Clinical Predictors of Early and Late Endoscopic Recurrence Following Ileocolonic Resection in Crohn's Disease

Background and Aims Multiple factors are suggested to place Crohn's disease patients at risk of recurrence after ileocolic resection with conflicting associations. We aimed to identify clinical predictors of recurrence at first [early] and further [late] postoperative colonoscopy.Methods Crohn's disease patients undergoing ileocolic resection were prospectively recruited at six North American centres. Clinical data were collected and endoscopic recurrence was defined as Rutgeerts score >= i2. A multivariable model was fitted to analyse variables independently associated with recurrence.Results A total of 365 patients undergoing 674 postoperative colonoscopies were included with a median age of 32 years, 189 [51.8%] were male, and 37 [10.1%] were non-Whites. Postoperatively, 133 [36.4%] used anti-tumour necrosis factor [anti-TNF] and 30 [8.2%] were smokers. At first colonoscopy, 109 [29.9%] had recurrence. Male gender (odds ratio [OR] = 1.95, 95% confidence interval [CI] 1.12-3.40), non-White ethnicity [OR = 2.48, 95% CI 1.09-5.63], longer interval between surgery and colonoscopy [OR = 1.09, 95% CI 1.002-1.18], and postoperative smoking [OR = 2.78, 95% CI 1.16-6.67] were associated with recurrence, while prophylactic anti-TNF reduced the risk [OR = 0.28, 95% CI 0.14-0.55]. Postoperative anti-TNF prophylaxis had a protective effect on anti-TNF experienced patients but not on anti-TNF naive patients. Among patients without recurrence at first colonoscopy, Rutgeerts score i1 was associated with subsequent recurrence [OR = 4.43, 95% CI 1.73-11.35].Conclusions We identified independent clinical predictors of early and late Crohn's disease postoperative endoscopic recurrence. Clinical factors traditionally used for risk stratification failed to predict recurrence and need to be revised.