Does Colonoscopy Preparation Cause a Flare of Ulcerative Colitis?

Abstract

Purpose: Anecdotally, colonoscopy preparation seems to increase ulcerative colitis (UC) symptoms in some patients. The goal of this study is to determine if there is an association between colonoscopy preparation and disease relapse and to identify demographic factors, disease characteristics, or medications that are associated with increased symptoms after colonoscopy preparation. Methods: Initial assumption of a 20% flare rate led to a planned study size of 30. Thirty-five outpatients with a history of UC, intact colon and quiescent disease as determined by the patient were enrolled in a prospective ABA crossover study in which the patients served as their own controls (A = no prep B = colonoscopy prep). Patients were clinically evaluated with the Simple Clinical Colitis Activity Index (SCCAI) for the week prior to colonoscopy(A), week 1(B) and week 4(A) after colonoscopy. Data analysis used a mixed model to accommodate non-independence of repeated measurements. Results: 31 patients completed the study. Mean age was 48 yrs, with 97% Caucasian and 55% men. Patients had UC for an average of 14.1 yrs (1.5–54 yrs), with an average of 0.64 hospitalizations/5yrs, 3.7 flares/yr, and 1.7 episodes requiring steroids/5yrs. 68% used 5-ASA, 36% used thiopurines, and 23% used chronic prednisone. 77% used fleets phospho-soda prep. Pre-colonoscopy, 2(6%) patients felt that preps lead to UC flares, while 3(10%) were unsure, and the majority (84%) thought there was no effect. Six (19%) pts required an escalation of their 5-ASA meds post-colonoscopy, 2 of these 6 pts flared as defined by a score ≥5 on the SCCAI at 1 wk post colonoscopy and 4 of 6 patients had a 1–3 point increase in their SCCAI. None required treatment with steroids or hospitalization. Multivariate modeling demonstrated that baseline SCCAI was predictive of increased SCCAI (p value < .001). Thiopurines were protective against increased symptoms (p = .01) and baseline prednisone use predicted increased post-colonoscopy symptoms (p = .02). There was a trend toward significance (p = 0.10) for the association between colonoscopy preparation and increased disease activity by SCCAI. Conclusions: Nineteen percent of patients required an increase in their 5-ASA meds post-colonoscopy. Baseline SCCAI and use of prednisone were predictive of disease flare, while thiopurines were protective. There was a trend toward significance in an association between colonoscopy prep and an increase in the SCCAI, additional subjects are being enrolled.