摘要
目的探讨老年创伤性骨折患者术中非计划性低体温(IPH)的风险因素,建立随机森林模型并评估其预测效能,为临床早期防控术中IPH提供参考。方法前瞻性选取2023年7月至2024年11月于河南省洛阳正骨医院(河南省骨科医院)择期行外科手术治疗的326例老年创伤性骨折患者作为研究对象,根据术中是否发生IPH分为发生组与未发生组。比较两组患者的临床资料,采用Logistic回归分析老年创伤性骨折患者术中IPH的影响因素,R-studio软件建立老年创伤性骨折患者术中IPH的随机森林模型,绘制受试者工作特征(ROC)曲线评估其预测效能。结果326例老年创伤性骨折患者术中IPH发生率为23.31%;单因素分析结果显示,两组患者的年龄、术前焦虑、衰弱、主动加温、手术时长、入室体温、术中输液量、术中出血量比较差异均有统计学意义(P<0.05);Logistic回归分析结果显示,年龄、术前焦虑、衰弱、术中输液量均是老年创伤性骨折患者术中IPH的危险因素(P<0.05),而主动加温、入室体温是其保护因素(P<0.05);基于上述因素建立老年创伤性骨折患者术中IPH的随机森林模型,ROC分析结果显示该模型曲线下面积(AUC)为0.967(95%CI:0.910~0.986),重要性排序显示影响术中IPH发生的前5位因素依次为衰弱、术中输液量、年龄、主动加温、术前焦虑。结论通过随机森林算法构建的老年创伤性骨折患者术中IPH预测模型预测性能良好,可为医护人员早期识别术中IPH发生并制定防控措施提供参考。
Objective To explore the risk factors for intraoperative inadvertent perioperative hypothermia(IPH)in elderly patients with traumatic fractures,establish a random forest model,and evaluate its predictive efficacy to provide a reference for early prevention and control of intraoperative IPH in clinical practice.Methods A prospective study included 326 elderly patients with traumatic fractures who underwent elective surgical treatment in Luoyang Orthopedic-Traumatological Hospital of Henan Province(Henan Provincial Orthopedic Hospital)from July 2023 to November 2024.According to whether IPH occurred during the operation,they were divided into an IPH occurrence group and a non-occurrence group.The clinical data of the two groups were compared.Logistic regression analysis was used to identify the influencing factors of intraoperative IPH in elderly patients with traumatic fractures.R-studio software was used to establish a random forest model for intraoperative IPH,and receiver operating characteristic(ROC)curves were drawn to evaluate its predictive efficacy.Results The incidence of IPH among 326 elderly patients with traumatic fractures was 23.31%.Univariate analysis showed that statistically significant differences between the groups were observed in age,preoperative anxiety,frailty,active heating,operation duration,temperature upon entering the operating room,intraoperative infusion volume,and intraoperative blood loss(all P<0.05).Logistic regression analysis indicated that age,preoperative anxiety,frailty,and intraoperative infusion volume were risk factors for intraoperative IPH(P<0.05),while active heating and temperature upon entering the operating room were protective factors(P<0.05).Based on the above factors,a random forest model for intraoperative IPH in elderly patients with traumatic fractures was established.ROC analysis showed that the area under the curve(AUC)of the model was 0.967(95%CI:0.910-0.986).Variable importance ranking identified the top five factors influencing intraoperative IPH occurrence,which were frailty,intraoperative infusion volume,age,active heating,and preoperative anxiety.Conclusion The prediction model of intraoperative IPH in elderly patients with traumatic fracture constructed by the random forest algorithm demonstrates good prediction performance,which can provide a reference for medical staff to identify the occurrence of intraoperative IPH early and formulate prevention and control measures.
作者
沈芳
张航航
黄蒙蒙
张毅
SHEN Fang;ZHANG Hang-hang;HUANG Meng-meng;ZHANG Yi(Operating Room,Luoyang Orthopedic-Traumatological Hospital of Henan Province(Henan Provincial Orthopedic Hospital),Zhengzhou 450000,Henan,CHINA)
出处
《海南医学》
2025年第24期3624-3629,共6页
Hainan Medical Journal
基金
2022年河南省医学科技攻关联合共建项目(编号:LHGJ20221254)。
关键词
老年
创伤性骨折
术中非计划性低体温
随机森林模型
危险因素
Elderly
Traumatic fracture
Inadvertent perioperative hypothermia
Random forest model
Risk factors
