P235 Clinical utility of the Lémann index and Rutgeerts score to predict postoperative course of Crohn's disease: a retrospective single-center cohort study.
G. Fiorino*1, C. Bonifacio2, M. Allocca1, M. Alfieri1, M. Sacchi3, I. Bravatà4, D. Gilardi1, P. Turri1, M. Delliponti1, A. Spinelli3, L. Balzarini2, A. Repici4, L. Peyrin-Biroulet5, S. Danese1
1Humanitas Research Hospital, IBD Center, Gastroenterology, Rozzano, Milan, Italy, 2Humanitas Research Hospital, Radiology, Rozzano, Milan, Italy, 3Humanitas Research Hospital, IBD Surgery, Colo-rectal Surgery, Rozzano, Milan, Italy, 4Humanitas Research Hospital, Endoscopy Unit, Gastroenterology, Rozzano, Milan, Italy, 5Nancy University Hospital, Université de Lorraine, Gastroenterology and Hepatology, Vandoeuvre-les-Nancy, France
The Rutgeerts' score (RS) predicts the risk for clinical postoperative Crohn's disease (CD) recurrence. Recently, the Lémann Index (LI) has been developed to quantify bowel damage (abscess, fistula, stricture).
We evaluated CD subjects undergoing intestinal resection from 2007 to 2013 and having both endoscopy within 6 and 12 months after surgery and magnetic resonance enterography (MRE) at the same time point (±60 days) and then followed every 12-18 months with the same examinations. Subjects with active perianal disease were excluded because of the relevant impact on the LI. Data on preventive medications, clinical relapse, further surgery or complications were analyzed. The correlation between RS and the LI was assessed by the Spearman's correlation test, and the predictive value of LI increase was assessed by logistic regression and Kaplan-Meyer curves.
Thirty-nine subjects were included in the analysis. Median follow-up (FU) was 29.0 months. Nineteen subjects (48.7%) had a RS of 3-4 within 12 months after surgery and 21 subjects (53.8%) had an increase in the LI during the same period as assessed by MRE. No subjects underwent further surgery, 7 subjects (18%) had clinical relapse, and 5 (12.8%) developed bowel damage (all strictures) during follow-up. The vast majority of subjects (84%) with early endoscopic recurrence (RS of 3-4) also had a an increase in the LI (p=0.0007). No significant correlation was found between RS and LI ( ρ=0.26, CI 95% -0.05-53.8, p=0.09) MRE was able to see pre-anastomotic ulcers in 40% of subjects with RS>2, while pre-anastomotic bowel wall thickening was seen in 89% of subjects with endoscopic recurrence compared to 25% with no endoscopic recurrence (p=0.0002) . Subjects with bowel wall thickening were more likely to have endoscopic recurrence (OR 25.5, CI 95% 4.3-151, p=0.0004). Subjects with an increase in the LI alone within 12 months were more likely to have a clinical relapse in the FU period (HR 0.0, CI 95% 0.04-0.87, p=0.03). Combined endoscopic recurrence (RS of 3-4) and LI increase within 12 months after surgery were also associated to a higher risk of clinical relapse during FU (HR 0.03, CI 95% 0.0000 to 0.0051, p<0.0001), while it was not predictive of further disease complications (OR 1.87, p=0.61).
Although in a small cohort, the assessment of endoscopic recurrence by RS and bowel damage by LI within 12 months since surgery may predict clinical recurrence. Bowel thickening after surgery is associated with endoscopic recurrence. Bowel damage as assessed by LI seems to be independent from endoscopic disease activity assessed by RS.