基于影像组学列线图模型预测急性缺血性脑卒中血管内治疗后恶性脑水肿的发生风险OA
Predicting Malignant Brain Edema Risk After Endovascular Therapy in Acute Ischemic Stroke Using a Radiomics Nomogram Model
目的 基于影像学指标探讨血管内治疗(EVT)后急性缺血性脑卒中(AIS)患者发生恶性脑水肿(MBE)风险的预测因素.方法 选取2020年8月-2024年8月于我院收治的480例接受EVT后的AIS患者为研究对象,建模组300例,根据是否发生MBE分为并发组(n=76)和非并发组(n=224);验模组180例,根据是否发生MBE分为并发组(n=46)和非并发组(n=134).收集患者临床资料,采用Logistic回归风险模型分析建模组AIS患者接受EVT后发生MBE的相关因素;利用R软件的"rms"包构建列线图模型并验证其临床效能.结果 在建模组中,与非并发组相比,并发组患者年龄较高,高血压、基底池消失占比较高,Alberta卒中项目早期CT(ASPECT)评分较低,入院NIHSS评分较高;并发组与非并发组相比,闭塞血管部位、缺血性卒中的病因(TOAST)分型相比差异有统计学意义(P<0.05);年龄、高血压、基底池消失、ASPECT评分、入院NIHSS评分、闭塞血管部位为AIS患者接受EVT治疗后发生MBE的独立影响因素(P<0.05);ROC曲线分析结果显示,建模组AUC=0.956(95%CI:0.933~0.979),校准曲线分析显示该模型C-index为0.882(95%CI:0.841~0.918);决策曲线显示,基于列线图预测结果进行干预后,当风险阈值>0.08时,能提供更多临床净收益;验模组校正曲线显示患者并发MBE的C-index为0.851(95%CI:0.800~0.897);H-L拟合优度检验为x2=8.011,P=0.432;DCA曲线显示,利用该模型对接受EVT治疗的AIS患者进行预测可获得较高的正向净收益.结论 年龄、高血压、基底池消失、ASPECT评分、入院NIHSS评分、闭塞血管部位对于AIS患者接受EVT后发生MBE的早期预测具有重要意义,可以为临床评估和防治提供决策.
Objective To explore the predictive factors of malignant brain edema(MBE)in patients with acute ischemic stroke(AIS)after endovascular therapy(EVT)based on imaging indicators.Methods A total of 480 patients with AIS who underwent EVT at our hospital from August 2020 to August 2024 were enrolled as study subjects.Among them,300 patients in the modeling group were divided into concurrent group(n=76)and non-concurrent group(n=224)according to whether MBE occurred.According to the occurrence of MBE,180 patients in the test group were divided into concurrent group(n=46)and non-concurrent group(n=134).The clinical data of patients were collected.Logistic regression risk model was used to analyze the related factors of MBE after EVT in patients with AIS in the modeling group.The"rms"package of R software was used to construct a nomogram model and verify its clinical efficacy.Results In the modeling group,compared with the non-concurrent group,the patients in the concurrent group were older,the proportion of hypertension and basal cistern disappearance was higher,the Alberta stroke project early CT(ASPECT)score was lower,and the admission NIHSS score was higher.There were significant differences in the location of occlusive vessels and the etiology of ischemic stroke(TOAST)between the concurrent group and the non-concurrent group(P<0.05).Age,hypertension,basal cistern disappearance,ASPECT score,NIHSS score on admission,and occlusive vascular site were independent influencing factors of MBE in AIS patients after EVT(P<0.05).The results of ROC curve analysis showed that the AUC of the modeling group was 0.956(95%CI:0.933-0.979),and the calibration curve analysis showed that the C-index of the model was 0.882(95%CI:0.841-0.918).The decision curve showed that after the intervention based on the prediction results of the nomogram,when the risk threshold was greater than 0.08,it could provide more clinical net income;the C-index of MBE was 0.851(95%CI:0.800-0.897).The H-L goodness of fit test was x2=8.011,P=0.432;the DCA curve showed that using this model to predict AIS patients receiving EVT treatment coud obtain higher positive net benefits.Conclusion Age,hypertension,basal cistern disappearance,ASPECT score,admission NIHSS score,and occluded vascular site are of great significance for the early prediction of MBE in patients with AIS after EVT,which can provide decision-making for clinical evaluation and prevention.
余志辉;姜军
乐平市中医医院医学影像科,江西 乐平 333300乐平市中医医院医学影像科,江西 乐平 333300
医药卫生
急性缺血性脑卒中血管内治疗恶性脑水肿影响因素预测模型
Acute ischemic strokeEndovascular therapyMalignant cerebral edemaInfluencing factorsPrediction model
《医学信息》 2026 (18)
22-28,7
江西省卫生健康委科技计划(编号:SKJP202018343)
评论