Usefulness of ultrasonography for small intestinal evaluations in pediatric Crohn's disease.

Abstract

Small bowel capsule endoscopy (SBCE) for Crohn's disease is useful; however, its use has some limitations, such as invasiveness when endoscopic assistance is required in patients who cannot swallow the capsule, and the burden of interpretation on a physician. In contrast, intestinal ultrasonography (IUS) is a non-invasive modality for children. The purpose of this study is to evaluate the accuracy of IUS for pediatric patients with established Crohn's disease. Small bowel capsule endoscopy and IUS findings from the same period in pediatric patients with established Crohn's disease were analyzed retrospectively. First, we compared the Lewis score (LS), small bowel endoscopic activity, and IUS findings by small bowel wall thickness (SBWT) and mesenteric lymph node size (MLNS). Second, we compared the performance of IUS findings with those of some biomarkers. In 22 procedures, SBWT and MLNS were correlated with LS (r = 0.52, P < 0.05, and r = 0.45, P < 0.05, respectively). Small bowel wall thickness, erythrocyte sedimentation rate, and fecal calprotectin levels had the highest accuracy (81.8%, 81.8%, and 81.8%, respectively). The combination of SBWT and MLNS had the highest positive predictive value and negative predictive value (100% and 83.3%, respectively). Intestinal ultrasonography findings, including SBWT and MLNS, are useful for monitoring small bowel lesions in pediatric patients with established Crohn's disease. We suggest first evaluating small bowel inflammation by IUS in pediatric patients with Crohn's disease before SBCE because IUS is less invasive than SBCE.