Hosch W P, Schmidt S M, Plaza S, Dechow C, Schmidt J, Ley S, Kauffmann G W, Hansmann J
Department of Radiology, University Clinic Heidelberg, INF 110, Heidelberg 69120, Germany.
AJR Am J Roentgenol. 2006 Jun;186(6):1502-11. doi: 10.2214/AJR.05.0879.
This study compared MR during arterial portography (MRAP) with CT during arterial portography (CTAP) with regard to the detection and differentiation of liver metastases before surgery.
Fifteen patients with liver metastases were enrolled before surgery according to the guidelines of our institutional review board and good clinical practice. After mesentericography, unenhanced scans (Volume Zoom) were performed initially. For CTAP, the contrast medium was injected through the superior mesenteric artery. Images were acquired in portal and delayed enhancement. The MR protocol (1.5 T; Magnetom Symphony) started with T1-weighted fast low-angle shot (FLASH) T2-weighted turbo spin echo (TSE). MRAP followed with gadolinium-enhanced dynamic T1-weighted 3D FLASH. Delayed-phase T1-weighted 2D FLASH axial images were performed 2 min after IV injection of the contrast medium. Qualitative and quantitative evaluation of CTAP and MRAP was performed by three blinded radiologists regarding the number of lesions and their size, localization, and differential diagnosis.
The overall sensitivity in detecting liver metastases was 97% with MRAP and 93% with CTAP (p > 0.05, not significant [n.s.]). The specificity was calculated to be 97% for MRAP and 82% for CTAP (p < 0.0001, statistically significant [s.s.]). The differences in sensitivity were more accentuated if only lesions 10 mm or smaller were considered (95% vs 88%, p > 0.05, n.s.), for which the respective specificities were 95% and 80% (p < 0.0014, s.s.). Improvements in sensitivity and specificity were associated with a higher lesion-to-liver contrast-to-noise ratio (59.4 +/- 51.0 for MRAP vs 10.4 +/- 7.3 for CTAP) and resulted in higher diagnostic confidence in the differential diagnosis of liver lesions (p < 0.001, s.s.) and better interobserver agreement (median kappa value, 0.88 vs 0.63).
MRAP proved to be a reliable method in the preoperative detection of small liver metastases in particular, with a higher sensitivity and specificity than CTAP. If organizational difficulties of MRAP can be overcome, MRAP could be considered instead of CTAP in the preoperative invasive evaluation of metastatic liver disease.
本研究比较了动脉门静脉造影磁共振成像（MRAP）与动脉门静脉造影计算机断层扫描（CTAP）在术前检测和鉴别肝转移瘤方面的效果。
根据我们机构审查委员会的指南和良好临床实践，在术前纳入了15例肝转移瘤患者。肠系膜造影后，首先进行非增强扫描（容积变焦）。对于CTAP，通过肠系膜上动脉注射造影剂。在门静脉期和延迟期增强时采集图像。MR协议（1.5T；Magnetom Symphony）以T1加权快速低角度激发（FLASH）T2加权涡轮自旋回波（TSE）开始。随后进行钆增强动态T1加权3D FLASH的MRAP。在静脉注射造影剂2分钟后进行延迟期T1加权2D FLASH轴位图像。由三位不知情的放射科医生对CTAP和MRAP进行定性和定量评估，涉及病变数量及其大小、位置和鉴别诊断。
MRAP检测肝转移瘤的总体敏感性为97%，CTAP为93%（p>0.05，无显著性差异[n.s.]）。MRAP的特异性计算为97%，CTAP为82%（p<0.0001，有统计学显著性差异[s.s.]）。如果仅考虑10mm或更小的病变，敏感性差异更为明显（95%对88%，p>0.05，无显著性差异），其各自的特异性分别为95%和80%（p<0.0014，有统计学显著性差异）。敏感性和特异性的提高与更高的病变与肝脏对比噪声比相关（MRAP为59.4±51.0，CTAP为10.4±7.3），并导致对肝病变鉴别诊断的更高诊断信心（p<0.001，有统计学显著性差异）和更好的观察者间一致性（中位数kappa值，0.88对0.63）。
MRAP被证明是一种可靠的方法，尤其在术前检测小肝转移瘤方面，其敏感性和特异性高于CTAP。如果能够克服MRAP的组织困难，在转移性肝病的术前侵入性评估中，可以考虑用MRAP代替CTAP。
