Assessing Disease Activity in Pediatric Crohn's Disease Using Ultrasound: The Pediatric Crohn Disease Intestinal Ultrasound Score.

ABSTRACT

OBJECTIVES: Currently, there is no consensus on how to score Crohn disease (CD) activity assessed by intestinal ultrasound (IUS) in children. This study aimed to design an easy-to-use IUS score for disease activity in pediatric CD. METHODS: Children undergoing ileo-colonoscopy for CD assessment underwent IUS the day before ileo-colonoscopy, assessed with simple endoscopic score for CD (SES-CD). IUS features were compared to the SES-CD on segmental level. Multiple regression analyses, separately for terminal ileum (TI) and colon, were done to assess predictors of disease activity and to develop a model. RESULTS: In 74 CD patients (median 15 years, 48% female), 67 TI and 364 colon segments were assessed. Based on receiver operating characteristics curves, bowel wall thickness (BWT) was categorized into low [1 point 2-3 mm (TI) and 1.6-2 mm (colon)], medium [2 points 3.0-3.7 mm (TI) and 2.0-2.7 mm (colon)], and high [3 points >3.7 mm (TI) and >2.7 mm (colon)]. In TI, only BWT was retained in the model [high BWT odds ratio (OR) 11.50, P < 0.001]. In colon, BWT (high BWT OR 8.63, P < 0.001) and mesenteric fat (1 point OR 3.02, P < 0.001) were independent predictors. A pediatric Crohn disease IUS score (PCD-US) cut-off of 1 resulted in a sensitivity of 82% (95% confidence interval, CI 65%-93%) and 85% (95% CI 80%-89%) and a cut-off of 3 in a specificity of 88% (72%-97%) and 92% (87%-96%) for TI and colon, respectively. Inter-observer agreement was moderate for TI and colon ( K 0.42, K 0.49, respectively). CONCLUSIONS: The PCD-US score is an easy-to-use and reliable score to detect or rule out CD activity on segmental level in children. External validation is needed before applying this score in clinical practice.