创伤性脑损伤合并脓毒血症患者30 d内死亡风险预测模型的构建和评价  

Construction and evaluation of 30-day death risk prediction model in patients with traumatic brain injury complicated with sepsis

作　　者：牛瑞娜[1] 尚小平[2] 赵亚楠[1] 邵慧杰[1] 李东瑞[1] NIU Ruina;SHANG Xiaoping;ZHAO Yanan;SHAO Huijie;LI Dongrui(Neurological ICU,the First Affiliated Hospital,Zhengzhou University,Zhengzhou 450052;Department of Medical Records Management,the First Affiliated Hospital,Zhengzhou University,Zhengzhou 450052)

机构地区：[1]郑州大学第一附属医院神经ICU,郑州450052 [2]郑州大学第一附属医院病案管理科,郑州450052

出　　处：《郑州大学学报(医学版)》2025年第2期270-273,共4页Journal of Zhengzhou University(Medical Sciences)

基　　金：河南省医学科技攻关计划项目(LHGJ20200267)。

摘　　要：目的:构建预测创伤性脑损伤合并脓毒血症患者30 d内死亡的模型。方法:回顾性分析2021年1月至2024年6月在郑州大学第一附属医院神经ICU住院的创伤性脑损伤合并脓毒血症患者344例。根据患者入院后30 d内的存活情况,分为死亡组和存活组。采用Bootstrap 1000次法和Logistic回归建立预测创伤性脑损伤合并脓毒血症患者30 d内死亡的模型,并绘制列线图。采用ROC曲线、校准曲线和决策曲线评估模型的预测能力、校准能力和临床净获益。结果:344例患者中,116例30 d内死亡。CRP、APACHEⅡ评分、红细胞分布宽度/血小板比值(RPR)和菌种类型是创伤性脑损伤合并脓毒血症患者30 d内死亡的预测因素,OR(95%CI)分别为1.004(1.001~1.006)、1.158(1.112~1.207)、3.328(1.402~3.895)、2.071(1.207~3.552)。列线图模型预测创伤性脑损伤合并脓毒血症患者30 d内死亡的ROC曲线的AUC(95%CI)为0.803(0.752~0.854);校准曲线显示该模型预测曲线与理想曲线基本吻合;决策曲线显示该模型有良好的临床净获益。结论:CRP、APACHEⅡ评分、RPR和菌种类型是创伤性脑损伤合并脓毒血症患者30 d内死亡的预测因素,基于上述因素建立的风险预测模型具有良好的预测性能。Aim:To construct a 30-day death risk prediction model in patients with traumatic brain injury complicated with sepsis.Methods:This study retrospectively analyzed 344 patients with traumatic brain injury complicated with sepsis admitted to the Neurological ICU of the First Affiliated Hospital of Zhengzhou University from January 2021 to June 2024.According to the 30-day survival after admission,the patients were divided into death group and survival group.Logistic regression and Bootstrap 1000 sampling method were used to construct 30-day death risk prediction model,and plot a Nomogram model.ROC analysis,calibration curve,and decision curve were used to evaluate the predictive power,calibration power,and clinical net benefit of the Nomogram model.Results:Among the 344 patients,116 died within 30 days.CRP,APACHEⅡscore,red blood cell distribution width to platelet ratio(RPR)and strain types were the predictive factors for 30-day death in patients with traumatic brain injury combined with sepsis[OR(95%CI)were 1.004(1.001-1.006),1.158(1.112-1.207),3.328(1.402-3.895),2.071(1.207-3.552),respectively].The AUC(95%CI)of 30-day death of the Nomogram model were 0.803(0.752-0.854).The calibration curve showed that the predicted curve of the model was basically consistent with the ideal curve.The decision curve showed that the model had a good clinical net benefit.Conclusion:CRP,APACHEⅡscore,RPR and strain types are predictive factors for 30-day death in patients with traumatic brain injury complicated with sepsis.The risk prediction model based on these factors has good predictive performance.

关 键 词：创伤性脑损伤 脓毒血症 死亡风险预测 

分 类 号：R741.05[医药卫生—神经病学与精神病学]