P229 The modified postoperative endoscopic recurrence score for Crohn's disease: Does it really make a difference in predicting clinical recurrence?
Rivière P.*1, Vermeire S.2, Van Assche G.2, Rutgeerts P.2, De Buck van Overstraeten A.3, D'Hoore A.3, Ferrante M.2
1Bordeaux University Hospital, Bordeaux, France 2UZ Leuven, Campus Gasthuisberg, Department of Gastroenterology, IBD, Leuven, Belgium 3UZ Leuven, Campus Gasthuisberg, Department of Abdominal Surgery, Leuven, Belgium
The Rutgeerts' score (RS) is widely used to guide post-operative management of patients with Crohn's disease (CD). The modified RS differentiates lesions at the anastomosis with or without <5 isolated neo-terminal ileal erosions (i2a) from presence of ≥5 isolated neo-terminal ileal erosions with or without anastomotic lesions (i2b), but its predictive value and clinical relevance has not been validated. We investigated if clinical relapse (CR) and need for endoscopic/surgical intervention (ESI) differ between i2a and i2b endoscopic recurrence (ER).
This was a retrospective, single-center study including all patients operated between 2000 and 2013 with an i2 ER observed <12 months after right hemicolectomy with ileo-colonic anastomosis. The modified RS was attributed based on the available endoscopic report and images. CR was defined as the occurrence of CD related symptoms along with objective signs of disease activity. ESI was defined as the need for balloon dilatation at site of the anastomosis, or new ileocolonic resection. Kaplan-Meier curves were plotted for time from index endoscopy to CR and ESI.
The study population consisted of 94 patients [43 males, median age 37 years]. At index endoscopy, 53 patients (56%) had an i2a ER, and 41 (44%) an i2b ER. Groups were not different regarding disease characteristics and post-operative prophylactic therapy. Medical treatment was optimized according to index colonoscopy in 8 (15%) patients with i2a and 20 (49%) with i2b ER (Odds ratio (OR) 5.2 (95% CI 2.0–14.6), p<0.001). During a median (IQR) follow-up of 78 (37–109) months, CR and ESI were observed in 47 (50%) and 21 (22%) patients, respectively. As shown in Figure 1, i2a and i2b scores were not predictive of CR or ESI (Log Rank p=0.37 and p=0.10, respectively). Also after exclusion of patients with immediate post-endoscopy treatment optimization, i2a and i2b scores were not predictive (Log Rank p=0.73 and p=0.34, respectively). A previous ileocolonic resection (OR 2.0 (1.1–3.9), p=0.04) was associated with CR; immediate post-operative prophylactic therapy by anti-TNF was protective against CR (p=0.03). Post-operative prophylactic therapy by thiopurine was protective against ESI (p=0.02).
No difference was observed in terms of clinical relapse and need for endoscopic/surgical intervention between patients with i2a or i2b endoscopic recurrence after right hemicolectomy with ileocolonic anastomosis. Further study is needed to confirm these results and evaluate the outcome of Rutgeerts' score i2 patients.