Search in the Abstract Database

Search Abstracts 2012

* = Presenting author

P188. A systematic review of MR enterography signs of inflammation and damage in Crohn's disease


P. Church1, D. Turner2, T. Walters3, M.‑L. Greer4, M. Amitai5, A. Griffiths1

1Hospital for Sick Children, Paediatrics, Toronto, Canada; 2Shaare Zedak Medical Center, Pediatric Gastroenterology Unit, Jerusalem, Israel; 3Hospital for Sick Children, Div GI, Hepatology and Nutrition, Toronto, Canada; 4Hospital for Sick Children, Radiology, Toronto, Canada; 5Sheba Medical Center, Radiology, Tel Aviv, Israel



Background: In the treatment of Crohn's disease (CD), mucosal healing has become a major goal, with the hope of avoiding intestinal damage from chronic inflammation. MRE (magnetic resonance enterography) has emerged as a non-invasive means of monitoring inflammation and damage. This is especially important in paediatrics, as repeated endoscopy is poorly tolerated. As part of the item-generation phase in the development of MR-based multi-item measures of inflammation and damage for use in paediatric studies, we carried out a systematic review to identify MR variables used to describe these 2 distinct concepts.

Methods: 692 studies of MRI and CD were identified. Studies not using a clinical, surgical, endoscopic or pathologic reference standard of inflammation or damage were excluded. 142 studies were retrieved for full review, of which 80 were included (80 examining inflammation and 18 examining damage). 71 studies were adult only, 6 were pediatric only and 3 included both adults and children. Pooled weighted sensitivity and specificity were calculated including area under the ROC curve, with a per-patient followed by per-segment analysis.

Results: A total of 22 MRI variables were used to reflect inflammation, of which bowel wall enhancement (69 studies) and thickness (63 studies) were most frequently reported. Of 9 MRI variables used to reflect intestinal damage, the most common were presence of abscess (8 studies) and stricture (9 studies). Tables 1 and 2 summarize the variables with the greatest sensitivity and specificity for inflammatory activity and for damage.

Table 1: MRE Signs of Inflammatory Activity
Sensitivity (%)Specificity (%)
Per-patient
Enhancement kinetics (100)Abscess (100)
Lymph node enhancement (100)Wall signal intensity (98)
Per-segment
Abnormal motility (85)Wall edema (100)
Diffusion-weighted hyperintensity (82)Abscess (100)
 Fistula (100)
Table 2: MRE Signs of Intestinal Damage
Sensitivity (%)Specificity (%)
Per-patient
Mesenteric signs (87)Abscess (99)
Abscess (80)Fistula (97)
Per-segment
Stricture (83)Stricture (100)
Fistula (71)Fistula (98)

Conclusions: Identifying the best MRI variables to reflect inflammation and damage will maximize the utility of this rapidly emerging technique and is the first stage of constructing MR-based indices for evaluating mucosal healing and intestinal damage.