S146 Poster presentations Patients with personal or family colon cancer history and personal polyp history were excluded. Results: Two hundred and thirty-six patients with IBD, 126 male (53.4%) and 110 female (46.6%), 63.6% with UC and 36.4% with CD, performed at least one colonoscopy during the time period, 179 (75.8%) had two or more procedures during this period. Nine hundred and fifty-three patients underwent screening colono- scopies, and 708 patients were randomly selected to comparison. The mean patient age was 57.1 years old for those with average risk and 56.4 years old for those with IBD (p=0.22). There were no significant statistically differences between the 2 groups regarding the variable gender. SAs were detected in 30 patients with IBD (12.7%), 23 in UC (15.3%) and 7 in CD (8.1%) patients, and in 206 (30.3%) patients with average risk, p<0.001. There was no significant difference in the rate of SAs according to type of IBD (UC vs CD (p=0.1) or age (p=0.98). There were no significant statistically differences in the prevalence of SAs and the presence or absence of colic inflammation, 14.5% vs 10.5%,respectively, p=0.43. The majority of patients with IBD took IBD medication (77.1%), most of them with mesalamine. Conclusions: We found a significantly lower prevalence of sporadic adenomas in patients over 50 years old with IBD than in control patients. This may reflect the effect of IBD medications in stabilizing the mucosa. Further studies are needed to corroborate our findings and determine which factors in IBD influence adenoma-carcinoma sequence. P137 Ulcerative colitis in the elderly – Different disease course? C. Teixeira*, C. Martins, S. Ribeiro, D. Trabulo, A.L. Alves, I. Cremers, A.P. Oliveira Centro Hospitalar de Setúbal, Gastroenterology, Setúbal, Portugal Background: The reported prevalence of ulcerative colitis (UC) in patients aged 60 years or older is 15%. Several studies have evaluated the differences between disease course in younger and older patients. Methods: Retrospective analysis of patients with UC followed in an inflammatory bowel disease outpatient clinic, with follow up >2 years. Patients were divided in 2 groups; patients with UC diagnosed before the age of 60 years-old (adult-onset UC) and patients with diagnosis with ≥60 years (late-onset UC). Demographic data, dis- ease extent, colectomy rates, need for hospitalization, treatment and infections were evaluated and compared between the two groups. Statistical analysis performed using SPSS 21, considering statistical significance, p<0.05. Results: 115 patients, 60 males, 29 patients (25%) in the late-onset group. Mean time follow-up was 12,7 years in the adult-onset ver- sus 9,8 years in the late-onset group. There were no differences be- tween the 2 groups regarding family history of IBD (p=0.712) and smoking habits (p=0.193). Regarding disease extent, in the adult- onset group 25% had proctitis, 48% had left-sided disease and 27% had extensive disease versus 28%, 48% and 24% respectively in the late-onset group (p=0.932). Progression of disease extent occurred in 9.6% in the late-onset group and 10.3% in the late-onset group (p=0.912). Colectomy was performed in only 1 patient, with adult- onset UC. There were no differences in the need for hospitalization (29% in the adult-onset; 21% in the late-onset group; p=0.379)or in the corti- costeroids use (57% versus 48% respectively, p=0.315). There was a significant higher use of immunosuppression in the adult-onset group (27%) than in the late-onset group (6.9%), p=0.025. Biologic ther- apeutic was used in 10.6% in the early onset group and 3.4% in the late-onset group but there was no statistical difference p=0.243. There were no differences in the Cytomegalovirus and Clostridium difficile infections (p=0.658; p=0.904, respectively). Conclusions: Although several studies have shown that late-onset UC has a more favorable clinical course, in our series there were no sig- nificant difference in the disease course between the late-onset and adult onset, except for the use of immunossuppression, which was higher in the adult-onset group. P138 Does depression at IBD diagnosis impact on disease outcomes in Crohn’s? C. Alexakis* 1 , S. Saxena 2 , V. Chhaya 1 , E. Cecil 2 , R. Pollok 1 1 St George’s University Hospital, Gastroenterology, London, United Kingdom; 2 Department of Primary Care and Public Health, Imperial College, London, United Kingdom Background: There is growing interest in how psychological distur- bances can impact on disease activity in IBD through neuro-humoral mechanisms, but research in this field is limited and conflicting. The impact of co-existing depression at the time of IBD diagnosis and its impact on subsequent disease activity through IBD medication usage and surgery has not previously been studied. We aimed to investigate whether baseline depression in patients diagnosed with Crohn’s dis- ease (CD) affects the requirement for steroids, thiopurines and intesti- nal surgery using a nationally representative UK research database. Methods: We used the Clinical Practice Research Datalink to iden- tify incident cases of patients with CD between 1998–2014. Demo- graphic data including age at diagnosis, sex, body mass index, social and smoking status were extracted as well as prescription data for IBD medications including 5-aminosalicylic acids (5ASA), corticos- teroids (CS) and thiopurines (TP). Additionally, prescribing data for the most commonly used anti-depressant medications (ADM) were obtained. A patient was considered to have baseline depression if they had either a prescription for ADM or a Read code for depres- sion within the period 6 months prior to and up to 3 months after the formal IBD diagnosis using an established methodology [1]. We com- pared the proportion of patients receiving 5ASA, CS and TP at 6 and 12 months from IBD diagnosis between patients with and without a diagnosis of depression at diagnosis. We generated a Cox regression model to estimate the risk of CS use, TP use and requirement for surgery in patients with baseline depression. Results: We identified 6237 patients with CD in the study period. De- pression at IBD diagnosis was present in 13%. Amongst CD patients with depression at diagnosis there was a higher proportion of females (72% vs 51%, p<0.001), a higher proportion of smokers (39% vs 29%, p<0.001) and a higher proportion of IBS co-diagnosis (25% vs 16%, p<0.001) compared to those without depression. There were no significant differences in medication usage at 6 months between patients with and without baseline depression (5ASA: 45% vs 47%, CS: 19% vs 20%, TP: 17% vs 16%). Similarly, there were no differ- ences in medication usage at 12 months between patients with and without baseline depression (5ASA: 48% vs 50%, CS: 24% vs 25%, TP: 21% vs 22%). In the multivariate regression analysis, depression at baseline was not significantly associated with CS use (HR 0.93, 95% CI 0.72–1.19, p=0.55), TP use (HR 1.13 95% CI 0.94–1.35, p=0.20) or intestinal surgery (HR 0.72 95% CI 0.48–1.10, p=0.13). Downloaded from https://academic.oup.com/ecco-jcc/article-abstract/11/suppl_1/S146/2961082 by guest on 15 June 2020