P177 Validation of UC Intestinal Ultrasound (UC-IUS) Index for children with Ulcerative Colitis
HuynhDr, H.(1);Ma, H.(1);Isaac, D.(1);Novak, K.(2);Almeida, P.(1);Kim, J.(1);Kuc, A.(1);Carroll, M.(1);Wine, E.(1);
(1)University of Alberta, Department of Paediatrics- Gastroenterology and Nutrition, Edmonton, Canada;(2)University of Calgary, Division of Gastroenterology and Hepatology, Calgary, Canada;
Background
Transabdominal bowel ultrasound (TABUS) is an ideal tool to assess transmural inflammation in children with Ulcerative Colitis (UC). The UC intestinal US (UC-IUS) Index was developed and validated using endoscopy with strong correlation between UC-IUS Index and Mayo subscore (ρ 0.830; p<0.001). Our aim was to determine how bowel wall thickness (BWT) and UC-IUS Index performed in children at diagnosis in comparison to endoscopy using the Mayo score.
Methods
Subjects (0-18 years old) with suspected inflammatory bowel disease (IBD) were prospectively enrolled. Baseline TABUS (excluding rectum due to poorly seen) done prior to endoscopy. Pediatric Ulcerative Colitis disease Activity Index (PUCAI) and Mayo were calculated, albumin, C-reactive protein (CRP) and fecal calprotectin (FCP) collected. The UC-IUS Index was calculated for each segment – sigmoid (SC), descending (DC), transverse (TC) and ascending (AC) [bowel wall thickness (BWT) (mm) : > 2 =1, >3 =2 and >4 =3; doppler: spots=1 and stretches =2, abnormal haustrations = 1 and fat wrapping =1]. Spearman’s rank (rho) and Pearson’s correlation (r) assess for a correlation. Receiver operating characteristic [ROC] analysis performed for BWT to determine sensitivity and specificity of BWT cut-offs in the UC-IUS.
Results
Of the 75 subjects recruited for suspected IBD, 26 had UC with mean age 15 years (SD 3.33) and 5 had normal US. Twenty three have extensive pancolitis. Mean PUCAI score : 60 (SD 23.28), CRP 22.62 (SD 39.5) mg/l, albumin 35.6 (SD 6.96) g/l and fecal calprotectin of 2223 (SD 1757) mg/kg. ROC curves generated using a total of 122 colonic segments. BWT of 2 mm discriminates between active and inactive :Mayo 0 and Mayo1-3 [sensitivity 84.9%; specificity 86.2%; an area under the curve [AUC] 0.900]; a cut-off of 3mm discriminates between Mayo 1 from Mayo 2-3 [sensitivity of 61.3%; specificity 90.5%; AUC 0.858]; a cut of 4mm discriminates Mayo 3 (sensitivity 64.7%; specificity 85.7 %; AUC 0.876). BWT and UC-IUS scores of all colonic segments correlated highly positively with the Mayo score of corresponding segments (rho=0.684, r=0.660, p<0.001) and (rho=0.750, r=0.722, P<0.001) respectively. The UC-IUS Index correlates poorly with CRP, ESR, and Fecal Calprotectin - r = 0.262, 0.346 and 0.100 respectively; p>0.05.
Conclusion
BWT and UC-IUS Index correlated highly positively with the Mayo subscore in children with UC. UC-IUS has a better correlation than BWT. BWT cut-off of 2mm discriminate between normal and mild Mayo is optimal. Cut-offs of 3 mm and 4 mm for moderate and severe Mayo may be too high in children with lower sensitivity. A larger cohort of children with UC will need to be studied to determine optimal BWT cut-off for moderate and severe disease.