DOP039. Development of MRE based multi item measures of inflammation and intestinal damage in paediatric Crohn's disease: the ImageKids study
P. Church1, M.-L. Greer2, A. Griffiths1, M.M. Amitai3, T. Walters1, R. Cytter-Kuint2, G. Focht4, D. Turner5, 1University of Toronto, The Hospital for Sick Children, Division of Gastroenterology, Hepatology and Nutrition, Toronto, Canada, 2The Hospital for Sick Children/University of Toronto, Diagnostic Imaging, Toronto, Canada, 3Sheba Medical Center, Department of Diagnostic Imaging, Tel Hashomer, Israel, 4Shaare Zedek Medical Centre, Pediatric Gastroenterology and Nutrition, Jerusalem, Israel, 5Shaare Zedek Medical Center, The Hebrew University of Jerusalem, Pediatric Gastroenterology, Jerusalem, Israel
Treatment goals in paediatric Crohn's disease (CD) have moved beyond symptom control to intestinal healing. Magnetic resonance enterography (MRE) has become the modality of choice for sequentially imaging small bowel, but description and interpretation of MRE findings must be standardized so that impact of therapies can be assessed.
We undertook to develop 2 multi-item MRE-based measures in children with CD; one of inflammation, (Pediatric MRE-based Crohn's Activity Index, P-MECAI) and one of structural change, (Pediatric Crohn's Disease Intestinal Damage, PECDID) through a rigorous process of item generation and reduction. First, comprehensive lists of MRE parameters reflecting inflammation and structural change were prepared by systematically reviewing the literature. In an iterative process over email, a Delphi group of 30 expert radiologists contributed items to the lists which were reviewed and consolidated by the steering committee. Items were scored and rank-ordered by the group according to importance and frequency. The highest ranked items were retained and judgmentally formatted in a face to face meeting.
The literature review yielded 80 relevant manuscripts which contributed 35 items; the Delphi process added a further 15 items totaling 33 items reflecting inflammation and 17 reflecting damage. Item reduction trimmed this to 6 items reflecting inflammation and 7 reflecting structural change. Some variables were common to inflammation and damage (Table 1), while others were specific (Table 2).
Motility (normal/impaired), wall enhancement pattern (mucosal, striated, transmural) and fibrofatty proliferation (present/absent) will be considered separately by data driven analysis of the ongoing prospective cohorts.
|Wall T2 hyperintensity||None, Mild, Marked|
|Wall enhancement||None, Mild, Marked|
|Wall restricted diffusion||None, Mild, Marked|
|Lumen calibre||Normal, Narrow, Narrow with prestenotic dilation|
|Comb sign||None, Mild, Marked|
|Penetrating lesions||None, Sinus/phlegmon, Fistula/abscess|
|History of surgical intervention||None, Endoscopic dilation, Stricturoplasty, Resection|
The formatted indices are being prospectively validated in the ongoing multi-centre international ImageKids study. These measures could be used to assess the effect of therapies on the control of inflammation and the progression of bowel damage in pediatric Crohn's disease.