AGA Abstracts analysis incorporating complete antibody testing, NOD2 status and historical features to identify factors that increase the risk for an earlier second ICR in a cohort of CD patients. METHODS: The cohort comprised 176 CD patients with ileal disease and at least one ICR. Historical data, such as date of diagnosis, date of all ICR's, disease behavior (Montreal classification), medication use, family history of inflammatory bowel disease (IBD), and smoking status were documented in detail and followed prospectively. All subjects were tested for the presence of all commercially available CD-related antibodies (ASCA, OmpC, CBir1, and ANCA) and genotyped for 3 major NOD2 allelic variants (Leu1007InsC, R702W, G908R). We performed Cox regression to determine which factors influence the time to second ICR with results reported as hazard ratios (HR). RESULTS: The median follow-up time was 5 years (range 1-31) and 53 patients (30%) required more than one ICR. In a multivariate model, a positive ASCA IgA was associated with earlier need for second ICR (HR=2.6, p=0.04). Positive or high titers of other antibodies did not influence the risk. Family history of IBD increased the risk of earlier repeat ICR (HR=2.2, p=0.02). The presence of any NOD2 variant did not predict earlier second ICR (HR = 0.8, p=0.5) but use of immunomodulators reduced the risk (HR=0.5, p=0.02). Smoking was associated with need for second ICR on univariate analysis but not in the multivariate model (HR=1.8, p=0.08) which suggests that an interaction with another factor is possible. CONCLUSIONS: A positive ASCA IgA in patients with ileal CD may serve as a biomarker predicting earlier second ICR. Early detection of ASCA IgA may identify a subset of patients who could benefit from more aggressive medical therapy. Patients with a family history of IBD are more likely to require earlier second ICR. NOD2 status does not appear to influence the time to second ICR. In our cohort, after adjusting for antibody and genotype status, smoking does not maintain significance as a risk factor for earlier repeat ICR. W1123 Rigorous Histopathologic Assessment of the Colectomy Specimen for Features of Indeterminate Colitis (IC) Does Not Predict Outcome After Ileal Pouch- Anal Anastomosis (IPAA) Gil Y. Melmed, Hanlin L. Wang, Eric A. Vasiliauskas, Marla C. Dubinsky, Andrew Ippoliti, Dermot P. McGovern, Stephan R. Targan, Phillip Fleshner INTRODUCTION: Indeterminate colitis (IC) is advocated as a classification of chronic inflammatory bowel disease (IBD) involving the colon wherein histopathologic examination of the colectomy specimen fails to identify features favoring typical ulcerative colitis (UC) or Crohn's disease (CD). Multiple features of IC have been proposed in the literature, without generally accepted consensus. In this study, we examined which of these features, if any, are associated with adverse outcomes following IPAA. METHODS: Consecutive patients with UC undergoing IPAA by a single surgeon were prospectively enrolled into a longitudinal cohort. Demographic and clinical data were obtained preoperatively. A checklist of 17 histopathologic features atypical for UC was developed after a comprehensive literature search, discussion, and consensus among a panel comprised of a gastrointestinal pathologist, gastroenterologist, and colorectal surgeon with expertise in IBD. Patients with preoperative typical UC as well as those with clinical features atypical for UC were selected for histopathol- ogic reassessment of the colectomy specimen; this was performed by a single gastrointestinal pathologist blinded to long-term outcomes. Outcomes of acute pouchitis (AP), chronic pouchitis (CP), CD, and the aggregate outcomes AP/CP/CD and CP/CD, were assessed using univariate and time-dependent multivariate regression. RESULTS: Of the 153 study patients, 148 (97%) had at least one feature of IC. These included broad-based ulcers (n=99), appendiceal involvement (n=78), v-shaped ulcer (n=48), crypt-associated granuloma (n= 42), isolated giant cells (n=39), discontinuous active inflammation (n=36), slit-like fissure (n=32), discontinuous chronic inflammation (n=16), ileal villous architectural distortion (n= 12), neural hypertrophy (n=10), backwash ileitis (n=10), transmural inflammation (n=8), discontinuous ileitis (n=8), muscle hypertrophy (n=5), ileal ulcer (n=4) and ileal pyloric metaplasia (n=1). The only feature assessed for but not identified in any specimen was the presence of a granuloma distant from a ruptured crypt. After a median followup of 26 months (range, 1-144) after ileostomy closure, AP, CP and CD were seen in 19%, 11%, and 8% of the study patients, respectively. On univariate analysis, neural hypertrophy was associated with the onset of CD (p=0.01), but this association was not significant on multivari- ate analysis. No other feature was associated with any adverse pouch outcome. CONCLU- SION: Rigorous histopathologic assessment of the colectomy specimen for features of IC in those with colitis does not predict outcome after IPAA. W1124 Management of Female Fertility and Pregnancy Following Restorative Proctocolectomy Julie A. Cornish, Richard Lovegrove, Baljit Singh, Emile Tan, Neil Mortensen, R. J. Nicholls, Sue K. Clark, Paris P. Tekkis PURPOSE: Females of child-bearing age have been reported to have a two to three-fold increase in infertility after restorative proctocolectomy (RPC). This study aimed to assess aspects of infertility and pregnancy. METHOD: A postal questionnaire was sent to 790 females who had undergone primary RPC and who were registered on a pouch database in two tertiary centres in the UK. The following were assessed; infertility, the number of pregnancies, pregnancy outcomes, delivery methods and the use of fertility treatments. RESULTS: Three hundred and six (38.5%) females responded, with a mean age of 44.9 years at follow up and 32.1 years at the time of RPC. Eighty two percent (n=220) had ulcerative colitis. Forty five percent (n=138) had conceived prior to RPC, 5.2% (n=16) conceived both before and after RPC, 5.5% (n=17) conceived after RPC only and 44.1% (n=135) had never conceived. Fifty seven patients had attempted to conceive after RPC over a mean period of 17 months(±168), with 25(45.5%) being successful. Eighteen females had been referred to a fertility specialist; 16 received In Vitro fertilisation (IVF). Four (30.7%) conceived by IVF, one using Clomid and one required a sperm donor to conceive. There was no significant difference in the number of stillbirths (pre-RPC 2.9%, n=4 vs. post-RPC 5.9%, n=1; p=0.551), miscarriages (pre-RPC 12.3%, n=17 vs. post-RPC 23.5%, n=4; p= 0.203), ectopic pregnancies (pre-RPC 1.4%, n=2 vs. post-RPC 0%, n=0; p=0.617) or elective abortion (pre-RPC 1.4%, n=2 vs. post-RPC 5.9%, n=1; p=0.211) between patients conceiving A-638 AGA Abstracts before vs. after RPC. Females delivering before RPC had significantly more vaginal deliveries than those conceiving after (pre-RPC 69.6%, n=96 vs. post-RPC 35.3%, n=6; p=0.001). There was no difference in the number of vaginal deliveries for females conceiving both before and after RPC (pre-RPC 69.6%, n=96 vs. pre and and post-RPC 56.3%, n=9; p= 0.139). Eight (5.8%) females conceiving before RPC had a complicated vaginal delivery, while no patient conceiving before and after or only after had a complicated delivery. CONCLUSION: There was no difference in pregnancy outcomes of females conceiving before or after RPC. While RPC is known to be associated with infertility, only a small proportion of patients are promptly referred for fertility management. IVF success rates after RPC are similar to the general population. W1125 A Study of Quality of Life, Sexual, Urinary and Faecal Function in Females Following Restorative Proctocolectomy Julie A. Cornish, Katherine Wooding, Emile