Reliability of EUS indices to detect inflammation in ulcerative colitis

Abstract

Background and Aims EUS is a potentially useful modality to assess severity of inflammation in ulcerative colitis (UC). We assessed the reliability of existing EUS indices and correlated them with endoscopic and histologic scores. Methods Four blinded endosonographers assessed 58 endoscopic and EUS videos in triplicate, from patients with UC. Intrarater and interrater reliability of the hyperemia and Tsuga scores were estimated by using intra-class correlation coefficients (ICCs). Correlation with the Mayo endoscopy score, modified Baron score (MBS), Ulcerative Colitis Endoscopic Index of Severity (UCEIS), and Geboes histopathology score (GHS) were calculated by using bootstrapping methods. A RAND consensus process led to development of standardized definitions and a revised EUS-UC score. Results ICCs for intrarater reliability were 0.76 (95% confidence interval [CI], 0.71-0.80) for the hyperemia score and 0.85 (95% CI, 0.79-0.89) for the Tsuga score. Corresponding values for interrater reliability were 0.34 (95% CI, 0.25-0.42) and 0.36 (95% CI, 0.24-0.46). Correlation between hyperemia and Tsuga scores to Mayo scoring system, MBS, UCEIS, and the GHS were 0.39 (95% CI, 0.15-0.61) and 0.28 (95% CI, 0.04-0.51), 0.38 (95% CI, 0.16-0.57) and 0.25 (95% CI, -0.01-0.48), 0.41 (95% CI, 0.16-0.62) and 0.27 (95% CI, 0.01-0.50), 0.37 (95% CI, -0.01-0.48) and 0.24 (95% CI, 0.13-0.57), respectively. The revised EUS-UC score included bowel wall thickening, depth of inflammation, and hyperemia. Conclusions Although substantial to almost perfect intrarater agreement existed for EUS indices in UC, interrater agreement was fair. Standardization of item definitions with development of a revised evaluative instrument has potential application as an evaluative and prognostic tool for UC. (Clinical trial registration number: NCT01852760.)