P768 Long-term prognosis of Crohn’s disease and its temporal changes between 1986 and 2015 in a population-based cohort in the Songpa-Kangdong district of Seoul, Korea

Abstract Background No previous population-based study has evaluated the natural course of Crohn’s disease (CD) over three decades in non-Caucasians. We previously reported a 30-year trend in the epidemiology of inflammatory bowel disease (IBD) in the Songpa-Kangdong (SK) District of Seoul, Korea between 1986 and 2015 (1). In this study, we aimed to analyse the long-term natural course of Korean patients with CD in the SK-IBD population-based cohort. Methods All patients newly diagnosed with CD between 1986 and 2015 were enrolled in this study. To assess the temporal trends in treatment paradigms and in the prognosis of CD, patients were divided into two cohorts according to the year of CD diagnosis: cohort 1, 1986–2003 and cohort 2, 2004–2015 (the anti-tumour necrosis factor [anti-TNF] era). Disease characteristics at diagnosis, outcomes and their predictors were evaluated. Results A total of 418 patients were enrolled. There were 318 males (76.1%) and median age at CD diagnosis was 22 years (interquartile range [IQR], 18–29). Disease location at CD diagnosis was ileal in 104 patients (24.9%), colonic in 39 (9.3%), and ileocolonic in 275 (65.8%). Disease behaviour at CD diagnosis was inflammatory in 339 patients (81.1%), stricturing in 34 (8.1%), and penetrating in 45 (10.8%). Perianal fistula/abscess was present in 43.3% (n = 181) before or at CD diagnosis. During the median follow-up of 108.1 months, the overall use of systemic corticosteroids, thiopurines, and anti-TNF agents was 57.4%, 80.9%, and 34.2%, respectively. Compared with the cohort 1, the cumulative probability of commencing corticosteroids decreased (p = 0.001), whereas that of commencing thiopurines and anti-TNF agents increased (both p < 0.001) in the cohort 2. A total of 113 patients (27.0%) underwent intestinal resection, demonstrating cumulative risks of intestinal resection at 1, 5, 10, 20, and 25 years after diagnosis of 12.5%, 16.5%, 25.6%, 49.7%, and 55.5%, respectively. Multivariate Cox regression analysis revealed that stricturing behaviour at diagnosis (hazard ratio [HR] 2.393, 95% confidence interval [CI] 1.234–4.641), penetrating behaviour at diagnosis (HR 4.514, 95% CI 1.752–11.629), and the cohort 2 (HR 0.530, 95% CI 0.297–0.945) were independent predictors of intestinal resection. The standardised mortality ratio was 1.867 (95% confidence interval, 0.502–4.780). Conclusion Korean patients showed a similar clinical course and intestinal resection rate compared with Western patients. The risk of intestinal resection has decreased in the anti-TNF era. Reference