PWE-107 Accuracy of Magnetic Resonance Enterograpgy in Predicting Anastomotic Stenosis in Recurrent Crohn’s Disease

    loading  Checking for direct PDF access through Ovid



Up to 80% of patient with Crohn’s disease (CD) undergo intestinal resection, commonly an ileocaecal resection. Eighty percent of patients develop endoscopic recurrence at the anastamotic site at 1 year and 50% develop clinical recurrence at 3 years. The severity of endoscopic recurrence varies from no endoscopic lesions to anastamotic stenosis and/or diffuse inflammation. Direct endoscopic visualisation is recommended to detect recurrence but it is invasive. Magnetic resonance enterography (MRE) has become a standard imaging investigation for CD but only few studies have evaluated its utility in recurrent anastamotic stenosis. Accurate characterisation of recurrence grade is critical as strictures up to 5cms can be successfully treated with endoscopic balloon dilatation. We evaluated the utility of MRE in the assessment of anastamotic stenosis in recurrent CD.


This retrospective study was done at the Royal Liverpool Hospital and included all CD patients who underwent endoscopic balloon dilatation for anastomotic stenosis between 2009–2013. Patients who had an MRE done within 6 months of the endsocopic procedure were eligible for inclusion. MRE was done following administration of oral polyethylene glycol solution and sequences were analysed for the presence of stenosis, length of stenosis, pre-stenotic dilation and the presence of enhancement by an experienced gastrointestinal radiologist. The length and presence of stenosis was extracted from the endoscopy report and correlated against the MRE findings.


16 patients were included in the study (5 male, 11 female). There was good agreement between endoscopy and MRE for the presence of anastamotic stenosis. Using endoscopy as the gold standard, the sensitivity and specificity of MRE in detecting anastomotic stenosis was 86% and 100% respectively. A significant variation was noted in the length of stenosis as assessed by MRE (45 ± 12 mm, mean ± SE) and endoscopy (20 ± 3 mm, mean ± SE), two sided p < 0.05. All patients underwent successful endoscopic dilatation.


MRE is an accurate tool for predicting the presence of anastomotic stenosis in recurrent CD. However, the discrepancy observed in the length of stenosis between the two modalities may mean suitable patients for endoscopic dilatation are missed. Therefore, endoscopy and MRE should be used as complementary tools in the assessment of anastamotic stenosis.

Disclosure of Interest

None Declared.

Related Topics

    loading  Loading Related Articles