From the Departments of Radiology (M.J.P., K.H., C.A., H.J.K., H.K., M.S.P.) and Biostatistics (K.H., C.M.N.), Yonsei University College of Medicine, Severance Hospital, 50 Yonsei-Ro, Seodaemun-Gu, Seoul 120-752, Korea; Department of Radiology, Ulsan University College of Medicine, Asan Medical Center, Seoul, Korea (N.H., S.H.P.); Center for Colorectal Cancer, National Cancer Center, Ilsan, Korea (M.J.K.); Department of Radiology, Seoul National University College of Medicine, Bundang Hospital, Bundang, Korea (Y.J.L.); Department of Radiology, Korea University College of Medicine, Kuro Hospital, Seoul, Korea (Y.S.P., C.H.L.); Department of Radiology, Catholic University College of Medicine, Seoul St. Mary's Hospital, Seoul, Korea (S.E.R.); Department of Radiology, National Health Insurance Service Ilsan Hospital, Ilsan, Korea (S.P.); and Department of Radiology, Sungkyunkwan University College of Medicine, Samsung Medical Center, Seoul, Korea (W.J.L.).
Radiology. 2017 Aug;284(2):423-431. doi: 10.1148/radiol.2017161619. Epub 2017 Mar 22.
Purpose To compare the diagnostic performances of contrast agent-enhanced computed tomography (CT) and gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid-enhanced liver magnetic resonance (MR) imaging (referred to as EOB MR imaging) in the evaluation of disappearing colorectal liver metastases (CRLMs) after chemotherapy. Materials and Methods The eight institutional review boards approved this retrospective study and waived the requirement for informed consent. On the basis of retrospective searches in eight hospitals, 87 patients with 393 CRLMs, each patient with one or more CRLM that later disappeared on contrast-enhanced CT scans after chemotherapy, and subsequently underwent surgery for the CRLMs, were enrolled. The anonymized imaging data and case report forms were sent to the central review system and independently reviewed by four radiologists. All anonymized data were randomly allocated into two groups (groups A and B), which were read by two independent readers. True absence of tumor was defined as pathologic absence of tumor for resected lesions and no in situ recurrence within 1 year after surgery for lesions left unresected at each 3-month follow-up contrast-enhanced CT. Positive predictive values for absence of tumor and for residual tumor on contrast-enhanced CT and EOB MR images were compared by using a generalized estimating equation. Results Among 393 CRLMs, the positive predictive value for absence of tumor on EOB MR images (78.0%; 95% confidence interval [CI]: 63.68%, 87.74%) was significantly higher than that on contrast-enhanced CT scans (35.2%; 95% CI: 25.11%, 46.79%; P < .001). The positive predictive value for residual tumor on CT scans (86.0%; 95% CI: 78.61%, 91.16%) was higher than that on EOB MR images (83.8%; 95% CI: 77.50%, 88.67%) without statistical significance (P = .330). Conclusion EOB MR imaging was superior to contrast-enhanced CT imaging for assessment of disappearing CRLMs after chemotherapy. RSNA, 2017 Online supplemental material is available for this article.
目的 比较对比剂增强 CT（CECT）与钆塞酸二乙三胺五乙酸增强肝脏磁共振成像（EOB-MRI）在评估化疗后消失的结直肠癌肝转移瘤（CRLM）中的诊断效能。
材料与方法 8 家机构审查委员会批准了这项回顾性研究，豁免了知情同意书的要求。基于 8 家医院的回顾性检索，共纳入 87 例患者的 393 个 CRLM，每位患者的一个或多个 CRLM 在化疗后 CECT 扫描中消失，随后因这些 CRLM 而行手术治疗。将匿名的影像学数据和病例报告表发送至中央审查系统，由 4 位放射科医生独立进行审查。所有匿名数据被随机分为两组（A 组和 B 组），由两位独立的读者进行阅读。切除病变的病理无肿瘤定义为术后 1 年内无肿瘤原位复发；对于未切除病变，每 3 个月进行一次 CECT 随访，定义为无肿瘤为无肿瘤残留。采用广义估计方程比较 CECT 和 EOB-MRI 图像上无肿瘤和肿瘤残留的阳性预测值。
结果 在 393 个 CRLM 中，EOB-MRI 图像上无肿瘤的阳性预测值（78.0%；95%置信区间：63.68%，87.74%）显著高于 CECT 扫描（35.2%；95%置信区间：25.11%，46.79%；P<0.001）。CT 扫描上肿瘤残留的阳性预测值（86.0%；95%置信区间：78.61%，91.16%）高于 EOB-MRI（83.8%；95%置信区间：77.50%，88.67%），但差异无统计学意义（P=0.330）。
结论 在评估化疗后消失的 CRLM 方面，EOB-MRI 优于 CECT 成像。
放射学会，2017
在线补充材料本文相关。
