不同时间点改良早期预警评分对院内心脏骤停复苏后患者住院死亡的预测价值OA
Prognostic value of modified early warning scores at different time points for predicting in-hospital mortality after in-hospital cardiac arrest resuscitation
目的 探讨院内心脏骤停(IHCA)后自主循环恢复(ROSC)的患者,不同时间点改良早期预警评分(MEWS)对住院死亡的预测价值.方法 回顾性分析自贡市第四人民医院(2019-2023 年)、富顺县人民医院(2019-2020 年)收治的IH-CA后ROSC患者的临床资料.计算入院时、心脏骤停发生前最近一次及 ROSC 后即刻 MEWS 评分(MEWS1、MEWS2、MEWS3).采用多因素Logistic回归分析探讨住院死亡的独立危险因素.应用限制性立方样条(RCS)评估MEWS评分与住院死亡之间的剂量反应关系.通过受试者工作特征(ROC)曲线评价预测效能,采用Bootstrap法进行内部验证.根据DeLong检验对不同的AUC进行比较.进行K-means聚类分析探究是否存在潜在类别.结果 本研究共纳入 596 例IHCA后ROSC患者,住院期间死亡446 例(74.8%).多因素Logistic回归分析显示,MEWS2(OR=1.094,95%CI:1.002~1.195,P=0.045)及 MEWS3(OR=1.421,95%CI:1.282~1.575,P<0.001)评分为IHCA患者住院死亡的独立危险因素,RCS分析提示MEWS2 和MEWS3 评分与死亡风险呈线性关系(非线性P>0.05).ROC曲线分析结果表明,MEWS3 预测住院死亡的效能最高(AUC=0.729,95%CI:0.681~0.777),并经Bootstrap验证结果一致.DeLong检验结果同样提示,MEWS3预测效能更高(P<0.05).K-means聚类分析结果提示,3 次MEWS评分呈急剧上升型的住院死亡率最高,低分平稳型的存活率最高(P<0.05).结论 IHCA患者MEWS3 较MEWS1 及MEWS2 具有更高的住院死亡预测效能,可为复苏后患者的早期风险评估与临床决策提供可靠依据.
Objective To investigate the prognostic value of the modified early warning score(MEWS)assessed at different time points for predicting in-hospital mortality in patients with in-hospital cardiac arrest(IHCA)who achieved return of spontaneous circulation(ROSC).Methods We performed a retrospective cohort study invol-ving IHCA patients who attained ROSC and were hospitalized at Zigong Fourth People's Hospital(2019-2023)and Fushun County People's Hospital(2019-2020).MEWS was calculated on admission,at the most recent assessment before cardiac arrest,and immediately after ROSC,defined as MEWS1,MEWS2,and MEWS3,re-spectively.Multivariable Logistic regression was used to identify independent predictors of death.Restricted cu-bic spline(RCS)modelling explored the dose-response relationship between MEWS and mortality.Discrimina-tion was assessed with receiver operating characteristic(ROC)analysis,internally validation via bootstrap resa-mpling.Differences in area under the curve(AUC)were compared using the DeLong test.K-means cluster a-nalysis was performed to explore potential latent classes.Results A total of 596 patients were included,and 446(74.8%)died during hospitalization.Multivariable logistic regression showed that MEWS2(OR=1.094,95%CI:1.002~1.195,P=0.045)and MEWS3(OR=1.421,95%CI:1.282~1.575,P<0.001)were inde-pendent risk factors for in-hospital mortality.RCS analysis showed linear associations between MEWS2,MEWS3 and mortality risk(P>0.05 for nonlinearity).MEWS3 had the highest predictive value for in-hospital mortality(AUC=0.729,95%CI:0.681~0.777),which was confirmed by bootstrap validation.DeLong test showed that MEWS3 had superior predictive performance(P<0.05).K-means cluster analysis showed that patients with a sharply increasing MEWS pattern had the highest in-hospital mortality,whereas those with a low and sta-ble pattern had the highest survival rate(P<0.05).Conclusion MEWS measured immediately after ROSC of-fers the best accurate prediction of in-hospital death in IHCA patients.Incorporating this score into early post-re-suscitation evaluations could support timely risk stratification and guide clinical decision-making.
陈涌;钟志涛;高霞;黄霞;任艳;段海真;徐平
川北医学院 临床医学院,四川 南充 637000||自贡市第四人民医院 急诊科,四川 自贡 643000自贡市第四人民医院 急诊科,四川 自贡 643000||自贡市医学大数据与人工智能研究院 医学大数据研究所,四川 自贡 643000自贡市第四人民医院 急诊科,四川 自贡 643000自贡市第四人民医院 急诊科,四川 自贡 643000自贡市第四人民医院 急诊科,四川 自贡 643000遵义医科大学附属医院 急诊科,贵州 遵义 563000川北医学院 临床医学院,四川 南充 637000||自贡市第四人民医院 急诊科,四川 自贡 643000||自贡市医学大数据与人工智能研究院 医学大数据研究所,四川 自贡 643000
医药卫生
院内心脏骤停改良早期预警评分心脏骤停复苏后院内生存预测评分风险评估预测价值
in-hospital cardiac arrestmodified early warning scorecardiac arrest survival prediction scorerisk assessmentprognostic value
《遵义医科大学学报》 2026 (8)
857-864,8
国家自然科学基金资助项目(NO:82260385)四川省科技计划资助项目(NO:2024JDKP0021).
评论