Prince M R, Narasimham D L, Jacoby W T, Williams D M, Cho K J, Marx M V, Deeb G M
Department of Radiology, University of Michigan Hospital, Ann Arbor 48109-0030, USA.
AJR Am J Roentgenol. 1996 Jun;166(6):1387-97. doi: 10.2214/ajr.166.6.8633452.
Our objective was to evaluate image quality and preliminary clinical experience with three-dimensional gadolinium-enhanced MR angiography of the thoracic aorta.
Ninety patients with suspected thoracic aorta pathology underwent 97 MR examinations at 1.5 T with a 4-min, three-dimensional spoiled gradient-echo techniques. Gadolinium infusion was timed for maximum arterial contrast during acquisition of the central portion of K-space. No ECG gating or breath-holding was used. All MR examinations were evaluated retrospectively for intravascular signal-to-noise ratio (SNR). In 30 of the 90 patients, results from surgery (n = 11), angiography (n = 12), or both (n = 7) were available. Four radiologists who were unaware of the angiographic or surgical findings assessed each of these 30 examinations for three types of pathology: dissection, coarctation, or aneurysm. The observers also assessed aortic branch vessel patency and vascular anomalies in the 19 patients who had angiographic correlation.
Image quality (determined as SNR) was highest in the aortic arch, upper descending thoracic aorta, and upper abdominal aorta. We saw a small reduction in the SNR in the ascending aorta and lower descending thoracic aorta (p < .0001), attributable to cardiac and respiratory motion. Image quality was not affected by slow flow. MR imaging correctly diagnosed pathology in all 30 patients with angiographic or surgical correlation, including eight dissections, three coarctations, and 10 aneurysms. The type of the dissection was correctly determined in all eight patients. Stenoses of major branch vessel origins were detected with a sensitivity of 90% (95% bayesian confidence interval, 99-63%) and a specificity of 96% (95% bayesian confidence interval, 99-89%) in the 19 patients with angiographic correlation. Five vascular anomalies, including an aberrant right subclavian artery, a bovine arch, and three accessory renal arteries, were correctly identified.
Three-dimensional gadolinium-enhanced MR angiography has the potential to accurately diagnose aortic dissection, coarctation, and aneurysm. It does not require ECG gating or breath-holding and thereby extends the diagnostic utility of MR imaging for the thoracic aorta.
我们的目的是评估三维钆增强磁共振血管造影对胸主动脉成像的质量及初步临床经验。
90例怀疑有胸主动脉病变的患者在1.5T磁共振成像设备上接受了97次检查,采用4分钟的三维扰相梯度回波技术。在采集K空间中心部分时,钆剂注射时间被设定为获得最大动脉期对比。未使用心电图门控或屏气技术。所有磁共振检查均回顾性评估血管内信噪比(SNR)。90例患者中的30例有手术(n = 11)、血管造影(n = 12)或两者(n = 7)的结果可供参考。4名不知血管造影或手术结果的放射科医生对这30例检查中的三种病变类型进行评估:夹层、缩窄或动脉瘤。观察者还评估了19例有血管造影对照患者的主动脉分支血管通畅情况及血管异常。
图像质量(以信噪比衡量)在主动脉弓、胸主动脉降部上段及腹主动脉上段最高。升主动脉和胸主动脉降部下段的信噪比有小幅降低(p <.0001),这归因于心脏和呼吸运动。图像质量不受缓慢血流影响。磁共振成像正确诊断了所有30例有血管造影或手术对照的患者的病变,包括8例夹层、3例缩窄和10例动脉瘤。所有8例夹层患者的夹层类型均被正确判定。在19例有血管造影对照的患者中,主要分支血管起始处狭窄的检测灵敏度为90%(95%贝叶斯置信区间,99 - 63%),特异度为96%(95%贝叶斯置信区间,99 - 89%)。正确识别了5种血管异常,包括迷走右锁骨下动脉、牛型主动脉弓和3条副肾动脉。
三维钆增强磁共振血管造影有潜力准确诊断主动脉夹层、缩窄和动脉瘤。它不需要心电图门控或屏气,从而扩展了磁共振成像对胸主动脉的诊断效用。