Search in the Abstract Database

Search Abstracts 2013

* = Presenting author

P308. Advanced endoscopy for prediction of relapse in ulcerative colitis

A. Jauregui-Amezaga1, M. Lopez-Ceron1, M. Aceituno1, M. Jimeno2, C. Rodriguez de Miguel1, E. Sanabria1, P. Susana1, M. Zabalza1, M. Sans1, E. Ricart1, I. Ordas1, B. Gonzalez1, M. Cuatrecasas2, J. Llach1, J. Panés1, M. Pellise1, 1Hospital Clínic i Provincial, Gastroenterology, Barcelona, Spain, 2Hospital Clínic i Provincial, Anatomical Pathology, Barcelona, Spain


Subclinical inflammation is considered a risk factor for relapse in ulcerative colitis (UC) and markers of subclinical inflammation are required. Pit pattern assessment by high-resolution colonoscopy in quiescent UC patients has been shown as a significant predictor of relapse in a previous study. The aim of this prospective study was to determine the accuracy of advanced endoscopy for prediction of relapse in UC, in comparison with fecal and serum biomarkers.


UC patients in clinical remission for at least three months and a baseline Mayo Endoscopic Subscore 0 were followed-up for one year or until relapse (determined by Mayo Score ≥1 and Mayo Endoscopic Subscore ≥1) with clinical evaluations, fecal calprotectin (FC) and serum biomarkers every three months. High-resolution rectosigmoidoscopy (Olympus H180) was performed at baseline and at the end of study. Mucosal pit pattern was assessed using 0.5% blue methylene chromoendoscopy and Narrow Band imaging (NBI) according to Nishio classification and mucosal vascular pattern intensity was assessed with NBI. Histological activity was evaluated using the Matts score.


82 UC patients in clinical remission were screened, 12 were excluded due to endoscopic activity, 6 dropped off during the follow up. Only patients with complete follow-up (n = 64) were analyzed. 17/64 patients (27%) relapsed within the follow-up period. Baseline clinical characteristics in relapsers and non-relapsers were similar. There were no differences in Nishio score determined by chromoendoscopy or NBI between both groups and vascular pattern as assessed by NBI was not significantly associated with relapse (see Table 1). Serum biomarkers and histology were not predictors of relapse. FC was predictor of relapse in three-month period (mean/SD in relapsers 532/899 vs non-relapsers 179/421 µg/g, p = 0.03) but not in one-year period, although 59% of one-year relapsers' FC levels were higher than 250 µg/g. However, the accuracy of FC for predicting relapse was not high (low sensitivity).

Table 1. High-resolution rectosigmoidoscopy: Pit patterns were established by NBI and CE and vascular pattern was assessed by NBI. No statistical differences between relapsers and non-relapsers were found.
Endoscopic findingsRelapse (n = 17)No relapse (n = 47)P value
Chromoendoscopy pit pattern  0.69
Small and round pits7 (41%)24 (51%) 
Large and somewhat irregular pits9 (53%)19 (40%) 
Very irregular pits1 (6%)4 (9%) 
NBI pit pattern  0.57
Small and round pits12 (71%)33 (70%) 
Large and somewhat irregular pits3 (17%)13 (28%) 
Very irregular pits2 (12%)1 (2%) 
NBI vascular intensity  0.72
Weak2 (12%)2 (4%) 
Normal12 (70%)38 (81%) 
Strong3 (18%)7 (15%) 


In this study advanced endoscopy, serum/fecal biomarkers and histology were useful for prediction of relapse in one-year period in UC patients in clinical and endoscopic remission, although fecal calprotectin could predict relapse in three-month period with low accuracy.