Abstract
AIM: To compare the diagnostic yield of small intestinal contrast ultrasonography (SICUS) with magnetic resonance enterography (MRE) in routine clinical practice in a cohort of pediatric patients investigated for Crohn's disease (CD) attending a UK tertiary center. METHODS AND RESULTS: Patients with suspected or established CD who underwent SICUS were identified retrospectively. SICUS was compared to conventional transabdominal ultrasound (TUS), ileocolonoscopy (IC), and MRE. The accuracy and agreement of SICUS in detecting small bowel lesions and CD-related complications were assessed using kappa (κ) coefficient statistics. A total of 93 patients (median age 15 years, range 2-17, 49 male) underwent SICUS; 58 had suspected and 35 had established CD. In suspected CD, sensitivity and specificity of SICUS in detecting CD small bowel lesions were 81.8 and 100% and for TUS 85.7 and 87.5%, respectively. In established CD, sensitivity and specificity of SICUS were 98.7 and 100% and TUS 80 and 100%, respectively. Agreement between SICUS and IC was substantial for the presence of lesions (κ = 0.73) but fair in TUS (κ = 0.31). Agreement between SICUS and IC was almost perfect for detecting strictures (κ = 0.84), with a sensitivity of 100% and specificity of 97.6%. When comparing SICUS and TUS with MRE, agreement for the presence of lesions was substantial (κ = 0.63) and moderate (κ = 0.53), respectively. Agreement between SICUS and MRE was substantial for detecting strictures (κ = 0.77) and dilatation (κ = 0.68). CONCLUSIONS: SICUS offers a radiation-free alternative for assessing pediatric small bowel CD, with diagnostic accuracy that is comparable to MRE and IC, supporting its wider use in routine practice.