From the Centre for Medical Imaging, University College London, Charles Bell House, 43-45 Foley Street, 2nd Floor, London W1W 7TS, England (A.M., A.A.P., J.M., D.P., D.A., G.B., S.A.T.) and Department of Histopathology/Research Pathology, University College London Hospitals/University College London Cancer Institute, Rockefeller Building, 21 University Street, London WC1E 6DE, England (M.R.J.); Department of Radiology and Nuclear Medicine, Academic Medical Center, Amsterdam, the Netherlands (C.P., C.E.T.N., J.T., F.V., J.S.); Department of Pathology, UMC Utrecht, Utrecht, the Netherlands (L.A.B.); and Department of Quantitative Imaging, Technical University Delft, Delft, the Netherlands (F.V.).
Radiology. 2018 Nov;289(2):428-435. doi: 10.1148/radiol.2018180100. Epub 2018 Aug 21.
Purpose To evaluate the accuracy of MRI-quantified small bowel motility for Crohn disease activity against endoscopic and histopathologic reference standards. Materials and Methods For this prospective study, 82 participants (median age, 31 years; range, 16 to 70 years; 42 males [median age, 31 years; range, 17 to 70 years] and 40 females [median age, 31 years; range, 16 to 63 years) underwent colonoscopy and MR enterography within 14 days (from October 2011 to March 2014) at two centers. The Crohn disease endoscopic index of severity (CDEIS), histopathologic activity score (endoscopic biopsy acute histologic inflammatory score [EAIS]), and MR index of activity (MaRIA) were scored in the terminal ileum. Terminal ileal motility was quantified by using an image registration based-motility assessment algorithm (hereafter, Motility). Sensitivity and specificity of Motility (˂0.3 arbitrary units) and MaRIA (≥7 and ≥11) for disease activity (CDEIS ≥4 or EAIS ≥1) were compared by using the McNemar test. Receiver operating characteristic curves were constructed and areas under the curve were compared. Motility was correlated with reference standards by using Spearman rank estimates. Results Terminal ileal Motility was negatively correlated with EAIS (r =-0.61; 95% confidence interval [CI]: 0.7, -0.5) and CDEIS (r = -0.59; 95% CI: 0.7, -0.4). With CDEIS as the standard of reference, Motility had higher sensitivity than did MaRIA (≥11) (93% vs 78%, respectively; P = .03), but lower specificity (61% vs 81%, respectively; P = .04). With EAIS as the standard of reference, Motility had higher sensitivity than did MaRIA (≥7) (92% vs 75%, respectively; P = .03) but similar specificity (71% vs 74%, respectively; P >.99). The area under the receiver operating characteristic curve for Motility was 0.86 and 0.87 with CDEIS and EAIS as the standard of reference, respectively. Conclusion The terminal ileal Motility score showed good agreement with endoscopic and histopathologic activity in Crohn disease. © RSNA, 2018 Online supplemental material is available for this article.
目的 评估 MRI 定量小肠运动对克罗恩病活动的准确性,将其与内镜和组织病理学参考标准相对比。
材料与方法 这是一项前瞻性研究,共纳入 82 名参与者(中位年龄,31 岁;范围,16 岁至 70 岁;42 名男性[中位年龄,31 岁;范围,17 岁至 70 岁]和 40 名女性[中位年龄,31 岁;范围,16 岁至 63 岁]),分别于 2011 年 10 月至 2014 年 3 月在两个中心接受结肠镜检查和磁共振肠造影术,时间间隔不超过 14 天。在回肠末端评估克罗恩病内镜严重程度指数(CDEIS)、组织病理学活动评分(内镜活检急性组织学炎症评分[EAIS])和活动性磁共振指数(MaRIA)。采用基于图像配准的运动评估算法(以下简称“运动”)对回肠末端运动进行定量。使用 McNemar 检验比较运动(<0.3 个单位)和 MaRIA(≥7 和≥11)对疾病活动(CDEIS≥4 或 EAIS≥1)的敏感性和特异性。构建受试者工作特征曲线并比较曲线下面积。通过 Spearman 秩次估计值将运动与参考标准相关联。
结果 回肠末端运动与 EAIS(r=-0.61;95%置信区间[CI]:0.7,-0.5)和 CDEIS(r=-0.59;95%CI:0.7,-0.4)呈负相关。以 CDEIS 为参考标准,运动的敏感性高于 MaRIA(≥11)(分别为 93%和 78%;P=.03),但特异性较低(分别为 61%和 81%;P=.04)。以 EAIS 为参考标准,运动的敏感性高于 MaRIA(≥7)(分别为 92%和 75%;P=.03),但特异性相似(分别为 71%和 74%;P>.99)。以 CDEIS 和 EAIS 为参考标准,运动的受试者工作特征曲线下面积分别为 0.86 和 0.87。
结论 回肠末端运动评分与克罗恩病的内镜和组织病理学活动具有良好的一致性。
©2018 美国放射学会