The Ileal Pouch Database |
 |
|
|||||||||
About the Ileal Pouch Database Project: Development of the Association of Coloproctology of Great Britain and Ireland Introduction:Restorative proctocolectomy (RPC) has become the treatment choice for most patients with ulcerative colitis and familial adenomatous polyposis who require surgery. Despite the associated complications related with RPC, the majority of patients have a successful outcome, with ileal pouch failure and complications occurring in 5 to 10% of patients in large series. As surgery is moving into an era of objective assessment and professional accountability for clinical outcomes, increasing emphasis is being placed on the clinical effectiveness and quality of health care provision. The proposed project aims to:
Study endpoints: Ileoanal pouch failure, defined as excision of the ileo-anal pouch, formation of a permanent colostomy or pouch related mortality during the follow up period. Ileal pouch-related complications: classed as septic and non-septic adverse events. Functional outcomes: Bowel movements per 24 hours / at night, urgency, pad usage, incontinence. Study design and target population: Multi-centre, cohort study with prospective and retrospective data collection, patient questionnaire survey (future work) and case-note review. All patients undergoing RPC for inflammatory bowel disease or familial adenomatous polyposis coli. Methodology:Phase 1: Development of The St Mark’s ileal pouch database. All patients operated between 1983 to the present date (n~1400) have been identified. Patient-related information, intra-operative factors and post-operative events have been extracted from patients’ records and entered in a newly developed registry with data fields based on the Cleveland Clinic Ileal pouch registry. Phase 2: Development of the ACPGBI ileal pouch registry. An extended and minimum dataset for patients undergoing IPAA surgery has been proposed and a dedicated database is currently being piloted to that effect. The newly developed database will be freely provided via a dedicated website to all ACPGBI members during the Annual Scientific Meeting in Birmingham, June 2004. To the present time, thirty hospitals have shown interest to join the ACPGBI ileal pouch registry using existing and future patients. It is estimated that a total of 3000 patients should be collected from existing databases in the UK, with approximately 300-400 new cases added to the ACPGBI registry per year. Data will be collected locally and custom-made extraction tools used for the electronic forwarding of information to a central server. The information on each patient will be held on a secure database and will be periodically updated by queries from the source hospitals or by direct patient contact (for example quality of life questionnaires – future work). Data entry, cleaning and checking will be carried out by trained database administrators.Risk factors - patient related: age, gender, ASA grade (I-V), individual comorbid conditions (cardiac, respiratory, renal etc), extra-intestinal manifestations of IBD (cutaneous, skeletal, hepatobiliary etc), manometric studies (squeeze pressures, resting pressures), provisional diagnosis (FAP, Indeterminate colitis, Crohn’s disease), disease distribution, immunosuppressive therapy, steroid therapy, pre-operative investigations (Haemoglobin, white cell count, albumin, total protein), previous anal symptoms. Factors specific to the surgical intervention: staged procedures, previous colectomies, type of anastomosis, proximal diversion, adhesiolysis, blood transfusion, type of pouch (J,S,W). Post-operative factors: Fistula formation, type of fistula (abdominal wall, perineal, vaginal, pre-sacral), peri-operative bleeding, source of bleeding, anastomotic leak, septic episodes, anastomotic stricture, obstructive episodes, pouchitis (intermittent and chronic). Statistical analysis: Unifactorial analysis will be performed by cox-regression in order to identify possible factors related to adverse outcomes. Significant factors along with factors of clinical importance will be entered into a multi-factorial cox-regression model in order to identify independent predictors of outcome. Parametric survival models will be constructed to identify the best function that characterises the patient population. The models will be internally validated by randomly splitting the population into two groups, one group (70%) will be used for developing of the model and the remaining patients (30%) will be used for validating the model and the observed and expected outcomes compared using sub-group analysis. Learning curve for IPAA surgery: Cumulative risk-adjusted pouch survival charts will be constructed for evaluating the learning curve for IPAA surgery. Using the parametric survival model the cumulative observed minus expected pouch failure rate will be plotted sequentially (6) and control limits will be constructed to evaluate surgical outcomes over time (7). In order to establish the learning curve for IPAA surgery, control charts will be based on all surgeons as a group by plotting, the median “performance ratio” with no individual surgeon being identified. Sample size: Based on a failure rate of 4.1% (Cleveland clinic) and a sample size of 2000 patients the multi-factorial model should be able to detect odds ratios >1.3 or <0.75 at 80% power and 5% statistical significance. Based on a provision of ten adverse events per covariate a maximum of 8 factors (+/-) their interaction terms can be entered into the multi-factorial model. Ethical Considerations:Ethical approval has been obtained for Phase 1 of the study which is near to completion. MREC approval will be attained for prospective data collection but all retrospective data can be aggregated as it is classed as Category D (epidemiological study). Informed written consent will be obtained from each new patient who provides information on functional outcomes and quality of life measures. The right of any patient to refuse participation without giving reasons will be respected. Patients will remain free to abstain from follow-up at any time, without prejudicing his/her further treatment at the local hospital. In order to maintain patient confidentiality, all patient identifiers will be encrypted. Finance:There is no financial incentive to the contributing centres for entering patients into the present study. A grant application is currently being made to provide financial support for the project and the central server, which will run in parallel with the ACPGBI colorectal cancer database. About the Study Investigators:Paris Tekkis, MD FRCS Professor John Nicholls MChir, FRCS About the website:The website has been setup and produced by Jason Smith & Paris Tekkis to facilitate communication amongst users of the Ileal Pouch database. The site is being rapidly developed and should be fully fuinctional following the June 2004 ACPGBI meeting in Birmingham. If you have any comments please use our contact form or the discussion forums. Useful links: Users of the website may find the following useful:
|
Disclaimer This website is affiliated with The Association of Coloproctology of GB & I - www.acpgbi.org.uk |
|
Copyright © 1998-2004 Surginet. All rights reserved. |
|
