RATIONALE AND OBJECTIVES: The purpose of this study was to determine the accuracy of magnetic resonance enterography (MRE) compared to histopathology in the evaluation of pediatric inflammatory bowel disease and to assess interreader reliability for image interpretation. MATERIALS AND METHODS: All magnetic resonance enterography studies performed for known or suspected inflammatory bowel disease between July 2009 and July 2010 were retrospectively reviewed by two pediatric radiologists. Exams were evaluated for signs of enteric inflammation and extraenteric disease. A five-point, Likert-type scale was used to assess the overall likelihood of active inflammation, with scores ≥ 3 considered positive. Cohen's κ coefficient was calculated to assess interreader agreement. A subset of patients who had undergone ileocolonoscopy or surgery with confirmed histopathology within 45 days of MRE were used to assess the accuracy of MRE for detecting active inflammation in the terminal ileum and large bowel. RESULTS: A total of 91 magnetic resonance enterography studies were reviewed. Of these, 45 had comparison histopathology within 45 days. The overall sensitivity of MRE for detecting active inflammation compared to ileocolonoscopy was 92% for both readers, while specificity was 100% for reader 1 and 75% for reader 2. Of the individual parameters evaluated, mucosal hyperenhancement and bowel wall thickening were the most sensitive indicators of active inflammation, each having sensitivity of 86% and specificity of 88%. Cohen's κ coefficient was 0.59, indicating moderate agreement between the readers. CONCLUSIONS: MRE has high overall diagnostic accuracy for detecting active bowel inflammation in pediatric patients compared to ileocolonoscopy and demonstrates moderate interreader reliability.
RATIONALE AND OBJECTIVES: The purpose of this study was to determine the accuracy of magnetic resonance enterography (MRE) compared to histopathology in the evaluation of pediatric inflammatory bowel disease and to assess interreader reliability for image interpretation. MATERIALS AND METHODS: All magnetic resonance enterography studies performed for known or suspected inflammatory bowel disease between July 2009 and July 2010 were retrospectively reviewed by two pediatric radiologists. Exams were evaluated for signs of enteric inflammation and extraenteric disease. A five-point, Likert-type scale was used to assess the overall likelihood of active inflammation, with scores ≥ 3 considered positive. Cohen's κ coefficient was calculated to assess interreader agreement. A subset of patients who had undergone ileocolonoscopy or surgery with confirmed histopathology within 45 days of MRE were used to assess the accuracy of MRE for detecting active inflammation in the terminal ileum and large bowel. RESULTS: A total of 91 magnetic resonance enterography studies were reviewed. Of these, 45 had comparison histopathology within 45 days. The overall sensitivity of MRE for detecting active inflammation compared to ileocolonoscopy was 92% for both readers, while specificity was 100% for reader 1 and 75% for reader 2. Of the individual parameters evaluated, mucosal hyperenhancement and bowel wall thickening were the most sensitive indicators of active inflammation, each having sensitivity of 86% and specificity of 88%. Cohen's κ coefficient was 0.59, indicating moderate agreement between the readers. CONCLUSIONS: MRE has high overall diagnostic accuracy for detecting active bowel inflammation in pediatric patients compared to ileocolonoscopy and demonstrates moderate interreader reliability.
Authors: Dana M H Dykes; Alexander J Towbin; Erin Bonkowski; Claudia Chalk; Ramona Bezold; Kathleen Lake; Mi-Ok Kim; James E Heubi; Bruce C Trapnell; Daniel J Podberesky; Lee A Denson Journal: Inflamm Bowel Dis Date: 2013-09 Impact factor: 5.325