建立急性缺血性卒中的临床/多模式CT预后评分系统

摘要

目的 建立一个简便、有效的临床/多模式CT评分系统,用以指导急性缺血性卒中患者的临床治疗和评估90 d后临床功能恢复情况.方法 选择49例急性缺血性卒中(发病时间＜9 h)的患者行“多模式CT”扫描,包括平扫CT (noncontrast enhanced computed tomography,NCCT)、CT灌注成像(computed tomography perfusion,CTP)和CT血管成像(computed tomography angiography,CTA)；评价患者基线NCCT、动脉期CTP原始图(arterial phase CTP source images,ACTP-SI)、静脉期CTP原始图(venous phase CTP source images,VCTP-SI)、CTA卒中溶栓分级(thrombolysis in cerebral ischemia scale,TICI)、Alberta卒中项目早期CT评分(Alberta Stroke Program Early CT Score Study,ASPECTS)及基线美国国立卫生研究院卒中量表(National Institute of Health Stroke Scale,NIHSS)评分,并应用受试者工作特征曲线(receiver-operating characteristics,ROC)分析,判断90 d临床功能恢复良好[采用改良的Rankin量表(modified Rankin Scale,mRS) ＜2作为评判标准]的临床和CT参数阈值；按照获得的阈值进行评分,将多模式CT各参数的阈值评分整合在一起获得多模式CT评分系统,将基线NIHSS阈值评分加入多模式CT评分系统中获得临床/多模式CT评分.最后应用ROC曲线分析比较各评分模式预测临床功能恢复的效能.结果 判断90 d临床功能恢复良好的阈值:临床/多模式CT评分＞1,多模式CT评分＞1,基线NCCTASPECTS＞9,动脉期CTP原始图ASPECTS＞6.5,静脉期CTP原始图ASPECTS＞8.5,CTA TICI＞1及基线NIHSS≥7；临床/多模式CT评分ROC曲线下面积最大(0.87,95％可信区间0.75～0.95),其预测急性缺血性卒中患者90 d临床功能恢复的效能最高,接下来依次是多模式CT评分、ACTP-SI、VCTP-SI、NIHSS、NCCT及CTA,除临床/多模式CT评分与ACTP-SI (P=0.226)及NIHSS阈值评分(P=0.174)的差异显著性外,其余各参数阈值评分与临床/多模式CT评分的差异均有显著性(P＜0.05).结论 应用临床/多模式CT评分系统比多模式CT及NIHSS各参数单独预测90 d急性缺血性卒中患者的临床功能恢复的效能均高,临床/多模式CT评分系统是预测患者预后的有效评分方法.%Objective To develop a clinical/multimodal computed tomography score(CMCTS) system based on National Institute of Health Stroke Scale(NIHSS) and multimodal CT scores(MCTS), which was used to guide clinical treatment and clinical assessment of functional recovery after 90 days. Methods Multimodal CT examination including non-contrast enhanced CT(NCCT), CT Perfusion(CTP), CT angiography(CTA) were performed in 49 patients with symptoms of stroke less than 9 hours. The Alberta Stroke Program Early CT Score(ASPECTS) were analyzed on NCCT, arterial phase CT perfusion source images(ACTP-SI) and venous phase CTP-SI(VCTP-SI) then the follow up imaging ASPECTS. Thrombolysis in Cerebral Ischemia Scale(TICI) were analyzed on CTA, Baseline NIHSS and 90 days modified Rankin Scale(mRS) were assessed in each patient with the 90 days good clinical functional recovery(mRS<2) as a standard, Application of receiver operating characteristics(ROC) to determine the threshold of NIHSS and CT parameters;in accordance with the threshold score obtain multi-mode CT scoring system, the baseline NIHSSscore join multi-mode CT scoring system obtain clinical/multi-mode CT score. Finally, applicationof ROC curve analysis efficacy of each model predicting clinical outcome.Results The optimal threshold measured on CMCTS, MCTS, NCCT ASPECTS, arterial phaseCTP-SI, venous phase ASPECTS, CTA TICI and NIHSS were>l, >1, >9, >6.5, >8.5, >1 and 3*7,respectively. The parameter that most accurately describes good clinical outcome 3 months afterstroke is the CMCTS(area under the curve is 0.873, 95% confidence interval is 0.75-0.95). Thenext turn is MCTS, ACTP-SI, VCTP-SI, NIHSS, NCCT and CTA, there was a significant statisticalsignificance(P<0.05) for each parameter with CMCTS, except for the ACTP-SI(P=0.226) andNIHSS(/>=0.174).Conclusion The CMCTS based on NIHSS and multimodal CT is superior to NIHSS, MCTS,NCCT, CTA, and CTP in predicting clinical outcome in acute stroke.