P392 Risk factors for postoperative recurrence after ileocolic resection for Crohn’s Disease: long-term results from a population-based cohort.

Abstract

Abstract Background Crohn’s disease (CD) patients often require surgical interventions and post operative recurrence is common. Data regarding risk factors for recurrence in representative cohorts are scarce. We assessed the risk-factors of recurrence of CD after ileocolic resection (ICR) in a population based cohort. Methods Patients who underwent a primary or re-resection for ileocolic CD between 2003 and 2017 were included. The primary endpoint was clinical recurrence. The secondary outcome was surgical recurrence. A Cox regression model was used to assess recurrence free survival. Results 289 patients were included of which 120 male (41.5%). Most common indication for ICR was stricturing (B2) disease in 146 (50.5)%), followed by penetrating (B3) disease in 101 (34.9%). Recurrence occurred in 189/289 patients (65.4%) after a median follow up of 2.3 years and surgical recurrence in 22 (7.6%) after median follow up of 4.8 years. Multivariate analysis showed a significant higher chance of clinical recurrence in smokers (HR: 1.659 (95% CI: 1.230; 2.238)) (p<0.05), patients with prior resections (HR: 1.514 (95% CI: 1.014; 2.261)) (p<0.05), perianal involvement (HR: 1.453 (95% CI: 1.033; 2.044)) (p<0.05) and isolated involved proximal resection margin (HR: 1.615 (95% CI: 1.039; 2.509)) (p<0.05). B2 and B3 disease behaviour and post-operative medication was associated with less recurrence, respectively (HR 0.397 (95% CI: 0.253; 0.624)) (p<0.05), (HR 0.443 (95% CI: 0.282; 0.696)), (p<0.05) and (HR: 0.658 (95% CI: 0.473; 0.916)) (p<0.05). Considering surgical recurrence, underweight and overweight BMI and open surgery decreased the chance on recurrence. Conclusion Our study in a population based cohort confirmed risk factors for recurrence consistent with current evidence, but also identifies novel and less frequently found risk factors. Medical therapy reduces the risk for recurrence on a population level. We found an association between involved resection margin and recurrence. Differences between studies may be caused by differences in patient selection or definition of recurrence. Confirmation of this finding in independent representative cohort is necessary. This indicates a need for a standardised scientific database for future prospective research.