Construction and value analysis of a predictive model for poor prognosis after endovascular treatment in patients with large vessel occlusion⁃acute ischemic stroke
Author:
Affiliation:
Department of Neurology,Zhumadian Central Hospital,Zhumadian 463000 ,China
Fund Project:
摘要
|
图/表
|
访问统计
|
参考文献
|
相似文献
|
引证文献
|
资源附件
|
文章评论
摘要:
目的：构建大血管闭塞性急性缺血性卒中（large vessel occlusion-acute ischemic stroke，LVO-AIS）患者血管内治疗 （endovascular treatment，EVT）术后90 d预后不良的预测模型并探讨其预测价值。方法：回顾性选取2020年1月—2023年12月驻马店市中心医院诊治的LVO-AIS患者200例，依据预后情况分别纳入预后良好组（112例）和预后不良组（88例），比较2组临床资料及治疗相关指标，采用Lasso-Logistic回归分析LVO-AIS患者EVT术后90 d预后不良的影响因素，构建预测模型，以列线图进行可视化呈现，采用校正曲线、受试者工作特征（receiver operating characteristic，ROC）曲线评价模型的预测准确性和预测价值，并进行外部验证。结果：与预后良好组相比，预后不良组合并高血压比例、入院时美国国立卫生研究院卒中量表（National Institutes of Health stroke scale，NIHSS）评分、入院时中性粒细胞与淋巴细胞比值（neutrophil-to-lymphocyte ratio，NLR）、闭塞血管数量≥2 支比例较高，入院时 Alberta 卒中项目早期 CT（Alberta stroke program early CT，ASPECT）评分较低（P ＜ 0.05）；预后不良组发病至手术时间、发病至再通时间长于预后良好组，改良脑梗死溶栓分级（modified thrombolysis in cerebral infarction， mTICI）≥2b级的比例低于预后良好组（P ＜ 0.05）；高血压、入院时NHISS评分、入院时NLR、入院时ASPECT评分、发病至再通时间、mTICI分级是预后的独立影响因素（P ＜ 0.05）；基于各指标构建的联合模型预测LVO-AIS患者EVT术后90 d预后不良的曲线下面积（area under curve，AUC）为0.900，预测价值较高，经外部验证进一步表明，该模型具有可靠的预测价值。结论：高血压、入院时NHISS评分、入院时NLR、入院时ASPECT评分、发病至再通时间、mTICI分级与LVO-AIS患者预后有关，基于各指标构建的联合模型可为患者EVT术后90 d预后预测提供参考，并以此指导临床治疗。
Abstract:
Objective：To construct a predictive model for poor 90-day prognosis after endovascular treatment（EVT）in patients with large vessel occlusion-acute ischemic stroke（LVO-AIS）and to explore its predictive value. Methods：Two hundred patients with LVOAIS diagnosed and treated in Zhumadian Central Hospital from January 2020 to December 2023 were retrospectively selected. According to the prognosis，the patients were divided into a good prognosis group（112 cases）and a poor prognosis group（88 cases）. The clinical data and treatment-related indicators of the two groups were compared. Lasso-Logistic regression was used to analyze the influencing factors of poor 90-day prognosis after EVT in patients with LVO-AIS. The prediction model was constructed and visualized by nomogram. The prediction accuracy and predictive value of the prediction model were evaluated by calibration curve and receiver operating characteristic（ROC）curve，and external validation was performed. Results：Compared with the good prognosis group，the proportion of hypertension，the National Institutes of Health stroke scale（NIHSS）score at admission，the neutrophil-to-lymphocyte ratio （NLR）at admission，and the number of occluded vessels ≥ 2 were higher in the poor prognosis group. The Alberta stroke program early CT（ASPECT）score was lower at admission（P ＜ 0.05）. The time from onset to operation and the time from onset to recanalization in the poor prognosis group were longer than those in the good prognosis group，and the proportion of modified thrombolysis in cerebral infarction（mTICI）grade ≥ 2b in the poor prognosis group was lower than that in the good prognosis group（P ＜ 0.05）. Hypertension， NHISS score at admission，NLR at admission，ASPECT score at admission，time from onset to recanalization，and mTICI grade were independent prognostic factors（P ＜ 0.05）. The AUC of the combined model based on each index to predict the poor 90-day prognosis after EVT of LVO-AIS patients was 0.900，and the predictive value was high. The external validation further showed that the model had reliable predictive value. Conclusion：Hypertension，NHISS score at admission，NLR at admission，ASPECTS score at admission，time from onset to reperfusion，and mTICI classification are related to the prognosis of patients with LVO - AIS. The combined model constructed based on these indicators can provide a reference for predicting the 90 - day prognosis of patients after EVT，and guide clinical treatment accordingly.
