MR Enterography for the Evaluation of Small-Bowel Inflammation in Crohn Disease by Using Diffusion-weighted Imaging without Intravenous Contrast Material: A Prospective Noninferiority Study

作者：Seo Nieun; Park Seong Ho; Kim Kyung Jo; Kang Bo Kyeong; Lee Yedaun; Yang Suk Kyun; Ye Byong Duk; Park Sang Hyoung; Kim So Yeon; Baek Seunghee; Han Kyunghwa; Ha Hyun Kwon

来源：Radiology, 2016, 278(3): 762-772.

DOI：10.1148/radiol.2015150809

Purpose: To determine whether magnetic resonance (MR) enterography performed with diffusion-weighted imaging (DWI) without intravenous contrast material is noninferior to contrast material-enhanced (CE) MR enterography for the evaluation of small-bowel inflammation in Crohn disease.
Materials and Methods: Institutional review board approval and informed consent were obtained for this prospective noninferiority study. Fifty consecutive adults suspected of having Crohn disease underwent clinical assessment, MR enterography, and ileocolonoscopy within 1 week. MR enterography included conventional imaging and DWI (b = 900 sec/mm(2)). In 44 patients with Crohn disease, 171 small-bowel segments that were generally well distended and showed a wide range of findings, from normalcy to severe inflammation (34 men, 10 women; mean age +/- standard deviation, 26.9 years +/- 6.1), were selected for analysis. Image sets consisting of (a) T2-weighted sequences with DWI and (b) T2-weighted sequences with CE T1-weighted sequences were reviewed by using a crossover design with blinding and randomization. Statistical analyses included noninferiority testing regarding proportional agreement between DWI and CE MR enterography for the identification of bowel inflammation with a noninferiority margin of 80%, correlation between DWI and CE MR enterography scores of bowel inflammation severity, and comparison of accuracy between DWI and CE MR enterography for the diagnosis of terminal ileal inflammation by using endoscopic findings as the reference standard.
Results: The agreement between DWI and CE MR enterography for the identification of bowel inflammation was 91.8% (157 of 171 segments; one-sided 95% confidence interval:. >= 88.4%). The correlation coefficient between DWI and CE MR enterography scores was 0.937 (P < .001). DWI and CE MR enterography did not differ significantly regarding the sensitivity and specificity for the diagnosis of terminal ileal inflammation (P > .999). DWI and CE MR enterography concurred in the diagnosis of penetrating complications in five of eight segments.
Conclusion: DWI MR enterography was noninferior to CE MR enterography for the evaluation of inflammation in Crohn disease in generally well-distended small bowel, except for the diagnosis of penetration.