Introduction Due to the different presentations of patients with inflammatory bowel disease, several clinical severity indices were used in the past. Interestingly, most (if not all) of these indices were not properly validated and did not go through a robust methodology. Our aim is to develop a new clinical disease severity index that valid, easy obtainable in the clinic and suitable to all IBD patients. Methods The development of Swansea IBD clinical severity index (SICSI) followed a clinimetric approach. Items were devised using IBD experts opinions and through reviewing 17 clinical severity indices commonly used in studies for UC and CD. To ensure items are applicable, we asked a small focus group of IBD specialists, statisticians and methodologists to review these items and ensure good face and content validity. Psychometric properties were tested on 210 patients to remove redundant items and shorten the index. Construct validity was checked using biochemical markers like CRP, WBC, HB and albumin and clinical indices which are: Harvey Bradshaw index, simple clinical colitis activity index and perianal disease activity index. If patient is having an endoscopy, endoscopic indices will be recorded as well which are mayo clinic score, Rachmilewitz scores and simple endoscopic score. Results We found that 7 items account for 98% of the variance of the total score and they are: Abdominal pain or discomfort, stool consistency compared to the usual, blood in stool, number of stool frequency, general well being, nocturnal symptoms and urgency. Items that had high item total correlation > 0.8 like physician global assessment were removed from the index as they are redundant. Temperature and abdominal mass had a zero variance score and did not add any value to the total score and were removed during factor analysis. Internal consistency (Correlation of items with each other) was acceptable (Cronbach alpha = 0.827). SICSI had good correlation with the clinical, biochemical and endoscopic severity scores (r > 0.5). Conclusion It is clear that the Swansea IBD clinical severity index will perform well in clinical practise. Further studies are going on to implement the index in clincal practise. The index has been incorporated into our local IBD registry to follow up and moniter patients. There are plans to develop an iPhone application.

Introduction Due to the different presentations of patients with inflammatory bowel disease, several clinical severity indices were used in the past. Interestingly, most (if not all) of these indices were not properly validated and did not go through a robust methodology. Our aim is to develop a new clinical disease severity index that valid, easy obtainable in the clinic and suitable to all IBD patients. Methods The development of Swansea IBD clinical severity index (SICSI) followed a clinimetric approach. Items were devised using IBD experts opinions and through reviewing 17 clinical severity indices commonly used in studies for UC and CD. To ensure items are applicable, we asked a small focus group of IBD specialists, statisticians and methodologists to review these items and ensure good face and content validity. Psychometric properties were tested on 210 patients to remove redundant items and shorten the index. Construct validity was checked using biochemical markers like CRP, WBC, HB and albumin and clinical indices which are: Harvey Bradshaw index, simple clinical colitis activity index and perianal disease activity index. If patient is having an endoscopy, endoscopic indices will be recorded as well which are mayo clinic score, Rachmilewitz scores and simple endoscopic score. Results We found that 7 items account for 98% of the variance of the total score and they are: Abdominal pain or discomfort, stool consistency compared to the usual, blood in stool, number of stool frequency, general well being, nocturnal symptoms and urgency. Items that had high item total correlation > 0.8 like physician global assessment were removed from the index as they are redundant. Temperature and abdominal mass had a zero variance score and did not add any value to the total score and were removed during factor analysis. Internal consistency (Correlation of items with each other) was acceptable (Cronbach alpha = 0.827). SICSI had good correlation with the clinical, biochemical and endoscopic severity scores (r > 0.5). Conclusion It is clear that the Swansea IBD clinical severity index will perform well in clinical practise. Further studies are going on to implement the index in clincal practise. The index has been incorporated into our local IBD registry to follow up and moniter patients. There are plans to develop an iPhone application.

Disease and 10% had an alternative form of IBD (e.g. Proctitis, Lymphocytic Colitis or Collagenous Colitis). The ethnic mix in the responding cohort was 91% Caucasian, 6% Asian, 2% Mixed and 1% was not stated. The sample had a mean score of 7.8 (CI = 7 -8.6). 98 (40%) of patients' scores reflected "no depression"; 64 (26%) reflected "mild depression"; 33 (14%) reflected "moderate depression"; 36 (15%) reflected "moderately severe depression"; 12 (5%) of scores reflected "severe depression". Conclusion 20% of our responding IBD patients were shown to have clinically significant levels of depression (moderately severe + severe), with 5% demonstrating scores suggestive of severe depression (1% expressing suicidal ideation). Relapse rates are known to be closely correlated with the severity of depression, and yet very few are on active treatment or review for this. The prevalence and severity of depression in our cohort of responding IBD patients supports the argument for screening all new IBD patients in order to optimise clinical well-being and treatment efficacy.

provided stool for FC concentration analysis and the study was terminated once the last recruited patient reached a follow up period of 365 days. Remission was defined as a Crohn's disease activity index (CDAI) of < 150. Relapse was defined as either a need for escalation of medical therapy, surgery for active CD or progression of disease phenotype using the Montreal classification. The study was approved by the West of Scotland Research Ethics Service (REC reference 10/ S0704/1). The Receiver Operating Characteristic (ROC) curve of relapse by 12 months, based on FC value at baseline, was calculated. Kaplan-Meier curves of time to relapse, some of which were longer than 12 months, were based on the resulting best FC cut-off value for predicting relapse (with patients who had not relapsed being censored at end of follow-up) and compared using the log-rank tests. Results 98 patients were recruited. One patient was lost to follow up, 1 died and the care of 3 patients was transferred to another centre, before either relapsing or being followed up for 12 months. Of the 93 remaining patients 11 (12%) had relapsed by 12 months. The median FC was lower for non-relapsers, 96 µg/g (IQR 39-237), than for relapsers, 328 µg/g (IQR 189-574), (p = 0.008). The area under the ROC curve to predict relapse using FC was 74.8% ( Figure 1). A cut-off FC value of 240 µg/g to predict relapse of quiescent Crohn's disease over the course of one year was associated with a sensitivity of 72.7% and specificity of 74.3%. Negative predictive value was high at 95.3% and positive predictive value was 27.6%. There was a significant difference in time to relapse for those with the first FC value below or above 240 µg/g (p = 0.011). Conclusion In this prospective dataset, FC appears to be a useful, non-invasive tool to help identify quiescent Crohn's disease patients at a low risk of relapse over the ensuing 12 months. A FC value of 240 µg/g was deemed the best cut-off value in our patients. Disclosure of Interest None Declared. THIOPURINE WITHDRAWAL FOR SUSTAINED REMISSION Gastroenterology, Glasgow Royal Infirmary, Glasgow, UKIntroduction Thiopurine therapy is effective in maintaining clinical remission in IBD. However, long-term therapy is associated with an increased risk of lymphoma; therefore in clinical practise it may be appropriate to withdraw thiopurines after prolonged remission. Nevertheless, many patients will experience disease relapse within 12 months of drug withdrawal. The Aim of the present study was to retrospectively determine the relapse rate in ulcerative colitis (UC) and Crohn's disease (CD) following azathioprine (AZA) or mercaptopurine (MP) withdrawal and to determine factors predictive of relapse. Methods Patients were identified by electronic case note review of IBD patients in eight major centres around the United Kingdom. Major inclusion criteria were AZA and/or MP therapy for a minimum of 3 years, AZA/MP withdrawn due to sustained clinical remission no steroid therapy for 6 months prior to drug withdrawal, and minimu...

No abstract