P277 Comparative accuracy of US versus MRI and colonoscopy in assessing disease activity and complications and influencing the decision-making process in Crohn's disease
Allocca M.*1, Fiorino G.1, Bonifacio C.2, Furfaro F.1, Radice S.1, Gilardi D.1, Peyrin-Biroulet L.3, Danese S.1,4
1Humanitas Research Hospital, IBD Center, Gastroenterology, Rozzano, Milan, Italy 2Humanitas Research Hospital, Radiology, Rozzano, Milan, Italy 3Nancy University Hospital, Université de Lorraine, Gastroenterology and Hepatology, Vandoeuvre-les-Nancy, France 4Humanitas University, Biomedical Sciences, Rozzano, Milan, Italy
Bowel ultrasound (US) and MRI are accurate in assessing disease activity and complications in Crohn's disease (CD) patients. The comparative accuracy of US versus MRI + endoscopy in assessing disease activity and complications and influencing the decision-making process in Crohn's disease is unknown.
Ileo-colonic CD consecutive patients seen in a tertiary referral Center (Humanitas Research Hospital, Milan, Italy) were prospectively assessed by magnetic resonance imaging (MRI), colonoscopy (CS), and US, within 1 week. Sensitivity, specificity, accuracy, positive and negative predictive values (PPV and NPV) of US in assessing localization and extension, bowel wall enhancement (increase of vascularization at color Doppler), bowel wall thickening (>3 mm), strictures (narrowing of the lumen), fistulas and abscesses, and active disease (presence of ulcers at colonoscopy) were calculated using CS in combination with MRI findings as a reference standard. Two independent blinded IBD specialists reviewed separately MRI and US findings, and were asked to decide the therapeutic strategy (continue therapy vs. optimize/change therapy). Kappa agreement between MRI and US was also investigated.
Forty-one consecutive CD patients, irrespectively of disease activity and current therapy, were enrolled. Twenty-five patients had active disease as assessed by MRI and colonoscopy (60.9%), 16/41 (39.1%) had CD-related complications. Sensitivity, specificity, accuracy, PPV and NPV of US are showed in Table 1. Based on US findings alone when compared to MRI and CS, the management of IBD patients (continuing or changing/optimizing therapy) was judged accurate in 85% of patients compared to 75% managed by MRI only (p<0.001). Agreement between MRI and US findings was 80% (p<0.001).
Parameter Sensitivity Specificity Accuracy PPV NPV Localization/Extension 0.80 0.93 0.85 0.95 0.75 Bowel wall enhancement 0.80 0.84 0.82 0.84 0.80 Bowel wall thickening (>3 mm) 0.70 1.0 0.92 1.0 0.91 Strictures 0.76 0.79 0.78 0.72 0.82 Fistulas 1.0 0.97 0.97 0.66 1.0 Abscesses 1.0 0.95 0.95 0.35 1.0 Active disease 0.70 1.0 0.92 1.0 0.91
US was as accurate as the combination CS + MRI in assessing disease activity and complications. Therapeutic decisions based on US findings alone were appropriate in the vast majority of CD patients. US is a non-invasive, easy-to-use tool to manage CD patients in clinical practice.