Lin Ke, Do Kinh G, Ong Phat, Shapiro Maksim, Babb James S, Siller Keith A, Pramanik Bidyut K
Department of Radiology, New York University Langone Medical Center, New York, NY 10016, USA.
Cerebrovasc Dis. 2009;28(1):72-9. doi: 10.1159/000219300. Epub 2009 May 20.
Conventional noncontrast CT (NCCT) is insensitive to hyperacute cerebral infarction in the first 3 h. Our aim was to determine if CT perfusion (CTP) can improve diagnostic accuracy over NCCT for patients presenting with stroke symptoms in the 3-hour window.
Consecutive patients presenting to our emergency department with symptoms of ischemic stroke <3 h old and receiving NCCT and CTP as part of their triage evaluation were retrospectively reviewed. Patients with follow-up diffusion-weighted MRI (DWI) <7 days from ictus were included. Two readers rated the NCCT and CTP for evidence of acute infarct and its vascular territory. CTP selectively covered 24 mm of brain centered at the basal ganglia with low relative cerebral blood volume in a region of low cerebral blood flow or elevated time to peak as the operational definition for infarction. A third reader rated all follow-up DWI for acute infarct and its vascular territory as the reference standard. Sensitivity, specificity, and predictive values were calculated. An exact McNemar test and generalized estimating equations from a binary logistic regression model were used to assess the difference in detection rates between modalities. A two-sided p value <0.05 was considered significant.
100 patients were included. Sixty-five (65%) patients had follow-up DWI confirmation of acute infarct. NCCT revealed 17 (26.2%) acute infarcts without false positives. CTP revealed 42 (64.6%) acute infarcts with one false positive. Of the 23 infarcts missed on CTP, 10 (43.5%) were outside the volume of coverage while the remaining 13 (56.5%) were small cortical or lacunar type infarcts (<or=15 mm in size). CTP was significantly more sensitive (64.6 vs. 26.2%, p < 0.0001) and accurate (76.0 vs. 52%, p < 0.0001) and had a better negative predictive value (59.6 vs. 42.2%, p = 0.032) than NCCT.
In a retrospective cohort of 100 patients with symptoms of hyperacute stroke in the 3-hour window, CTP provided improved sensitivity and accuracy over NCCT.
传统的非增强CT（NCCT）在超急性脑梗死发病后的最初3小时内敏感度较低。我们的目的是确定对于在3小时时间窗内出现卒中症状的患者，CT灌注成像（CTP）是否能比NCCT提高诊断准确性。
对连续到我院急诊科就诊、发病时间小于3小时且接受了NCCT和CTP作为分诊评估一部分的缺血性卒中患者进行回顾性分析。纳入发病后7天内进行了随访扩散加权磁共振成像（DWI）的患者。两名阅片者对NCCT和CTP进行阅片，判断有无急性梗死及其血管分布区域。CTP以基底节为中心选择性覆盖24mm的脑组织，将脑血流量低或达峰时间延长区域内相对脑血容量低作为梗死的操作定义。第三名阅片者将所有随访DWI作为参考标准，判断有无急性梗死及其血管分布区域。计算敏感度、特异度和预测值。采用精确McNemar检验和二元逻辑回归模型的广义估计方程评估不同检查方法之间检出率的差异。双侧p值<0.05被认为具有统计学意义。
共纳入100例患者。65例（65%）患者随访DWI证实存在急性梗死。NCCT显示17例（26.2%）急性梗死，无假阳性。CTP显示42例（64.6%）急性梗死，1例假阳性。在CTP漏诊的23例梗死中，10例（43.5%）不在覆盖范围内，其余13例（56.5%）为小皮质或腔隙性梗死（大小≤15mm）。CTP的敏感度（64.6%对26.2%，p<0.0001）和准确性（76.0%对52%，p<0.0001）显著高于NCCT，阴性预测值也更好（59.6%对42.2%，p=0.032）。
在一个包含100例发病3小时内超急性卒中症状患者的回顾性队列研究中，CTP比NCCT具有更高的敏感度和准确性。
Zotero 插件