Background: There is a paucity of published literature describing ultrasound (US)-US and US-MR enterography (MRE) inter-radiologist agreement in pediatric small bowel Crohn disease.
Objective: To prospectively assess US-US and US-MRE inter-radiologist agreement in pediatric small bowel Crohn disease.
Materials and methods: Institutional Review Board approval and informed consent/assent were obtained for this HIPAA-compliant prospective cohort study of children with newly diagnosed distal small bowel Crohn disease (July 2012 to December 2014). Enrolled subjects (n = 29) underwent two small bowel US examinations performed by blinded independent radiologists both before and at multiple time points after initiation of medical therapy (231 unique US examinations, in total); 134 US examinations were associated with concurrent MRE. The MRE examination was interpreted by a third blinded radiologist. The following was documented on each examination: involved length of ileum (cm); maximum bowel wall thickness (mm); amount of bowel wall and mesenteric Doppler signal, and presence of stricture, penetrating disease and/or abscess. Inter-radiologist agreement was assessed with single-measure, three-way, mixed-model intra-class correlation coefficients (ICC) and prevalence-adjusted, bias-adjusted kappa statistics (κ). Numbers in brackets are 95% confidence intervals.
Results: Ultrasound-US agreement was moderate for involved length (ICC: 0.41 [0.35-0.49]); substantial for maximum bowel wall thickness (ICC: 0.67 [0.64-0.70]); moderate for bowel wall Doppler signal (ICC: 0.53 [0.48-0.59]); slight for mesenteric Doppler signal (ICC: 0.25 [0.18-0.42]), and moderate to almost perfect for stricture (κ: 0.54), penetrating disease (κ: 0.80), and abscess (κ: 0.96). US-MRE agreement was moderate for involved length (ICC: 0.42 [0.37-0.49]); substantial for maximum bowel wall thickness (ICC: 0.66 [0.65-0.69]), and substantial to almost perfect for stricture (κ: 0.61), penetrating disease (κ: 0.72) and abscess (κ: 0.88).
Conclusion: Ultrasound-US agreement was similar to US-MRE agreement for assessing pediatric small bowel Crohn disease. Discrepancies in US-US and US-MRE reporting question the utility of US as an accurate, reproducible radiologic biomarker for assessing response to medical therapy and disease-related complications.
Keywords: Adolescents; Children; Crohn disease; Inter-observer agreement; MR enterography; Ultrasound.