P248 The diagnostic accuracy of Intestinal ultrasound for assessing disease activity in the proximal ileum and jejunum in children with Crohn’s disease: Preliminary results of a prospective study

Vos, H.(1)*;van Wassenaer, E.(1);van Schuppen, J.(2);van Rijn, R.(2);D'Haens, G.(3);Benninga, M.(1);Koot, B.(4);

(1)Amsterdam University Medical Centers, Paediatric Gastroenterology and Nutrition, Amsterdam, The Netherlands;(2)Amsterdam University Medical Centers, Department of Radiology and Nuclear Medicine, Amsterdam, The Netherlands;(3)Amsterdam University Medical Centers, Gastroenterology and Hepatology, Amsterdam, The Netherlands;(4)Amsterdam University Medical Centers- Amsterdam, Paediatric Gastroenterology and Nutrition, Amsterdam, The Netherlands; RAINBOW group

Background

Crohn’s disease (CD) is an inflammatory bowel disease (IBD) involving mainly the small intestine and the colon. Current imaging modalities such as magnetic resonance enterography (MRE) to assess disease activity in the small bowel are expensive and are often poorly tolerated by children. Intestinal ultrasound (IUS) is a non-invasive and well tolerated monitoring tool for assessing disease activity in colon and terminal ileum. In children the diagnostic accuracy of IUS for proximal bowel has not been systematically studied. In this study we assess the diagnostic accuracy of IUS compared to MRE for the proximal ileum and jejunum in the paediatric population. Furthermore we aim to establish which IUS features are the best predictors for IBD activity in the proximal ileum and jejunum.

Methods

Paediatric CD patients were prospectively enrolled and underwent MRE and IUS within 7 days. MRE-images were scored using the segmental grading system by Tielbeek et al. The ultrasonographer was blinded for the MRE results and the paediatric radiologist who assessed the MRE images was blinded for IUS results.

Results

Forty-two paediatric CD patients underwent IUS paired with MRE until now. Eighty-four bowel segments were available for analyses. In total 24 bowel segments had mild (n=12) to moderate severe (n=12) disease activity based on MRE. The AUROC of BWT for moderate severe disease was 0.78 (95% CI 0.61-0.95). A cut-off value of BWT >2.5 mm showed had a specificity of 90%. A cut-off value of BWT <1.7 mm showed had a sensitivity of 91%. Of the affected segments 12 had increased Doppler signal and 13 showed mesenteric fat proliferation. No regression analysis was performed since the number of inflamed segments was too low.

Conclusion

IUS is an promising tool to non-invasively assess disease activity in the proximal ileum and jejunum. Our preliminary results suggest that a cut-off for BWT < 2,5mm IUS can be used to demonstrate disease activity. A lager sample size to perform regression analysis and to assess other IUS features for disease activity is currently being collected.