摘要
目的评估循环微小RNA155(miR-155)对缺血性卒中(AIS)患者随访90 d发生不良功能结局(mRS 3~6分)的独立相关性与增量预测价值。方法单中心回顾性队列。连续纳入2018年1月—2019年12月入住我院的AIS患者279例,按7∶3随机分为训练集(n=195)与验证集(n=84)。基于临床与影像学变量构建对照模型,在此基础上加入miR-155(Z值)形成扩展模型。主要结局为90 d不良功能结局。采用多因素logistic回归获得校正效应,比较两模型的区分度、再分层能力、临床净获益与校准价值。结果279例患者随访90 d期间,89例(31.9%)为mRS 3~6分,33例(11.8%)全因死亡,42例(15.1%)出现早期神经功能恶化,14例(5.0%)出现症状性颅内出血,16例(5.7%)合并恶性脑水肿,11例(3.9%)复发卒中/短暂性脑缺血发作,39例(14.0%)院内感染。多因素logistic回归分析结果显示,NIHSS、年龄、血糖、ASPECTS评分与miR-155为90 d不良功能结局的独立危险因素,血管内血栓切除术与较高ASPECTS评分呈保护效应(均P<0.05)。相比对照模型,扩展模型曲线下面积在训练集由0.813提升至0.844,在验证集由0.795提升至0.822,再分层能力也明显提升。结论miR-155能够为AIS患者随访90 d不良功能结局提供独立且具有临床意义的增量信息,能够在保持良好校准的同时提升区分度、再分层能力,可辅助个体化预后评估与管理。
Objective To evaluate the independent association and incremental predictive value of circulating miR-155 for 90-day poor functional outcome(modified Rankin Scale[mRS]3-6)in acute ischemic stroke(AIS).Methods This single-center retrospective cohort consecutively enrolled 279 AIS patients and randomly assigned them to a training set(n=195)and a validation set(n=84)in a 7∶3 ratio.A reference model was constructed from clinical and imaging variables(age,NIHSS,admission glucose,ASPECTS,EVT,and baseline mRS≥1).An expanded model was built by adding miR-155(z-score).The primary endpoint was 90-day poor functional outcome;secondary endpoints included 90-day all-cause mortality,early neurological deterioration(END),symptomatic intracranial hemorrhage(sICH),malignant cerebral edema,90-day recurrence/TIA,and in-hospital infection.Multivariable logistic regression was used to obtain adjusted effects.Discrimination(ROC/AUC),reclassification(NRI/IDI),clinical net benefit(decision curve analysis,DCA),and calibration were compared between the two models.Results Overall event rates were as follows:90-day mRS 3-6,31.9%;all-cause death,11.8%;END,15.1%;sICH,5.0%;malignant cerebral edema,5.7%;90-day recurrence/TIA,3.9%;and in-hospital infection,14.0%.Multivariable logistic regression showed that NIHSS,age,glucose,ASPECTS,and miR-155 remained independent risk factors(P<0.05),whereas EVT and higher ASPECTS exhibited protective effects(P<0.05).Compared with the reference model,the expanded model increased AUC from 0.813 to 0.844 in the training set and from 0.795 to 0.822 in the validation set;NRI/IDI indicated improved reclassification,and Hosmer-Lemeshow testing together with calibration curves suggested good model fit for both.Conclusions miR-155 provides independent and clinically meaningful incremental information for 90-day poor functional outcome after AIS,improving discrimination and reclassification while maintaining good calibration,and may assist individualized prognostic assessment and management.
作者
杨静
张书欣
李守兰
侯霞
YANG Jing;ZHANG Shuxin;LI Shoulan;HOU Xia(Department of Internal Medicine,Zhangjiakou Maternal and Child Health Hospital,Zhangjiakou,China;Department of Neurology,Zhangjiakou First Hospital,Zhangjiakou,China)
出处
《中国神经免疫学和神经病学杂志》
2025年第6期458-465,共8页
Chinese Journal of Neuroimmunology and Neurology
基金
张家口市重点研发计划项目(2322149D)。
