[关键词]
[摘要]
目的 分析老年颈椎手术后亚谵妄(subsyndromal delirium,SDD)转谵妄的影响因素,并构建列线图风险预测模型,为降低患者谵妄发生提供依据。 方法 以便利抽样法选取2021年5月至2024年8月某院老年颈椎手术后SDD患者228例及同期另一医院的同类患者118例为研究对象,根据术后是否存在SDD转谵妄,将其分为SDD组和SDD转谵妄组。按3∶1倾向性匹配后再按7∶3分为训练集和验证集。训练集构建列线图模型,验证集评估模型预测效能。 结果 共匹配248例患者,其中SDD 186例、SDD转谵妄62例。术前衰弱、第1次意识模糊评估特征数、术前禁食物时间、术中输注红细胞、术前营养风险指数均为SDD转谵妄的影响因素(均P<0.05)。训练集、验证集模型预测曲线下面积为0.843和0.822,敏感度为79.07%和73.70%,特异性为78.46%和73.20%,Youden指数为0.575和0.469。 结论 构建的列线图风险预测模型能预测老年颈椎术后患者SDD转谵妄风险,有较好的应用价值。
[Key word]
[Abstract]
Objective To analyze the influencing factors of subsyndromal delirium (SDD) progressing to delirium after cervical spine surgery in the elderly patients,and to construct a nomogram risk prediction model,for providing a basis for reducing the occurrence of delirium in patients.Methods A total of 228 elderly patients with SDD after cervical spine surgery in a hospital from May 2021 to August 2024 and 118 similar patients from another hospital during the same period were selected by the convenience sampling method,and were divided into the SDD group and the SDD-to-delirium group according to whether SDD turned into delirium after surgery.After propensity score matching at a ratio of 3∶1 for the two groups,they were further divided into a training set and a validation set at a ratio of 7∶3.The nomogram model was constructed using the training set,and the predictive performance of the model was evaluated using the validation set.Results A total of 248 patients were matched,including 186 from SDD group and 62 from SDD-to-delirium group.Preoperative frailty,the number of features in the first confusion assessment,preoperative fasting time,intraoperative red blood cell transfusion,and preoperative Nutritional Risk Index(NRI) were all influencing factors for SDD-to-delirium(all P<0.05).The areas under the predicted curves of the training set and validation set models were 0.843 and 0.822,the sensitivities were 79.07% and 73.70%,the specificities were 78.46% and 73.20%,and the Youden indices were 0.575 and 0.469.Conclusions The constructed risk prediction model can predict the risk of postoperative SDD progressing to delirium,and has good clinical application value.
[中图分类号]
R473;R473.59;R823
[基金项目]
江苏省重点研发计划专项课题(BE2019653);第九O四医院院管科研课题(MS202111);江苏省高层次卫生人才“六个一工程” 拔尖人才项目(LGY2019026)