Department of Radiology and Radiological Science, Division of Cardiovascular Imaging, Medical University of South Carolina, 25 Courtenay Drive, Charleston, SC 29425; Department of Diagnostic and Interventional Radiology, Division of Experimental and Translational Imaging, University Hospital Frankfurt, Frankfurt, Germany.
Department of Radiology and Radiological Science, Division of Cardiovascular Imaging, Medical University of South Carolina, 25 Courtenay Drive, Charleston, SC 29425.
Acad Radiol. 2019 Oct;26(10):1309-1317. doi: 10.1016/j.acra.2018.12.010. Epub 2019 Jan 14.
To evaluate the diagnostic accuracy of a prototype noncontrast, free-breathing, self-navigated 3D (SN3D) MR angiography (MRA) technique for the assessment of coronary artery anatomy in children with known or suspected coronary anomalies, using CT angiography (CTA) as the reference standard.
Twenty-one children (15 male, 12.3 ± 2.6 years) were prospectively enrolled between July 2014 and August 2016 in this IRB-approved, HIPAA-compliant study. Patients underwent same-day unenhanced SN3D-MRA and contrast-enhanced CTA. Two observers rated the visualization of coronary artery segments and diagnostic confidence on a 3-point scale and assessed coronary arteries for anomalous origin, as well as interarterial and intramural course. Sensitivity, specificity, positive (PPV) and negative predictive values (NPV) of SN3D-MRA for the detection of coronary artery abnormalities were calculated. Interobserver agreement was assessed using Intraclass Correlation Coefficients (ICC).
Fourteen children showed coronary artery abnormalities on CTA. The visualization of coronary segments was rated significantly higher for CTA compared to MRA (p <0.015), except for the left main coronary artery (p = 0.301), with good to excellent interobserver agreement (ICC = 0.62-0.94). Diagnostic confidence was higher for CTA (p = 0.046). Sensitivity, specificity, PPV, and NPV of MRA were 92%, 92%, 96%, and 87% for the detection of coronary artery anomalies, 85%, 85%, 74%, and 92% for high origin, 71%, 92%, 82%, and 87% for interarterial, and 41%, 96%, 87%, and 80% for intramural course.
Noncontrast SN3D-MRA is highly accurate for the detection of coronary artery anomalies in pediatric patients while diagnostic confidence and coronary artery visualization remain superior with CTA.
评估一种新型非对比剂、自由呼吸、自我导航 3D（SN3D）磁共振血管成像（MRA）技术在评估已知或疑似冠状动脉异常儿童的冠状动脉解剖结构中的诊断准确性，以 CT 血管造影（CTA）为参考标准。
2014 年 7 月至 2016 年 8 月期间，前瞻性纳入 21 名儿童（男 15 名，12.3±2.6 岁），本研究经机构审查委员会批准，符合 HIPAA 规定。患者行同日非增强 SN3D-MRA 和增强 CTA。两名观察者对冠状动脉节段的显示和诊断信心进行 3 分制评分，并评估冠状动脉的异常起源，以及动脉间和壁内走行。计算 SN3D-MRA 检测冠状动脉异常的敏感性、特异性、阳性（PPV）和阴性预测值（NPV）。采用组内相关系数（ICC）评估观察者间的一致性。
CTA 显示 14 名儿童存在冠状动脉异常。CTA 显示冠状动脉节段的能力明显优于 MRA（p<0.015），左主干冠状动脉除外（p=0.301），观察者间一致性良好至极好（ICC=0.62-0.94）。CTA 的诊断信心更高（p=0.046）。MRA 检测冠状动脉异常的敏感性、特异性、PPV 和 NPV 分别为 92%、92%、96%和 87%，高起源的敏感性、特异性、PPV 和 NPV 分别为 85%、85%、74%和 92%，动脉间的敏感性、特异性、PPV 和 NPV 分别为 71%、92%、82%和 87%，壁内的敏感性、特异性、PPV 和 NPV 分别为 41%、96%、87%和 80%。
非对比剂 SN3D-MRA 对儿童冠状动脉异常的检测具有高度准确性，而 CTA 则具有更高的诊断信心和冠状动脉显示能力。
