摘要
目的 构建老年髋部骨折患者围手术期并发肺炎的风险预测回归模型.方法 回顾性分析2017年5月至2020年12月就诊于首都医科大学附属北京积水潭医院的2669例老年髋部骨折患者的临床资料,根据围手术期是否并发肺炎将研究对象分为并发肺炎组(并发肺炎,322例)和未并发肺炎组(未并发肺炎,2347例).采用单因素分析和多因素logistic回归方法分析老年髋部骨折患者围手术期并发肺炎的独立影响因素,并构建logistic回归模型.采用受试者操作特征曲线分析回归模型对老年髋部骨折患者围手术期并发肺炎的预测价值.结果 单因素分析结果显示,性别、年龄、骨折类型、手术方式、住院时间、呼吸系统疾病、血液系统疾病、神经系统疾病、泌尿系统疾病、急性冠脉综合征、谵妄、低蛋白血症、电解质紊乱与老年髋部骨折患者围手术期并发肺炎显著相关(P＜0.05).多因素logistic回归分析结果显示,性别、年龄、骨折类型、手术方式、住院时间、呼吸系统疾病、急性冠脉综合征、低蛋白血症均为老年髋部骨折患者围手术期并发肺炎的独立影响因素(P＜0.05),构建回归模型为logit(P)=0.574×性别+0.05×年龄+0.705×骨折类型+0.545×手术方式+0.149×住院时间+0.772×呼吸系统疾病+0.849×急性冠脉综合征+0.702×低蛋白血症-8.002.受试者操作特征曲线分析结果显示,该回归模型预测老年髋部骨折患者围手术期并发肺炎的曲线下面积为0.751,敏感度为0.699,特异度为0.684.结论 性别、年龄、骨折类型、手术方式、住院时间、呼吸系统疾病、急性冠脉综合征、低蛋白血症为老年髋部骨折患者围手术期并发肺炎的独立影响因素,基于上述影响因素构建的风险预测回归模型对老年髋部骨折患者围手术期肺炎的发生具有良好预测价值.
Abstract
Objective To construct the regression model for predicting the risk of perioperative pneumonia in elderly patients with hip fracture.Method The clinical data of 2669 elderly patients with hip fracture admitted to Beijing Jishuitan Hospital,Capital Medical University from May 2017 to December 2020 were retrospectively analyzed.The subjects were divided into pneumonia group(322 cases with pneumonia)and non-pneumonia group(2347 cases without pneumonia)according to whether with pneumonia during perioperative period.Univariate and multivariate logistic regression methods were used to analyze the independent influencing factors of perioperative pneumonia in elderly patients with hip fracture,and logistic regression model was established.The predictive value of regression model for perioperative pneumonia in elderly patients with hip fracture was analyzed by ROC curve.Result The results of univariate analysis showed that gender,age,fracture type,operation method,length of stay,respiratory diseases,hematological diseases,nervous system diseases,urinary system diseases,acute coronary syndrome,delirium,hypoproteinemia and electrolyte disturbance were significantly correlated with perioperative pneumonia in elderly patients with hip fracture(P＜0.05).Multivariate logistic regression analysis showed that gender,age,fracture type,operation method,length of stay,respiratory disease,acute coronary syndrome and hypoproteinemia were independent influencing factors for perioperative pneumonia in elderly patients with hip fracture(P＜0.05).The regression model was established as follows:logit(P)= 0.574×gender+0.05×age+0.705×fracture type+0.545×operation method+0.149×length of stay+0.772×respiratory disease+0.849×acute coronary syndrome+0.702×hypoproteinemia-8.002.The ROC analysis showed that the area under the curve,sensitivity and specificity of this regression model for predicting perioperative pneumonia in elderly patients with hip fracture were 0.751,0.699 and 0.684.Conclusion Gender,age,fracture type,operation method,length of stay,respiratory disease,acute coronary syndrome and hypoproteinemia were independent influencing factors for perioperative pneumonia in elderly patients with hip fracture,and the regression model built based on the above influencing factors has good predictive value for the occurrence of perioperative pneumonia in elderly patients with hip fracture.
基金项目
北京市医院管理中心"青苗"计划专项(QML20210405)
北京积水潭医院"学科骨干"计划专项(XKGG202204)
NETL
NSTL
维普
万方数据