高风险老年喉癌患者。

Objective To explore the risk factors of postoperative delirium(POD)in elderly laryngeal cancer patients,establish a nomogram model and validate its effectiveness.Methods 426 elderly patients with laryngeal cancer admitted to Fujian Cancer Hospital from Jun 2017 to Jan 2023 were collected as the modeling group.LASSO regression and ten fold cross validation were used to screen for relevant variables of POD,Logistic regression analysis was used to confirm independent risk factors,and R software was used to establish predictive nomogram model.135 elderly laryngeal cancer patients treated at Fujian Medical University Union Hospital from Mar 2020 to Dec 2022 were selected as the validation group for external validation of the model.Results The difference in baseline data between the two groups was not statistically significant(P>0.05).In the modeling group,age,diabetes,American Society of Anesthesiologists(ASA)classification and preoperative serum albumin were independent risk factors for POD(P<0.05).The area under the receiver operator characteristic curve for the modeling group and the validation group were 0.803(95%CI:0.746~0.959)and 0.790(95%CI:0.701~0.879),respectively.The calibration curves of the two groups indicated that there was good consistency between the predicted results and actual results.Conclusion The predictive nomogram model for predicting POD in this study has good predictive effectiveness for elderly laryngeal cancer patients,which is helpful for clinical healthcare professionals to identify high-risk elderly laryngeal cancer patients in the early stage.