摘要
目的 探索智能多时相CTA(mCTA)侧支循环自动化评分联合CT灌注成像(CTP)定量参数及临床指标,综合评估影响急性缺血性脑卒中(AIS)患者行血管内机械取栓术(EVMT)预后的因素.方法 本研究为回顾性研究.按照纳入与排除标准,搜集2017年1月—2022年12月于苏州大学附属第一医院收治的111例行EVMT治疗的AIS患者的病例资料.mCTA图像采用NeuBrainCARE(NBC)软件处理,对侧支循环进行自动化评分,并与人工Menon评分进行一致性检验.CTP图像经后处理得到相关参数.按照90 d mRS评分,分为预后良好组(60例)、预后不良组(51例),比较两组患者的临床资料、脑卒中危险因素、大脑闭塞部位、缺血性脑卒中病因分型(TOAST)、发病至CTA检查与治疗时间、入院美国国立卫生研究所卒中量表(NIHSS)评分等.采用ROC曲线分析侧支循环自动化评分、CTP参数的预测效能.组间符合正态分布的计量资料以xˉ±s表示、采用t检验,非正态分布资料以M(Q1,Q3)表示、采用Mann-Whitney U检验;分类变量以例数(％)表示、采用x2检验或Fisher精确概率检验.结果 2名医师对侧支循环人工评分一致性检验较好(Kappa=0.667),自动化评分与2名医师评分的一致性检验较好(Kappa值分别为0.872、0.597).预后良好组患者的年龄、心房颤动例数、入院NIHSS评分、核心梗死体积(rCBF<30％)以及低灌注体积(Tmax>10 s)均显著低于预后不良组;mCTA侧支循环自动化评分显著高于预后不良组,差异均有统计学意义(P均<0.05).ROC曲线分析显示,临床资料(年龄、房颤例数、入院NIHSS评分)、侧支循环自动化评分、CTP参数对AIS患者EVMT临床预后的预测效能AUC分别为0.75、0.74、0.59,上述指标联合的预测效能AUC为0.81.结论 智能mCTA侧支循环自动化评分与高年资医师人工评分有较高的一致性,评估EVMT结局的效能较好.临床相关指标、智能mCTA侧支循环自动化评分及CTP参数联合评估AIS患者EVMT预后的效能高于其中任一因素,为临床AIS患者取栓诊疗提供可靠依据.
Abstract
Objective To explore the smart multitemporal CTA (mCTA) collateral circulation automation score combined with quantitative CT perfusion imaging (CTP) parameters and clinical indicators to comprehensively assess the factors affecting the prognosis of patients with acute ischaemic stroke (AIS) undergoing endovascular mechanical thrombectomy (EVMT). Methods This study was a retrospective study. According to the inclusion and exclusion criteria,case data were collected from 111 AIS patients treated with EVMT admitted to the First Affiliated Hospital of Soochow University from January 2017 to December 2022. mCTA images were processed using NeuBrainCARE (NBC) software for automated scoring of collateral circulation and tested for consistency with the manual Menon Scoring was tested for consistency. CTP images were post-processed to obtain relevant parameters. According to the 90 d mRS score,the patients were divided into a good prognosis group (60 cases) and a poor prognosis group (51 cases),and the two groups were compared in terms of their clinical data,stroke risk factors,cerebral occlusion sites,ischaemic stroke etiological subtyping (TOAST),time between onset of the disease and CTA examination and treatment,and National Institutes of Health Stroke Scale (NIHSS) score on admission to the hospital. ROC curves were used to analyse the predictive efficacy of the collateral circulation automation score and CTP parameters. Measurements with normal distribution between groups were expressed as xˉ±s by t-test,and non-normally distributed data were expressed as M(Q1,Q3) by Mann-Whitney U-test;categorical variables were expressed as the number of cases (％) by x2-test or Fisher's exact probability test. Results The test of agreement between the manual scoring of collateral circulation by 2 physicians was good (Kappa=0.667),and the test of agreement between the automated scoring and the scoring by 2 physicians was good (Kappa values of 0.872 and 0.597,respectively). The age,number of AF cases,admission NIHSS score,core infarct volume (rCBF<30％),and hypoperfusion volume (Tmax>10 s) of the patients in the good prognosis group were significantly lower than those in the poor prognosis group. mCTA collateral circulation automation scores were significantly higher than those in the poor prognosis group,and the differences were all statistically significant (all P<0.05). ROC curve analysis showed that the clinical data (age,number of AF cases,admission NIHSS score),collateral circulation automation score,and CTP parameters had an AUC of 0.75,0.74,and 0.59,respectively,and the combined predictive efficacy of the above metrics had an AUC of 0.81 for the clinical prognosis of EVMT in patients with AIS. Conclusion The smart mCTA automated scoring of collateral circulation has a high concordance with manual scoring by senior physicians and has a better efficacy in assessing EVMT outcomes. The combined efficacy of clinically relevant indexes,smart mCTA automated scoring and CTP parameters in assessing the prognosis of EVMT in AIS patients was higher than either of the factors,providing a reliable basis for the clinical diagnosis and treatment of AIS patients with thrombus extraction.
维普
万方数据