Search in the Abstract Database

Search Abstracts 2011

* = Presenting author

5. Depth of remission may not predict outcome of UC over 2 years

D.C. Burger, S.J. Thomas, A.J. Walsh, A. Von Herbay, O.C. Buchell, S. Keshav, B.F. Warren, S. Travis

John Radcliffe Hospital, Oxford, United Kingdom

Introduction: Remission is the goal of therapy in ulcerative colitis (UC) and a primary endpoint of clinical trials. Disparity between clinical, endoscopic and histological assessments of activity has long been recognised, but since the definition of remission affects clinical and regulatory decisions, the disparity and impact on outcome needs re-evaluation.

Aims and Methods: To quantify disparities between clinical, endoscopic and histological assessments of remission in patients with UC and effects on outcome during 2 yr follow up. Clinical activity was defined by the Simple Clinical Colitis Activity Index, endoscopic activity by the Baron score and microscopic activity by the Truelove & Richards' score (remission scores ≤2, ≤1 and ‘no significant inflammation’ respectively). Consecutive patients requesting an early appointment were assessed by 4 gastroenterologists, before video sigmoidoscopy and mucosal biopsy. Each scored clinical and endoscopic activity independently. Histological activity was scored by two pathologists. Outcome measures included treatment escalation, hospitalisation and colectomy. Fleiss' kappa was used to evaluate interobserver variation and two-tailed Fisher's exact test to compare outcomes.

Results: 91 patients were recruited, median age at diagnosis 31 yr (11–66); 60 were in remission by at least one definition: 37 (41%) were in clinical, 30/37 (81%) were also in histological remission, 30/37 (81%) were in endoscopic remission and 20/37 (76%) (20/91 = 22%) were in remission by all three measures. Of 56 in endoscopic remission: 26/56 (46%) had clinical activity, and 10/26 (38%) had moderate clinical activity (SCCAI 6–8). Of 47 in histological remission only 5/47 had endoscopic activity (42/47 (81%) were in endoscopic remission and 30/47 (64%) were in clinical remission). Agreement between histological and endoscopic assessment (κ = 0.58, moderate agreement) was better than between clinical and endoscopic (κ = 0.27), or clinical and histological (κ = 0.47), or between all three methods (κ = 0.44). Follow up data was available in 54/60 (90%), with 29 mo (5–35 mo) median follow-up; median age at review 48 yr (19–74), 57% female, 30% proctitis, 48% distal, 22% extensive. Most (83%) took maintenance therapy: 5ASA (31/54), immunomodulators (13/54), or infliximab (1/54). 43 (80%) maintained steroid-free remission at 12 mo and 34 (63%) during follow up; of those needing steroids, 9 (45%) had >1 course during follow up and 3 (6%) infliximab. 4 (7%) needed hospitalisation and 1 (2%) colectomy. 12 month steroid-free remission did not differ according to the depth of remission (80% all 3 criteria, 75% on 2 criteria, 83% on 1 criterion, p = 0.7).

Conclusion: Modest agreement between clinical, endoscopic or histological activity emphasises the differences between the measures, all of which may have prognostic value. Clinical assessment systematically over-estimates disease activity (and therefore under-estimates remission), but the outcome over 2 years was not affected by the definition of remission, possibly because of the sample size.