Premium
Evaluation of noncontrast MR enterography for pediatric inflammatory bowel disease assessment
Author(s) -
Lanier Michael Hunter,
Shetty Anup S.,
Salter Amber,
Khanna Geetika
Publication year - 2018
Publication title -
journal of magnetic resonance imaging
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.563
H-Index - 160
eISSN - 1522-2586
pISSN - 1053-1807
DOI - 10.1002/jmri.25990
Subject(s) - medicine , inflammatory bowel disease , retrospective cohort study , radiology , cohen's kappa , logistic regression , population , gastroenterology , disease , environmental health , machine learning , computer science
Background Gadolinium deposition in normal tissues is being increasingly recognized. Children with inflammatory bowel disease (IBD) undergo frequent imaging with contrast‐enhanced MR enterography (MRE). Purpose To determine the impact of intravenous (IV) gadolinium in assessment of pediatric IBD by MRE. Study Type Retrospective, case series. Population Radiology information system was searched to identify all children who underwent MRE and endoscopy within 30 days in 2016. Field Strength/Sequence 1.5T and 3T. Assessment Imaging studies were evaluated for bowel inflammation interpreted by two blinded radiologists in two sessions 6 weeks apart (session 1 pre‐MRE; session 2 pre/postcontrast‐MRE). Endoscopic histology was the reference standard. Statistical Tests A logistic regression model was evaluated using receiver operating characteristics curves and expressed by c‐statistics. Agreement between readers was evaluated using Cohen's or weighted kappa statistic, as appropriate. Two‐sided P < 0.05 was considered statistically significant. Descriptive statistics were used for assessment of IBD complications. Results In all, 52 children (46% female), mean age 13.2 (SD 3.42) years formed the study cohort. 77% (40/52) had inflammation on endoscopic biopsy. Pre/post‐MRE showed no significant increase in the c‐statistic compared to pre‐MRE for assessment of small bowel (Reader 1 P = 0.56, Reader 2 P = 1.00) or large bowel inflammation (Reader 1 P = 0.42, Reader 2 P = 1.00)). Intravenous contrast showed no improvement in interobserver agreement for assessment of inflammation in small (kappa 0.92 pre‐MRE, 0.88 pre/post‐MRE) or large bowel (kappa 0.83 pre‐MRE, 0.73 pre/post‐MRE). IV contrast had no meaningful impact on interobserver agreement for length of small bowel inflamed (intraclass correlation coefficient 0.90 pre‐MRE, 0.95 pre/post‐MRE). Assessment of IBD complications was improved with IV contrast, with 3/5 cases with perianal penetrating disease not recognized on pre‐MRE. Data Conclusion Routine administration of IV gadolinium has no impact on the assessment of bowel inflammation. However, there is potential for missing perianal complications using a noncontrast MRE protocol without dedicated pelvic imaging. Level of Evidence : 3 Technical Efficacy : Stage 2 J. MAGN. RESON. IMAGING 2018;48:341–348.