P661 Endoscopy-based therapeutic management in postoperative recurrent Crohn’s disease: Results of a multi-centre retrospective study

Abstract

Background: The Rutgeerts' score (RS) is used to guide management of patients with Crohn's disease (CD) after ileal resection with ileocolonic anastomosis. The modified RS differentiates lesions at the anastomosis ± <5 isolated neo-terminal ileal erosions (i2a) from presence of ≥5 isolated neo-terminal ileal erosions ± anastomotic lesions (i2b). We investigated if clinical relapse (CR) and need for new endoscopic/surgical intervention (NI) differ between i2a and i2b endoscopic recurrence (ER). Our secondary objective was to evaluate the effect of therapy modification based on postoperative endoscopic findings. Methods: This was a retrospective study including patients undergoing a colonoscopy <12 months after ileal resection with ileo-colonic anastomosis with >3 years of follow-up after endoscopy. CR was defined as the occurrence of CD-related symptoms along with objective signs of disease activity. NI included endoscopic dilatation, re-resection or stricturoplasty of ileo-colonic anastomosis. Kaplan-Meier curves were plotted for time from index endoscopy to CR and NI. Results: The study population consisted of 365 patients. At index endoscopy, 74 patients (20%) had a RS i0, 37 (10%) i1, 91 (25%) i2a, 89 (24%) i2b, 42 (12%) i3, and 32 (9%) i4. During a median (IQR) follow-up of 88 (67-118) months after index endoscopy, CR and NI were observed in 176 (48.2%) and 81 (22.2%) patients, respectively. Patients with an i2 score overall had a 49% risk for CR and a 27% risk for NI. No difference was observed concerning the risk of CR between i2a and i2b patients (44% and 54%, respectively, log-rank p = 0.31). Risk of NI was not different between i2a and i2b patients (34% and 19%, respectively, p = 0.11). Probability of CR and NI was significantly higher in patients with an i3 or i4 compared with patients with i0-i2 (log-rank p < 0.001 and p < 0.01, respectively). Exposure to anti-TNF before surgery, active smoking, an RS i3-i4, or CRP >5 mg/l were independent predictors of CR. A modest but significant decrease of CR risk was observed for patients with an RS i3-i4 when an immunosuppressant or biological therapy was introduced after endoscopy (47% vs. 79%, p = 0.03). Among i2a/i2b patients, no difference was observed in terms of CR depending on immunosuppressant or biological therapy initiation after endoscopy (46% vs. 44% in i2 overall cohort) (p = 0.46). Conclusions: No difference was observed in terms of CR and NI between patients with RS i2a or i2b. In patients with a RS i2a/i2b, no effect on probability of CR was observed when an immunosuppressant or a biological treatment was initiated after index endoscopy. Intensification of therapy in patients with i3-i4 resulted in significant but modest improvement of clinical relapse risk.