Getting the steak without the sizzle: is MR enterography as good as CT enterography?

S everal recent studies have identified an accelerating rate of radiation exposure in patients with poorly controlled complicated small bowel Crohn’s disease (CD), largely due to abdominal and pelvic computed tomography (CT) scans performed in local emergency departments with potentially long-term clinical consequences. This has motivated active research into alternatives for disease activity assessment, including magnetic resonance enterography (MRE). Dr. Siddiki et al designed this study to assess whether MRE is as accurate as CT enterography (CTE) in assessing disease activity. They performed clinical evaluation, ileocolonoscopy, CTE, and MRE in 30 patients (3 were unable to complete all 3 studies). The radiologist readers were blinded to clinical information when reading each scan. The interobserver agreement was substantial for both MRE (0.63, 95% confidence interval [CI] 0.31–0.92) and for CTE (0.76, 95% CI 0.5– 1.0). MRE had a sensitivity of 90.5% (95% CI 70–99) and specificity of 66.7% (95% CI 30–93) and CTE had a sensitivity of 95.2% (95% CI 76–100) and specificity of 88.9% (95% CI 52–100) for detecting small bowel disease activity. Twenty-three (77%) patients underwent MRE the same day as the CTE. The remainder underwent MRE within 21 days of CTE. The endoscopy was performed without input from the study team, and therefore no endoscopic disease severity scoring was performed, and biopsies were not consistently obtained. MRE and CTE scans identified disease activity in 8 patients (24%) with normal endoscopy, and in an additional 3 patients who did not have ileal intubation on colonoscopy. A composite ‘‘clinical reference standard’’ of active, inactive, or absent small bowel CD was constructed based on the impression of the treating clinician, the findings on endoscopy, biopsies (if taken), and readings of the CTE and MRE (after they had already been scored). Twenty-two patients (67%) had active disease, 2 (6%) had inactive disease, and 9 (27%) had no evidence of CD. MRE and CTE had similar sensitivities for detecting active small bowel inflammation.