Ramasundara S, Mitchell P J, Dowling R J
Department of Radiology, Royal Melbourne Hospital, Parkville, Victoria, Australia.
J Med Imaging Radiat Oncol. 2010 Dec;54(6):526-33. doi: 10.1111/j.1754-9485.2010.02211.x.
The aim of the study was to retrospectively analyse the accuracy of CT angiography (CTA) of the Circle of Willis and a prototype bone subtraction CT angiogram (BSCTA) compared with digital subtraction angiography (DSA) in the detection of intracranial aneurysms.
A total of 36 patients with suspected subarachnoid haemorrhage who had CTA scans that had matching DSA studies between November 2005 and December 2006 were retrospectively reviewed by two experienced neurointerventionalist. Three-dimensional volume-rendered (3D VR) CTA and BSCTA were reviewed, followed by review of the source data in multi-planar reformats (MPRs). The results were then assessed for sensitivity and specificity compared with the gold standard DSA. Time taken to review the studies, visibility of the ophthalmic, posterior communicating arteries, superior cerebella artery, posterior cerebral artery on the 3D VR BSCTA and CTA were analysed. BSCTA was assessed for residual bone and artificial stenosis. Further, the aneurysms were analysed for suitability of endovascular treatment versus surgical treatment based on CTA and BSCTA with MPR and 3D VR.
CTA 3D VR sensitivity and specificity was 94 and 80%, respectively. CTA MPR sensitivity and specificity was 100% and 90%, respectively. When both CTA 3D VR and MPR images were combined the sensitivity and specificity was 100 and 90%, respectively. BSCTA 3D VR sensitivity and specificity was 91 and 90%, respectively, and the MPR sensitivity and specificity was 97 and 90%, respectively. When both BSCTA 3D VR and MPR images were combined the sensitivity was 97% with a specificity of 90% Overall, an 84% correlation was found between both CTA and BSCTA when compared with DSA for assessing for suitability of endovascular treatment versus surgical treatment. A statistically significant difference was shown between the time taken to review the CTA and BSCTA (P < 0.001 confidence interval 3.1-3.9 min).
In this study there was no significant difference between the diagnostic ability of BSCTA compared with CTA. Importantly, both CTA and BSCTA had high sensitivities and specificities compared to DSA.
本研究的目的是回顾性分析 Willis 环的 CT 血管造影(CTA)及一种原型骨减影 CT 血管造影(BSCTA)与数字减影血管造影(DSA)相比在检测颅内动脉瘤方面的准确性。
对 2005 年 11 月至 2006 年 12 月期间 36 例疑似蛛网膜下腔出血且 CTA 扫描与 DSA 检查匹配的患者进行回顾性研究,由两名经验丰富的神经介入专家进行评估。先对三维容积再现(3D VR)CTA 和 BSCTA 进行评估,随后对多平面重组(MPR)的源数据进行评估。然后将结果与金标准 DSA 进行比较,评估其敏感性和特异性。分析评估各项检查所需时间,以及 3D VR BSCTA 和 CTA 上眼动脉、后交通动脉、小脑上动脉、大脑后动脉的显示情况。对 BSCTA 的残留骨和人工狭窄情况进行评估。此外,根据 CTA 和 BSCTA 以及 MPR 和 3D VR 分析动脉瘤适合血管内治疗还是手术治疗的情况。
CTA 3D VR 的敏感性和特异性分别为 94%和 80%。CTA MPR 的敏感性和特异性分别为 100%和 90%。当将 CTA 3D VR 和 MPR 图像结合时,敏感性和特异性分别为 100%和 90%。BSCTA 3D VR 的敏感性和特异性分别为 91%和 90%,MPR 的敏感性和特异性分别为 97%和 90%。当将 BSCTA 3D VR 和 MPR 图像结合时,敏感性为 97%,特异性为 90%。总体而言,在评估血管内治疗与手术治疗的适用性时,CTA 和 BSCTA 与 DSA 相比的相关性为 84%。评估 CTA 和 BSCTA 所需时间存在统计学显著差异(P < 0.001,置信区间 3.1 - 3.9 分钟)。
在本研究中,BSCTA 与 CTA 的诊断能力无显著差异。重要的是,与 DSA 相比,CTA 和 BSCTA 均具有较高的敏感性和特异性。