P0253 Identification of baseline Pediatric Crohn’s disease location and complications: How do intestinal ultrasound and magnetic resonance enterography compare?

Background Pediatric patients with newly diagnosed Crohn’s disease (CD) often undergo magnetic resonance enterography (MRE) to assess bowel not accessible by endoscopy and to evaluate for CD complications. In contrast to intestinal ultrasound (IUS), MRE access is quite limited, expensive, and not feasible for young patients. Little is known about how MRE and IUS compare, with a recent systematic review reporting inconclusive results when comparing diagnostic accuracy in pediatric CD. This study therefore aimed to compare MRE and IUS at diagnosis of pediatric CD. Methods Pediatric patients (≤18 years old) seen for suspected CD were prospectively consecutively enrolled (2019-2022) and underwent endoscopy (Simple Endoscopic Score for Crohn Disease (SES-CD)), MRE, IUS (performed by IUS trained pediatric gastroenterologists), calprotectin, blood work and assessment (weighted Pediatric Crohn Disease Activity Index (wPCDAI)). A blinded pediatric radiologist read all MREs. An interim analysis was done after 21 patients before the final analysis. Statistical analyses included descriptive statistics and Pearson chi-square. Results 58 patients (n=38, 66% male) were included; median age 13 years (IQR 10-15, range 6-17). Disease activity at diagnosis was moderate: median SES-CD 14 (IQR 10-21), wPCDAI 48 (IQR 34-59), calprotectin 1455ug/g (IQR 840-2454), C-reactive protein 47mg/L (IQR 27-72), and erythrocyte sedimentation rate 33mm/hr (IQR 18-46). Jejunum and proximal ileum involvement had high levels of agreement between MRE and IUS for non-inflamed bowel, but less agreement for inflammation (Figure 1). There was higher agreement for inflamed distal ileum compared to proximal ileum or jejunum (Figure 1). MRE and IUS performed similarly in identifying endoscopically inflamed bowel, with the highest accuracy for the terminal ileum (TI) (Figure 2). Both were more accurate for moderate/severe (segment SES-CD≥7) vs. mild/moderate/severe disease (segment SES-CD≥3) (Figure 2). For patients with endoscopic TI strictures (n=8/58, 14%), this was seen on MRE (n=8/8, 100%) more often than IUS (n=5/8, 63%). Those without endoscopic TI strictures were still reported on MRE (n=15/50, 30%), more than IUS (n=2/50, 4%). There were no significant differences in MRE vs. IUS bowel wall thickness measurements in all segments of the small and large bowel (p>0.05). Conclusion IUS and MRE performed similarly in distal small bowel and colonic CD assessment in both the identification and measurement of inflamed bowel. More proximal small bowel disease may be over reported by MRE or under reported by IUS. MRE also over reported TI stricturing disease. Both may miss mild CD. For initial and longitudinal pediatric CD assessment, repeating IUS rather than MRE is a viable option.