Tan, Ralph. J. Nicholls, Sue K. Clark, Paris P. Tekkis Aim: To investigate quality of life (QoL), sexual, faecal and urinary function in females undergoing restorative proctocolectomy (RPC). Methods: A retrospective case-control study was performed in two tertiary centres. Controls were females with ulcerative colitis or indeterminate colitis, without a stoma or RPC. Validated questionnaires on QoL (SF-36), sexual (FSFI; Female sexual function index), urinary (King's questionnaire) and faecal func- tion (Wexner) were administered in the outpatient setting. Pearson Chi2, t-test and Mann- Whitney U tests were used to assess significance. Results: A total of 255 females were identified and 49% (n=124) recruited. One hundred and nine patients fulfilled the inclusion criteria; 55 (50.5%) IBD, 54 (49.5%) RPC. The mean age of RPC patients was 41.8 years (±12.7) vs. 43.8years (±15.8) for IBD (p=0.491). Mean follow up time was longer for RPC patients' (RPC 5.75±6.5years vs. IBD 10.1±7.5years; p=0.041), but there was no significant difference in smoking (p=0.551), number of females with regular partners (p=0.115) and sexual orientation (p=0.641). There was no difference in the number of females with bladder problems (RPC n=31, 60.8% vs. IBD n=28, 54.9%; p=0.547), however RPC females with bladder problems were 10 years younger than controls (RPC mean age 37.6±7.3years vs. IBD 47.4±13.5; p=0.044). Quality of life scores for overall physical health (RPC median 56.00 ±11.50 IQR vs. IBD 61.00 ±14.00; p=0.020) were worse following RPC, however there was no significant difference for general health (p=0.112), overall mental health (p= 0.714) or social role (p=0.324). Urgency in faecal function was experienced by more IBD patients (IBD 75.0% vs. RPC 47.9%; p=0.006), although RPC patients had increased day (p<0.001) and night bowel frequency (p<0.001) and were more likely to experience night seepage (p=0.001). RPC females who had a vaginal delivery (VD) were more likely to have day seepage (p=0.046) and require pads (p=0.026) than RPC females who had not undergone VD. The number of females who were sexually active was similar between the groups (RPC n=31, 60.8% vs. IBD n=28, 54.9%; p=0.547) and there was no difference in the overall FSFI score (p=0.747) or in the domains of dyspareunia (p=0.713), arousal (p=0.644), lubrication (p=0.912) or orgasm (p=0.511). Conclusion: Restorative proctocolectomy impacts on urinary function and quality of life. Bowel frequency, seepage and pad usage are increased following RPC and faecal function may be worse following vaginal delivery. RPC does not adversely affect overall sexual function. W1126 Cost-Effectiveness of Natalizumab in Crohn's Disease Patients Who Have Failed Anti-TNF Alpha Therapy Bruce E. Sands, Douglas C. Wolf, Sumeet Panjabi, Timothy Niecko, Steven Hass, Loretto Lacey, Mike Spencer OBJECTIVE: To compare the cost-effectiveness (CE) of natalizumab (NAT) to FDA-approved tumor-necrosis factor alpha inhibitors (anti-TNFα) in Crohn's disease (CD) patients who failed previous anti-TNFα treatment. BACKGROUND: Understanding the economics of pharmaceutical care, such as comparative cost-effectiveness, can inform decisions regarding the appropriate use of pharmaceuticals. METHODS: A decision analytic framework was used to model treatment for patients with moderate-to-severe CD (Crohn's Disease Activity Index scores ≥220 and <450). Patients are assumed to have failed treatment with corticosteroids, immunomodulators, and at least one anti-TNFα agent. The model compared NAT 300 mg, infliximab (INF) 5 mg/kg or 10 mg/kg, and adalimumab (ADA) 40 mg dosed every other week (EOW) or weekly (EW). The model includes an induction period followed by a 2- year maintenance phase. At the end of induction and each of the 4 six-month maintenance cycles, patients were assigned to one of 3 efficacy states (remission, response, nonresponse) based on estimates from the published literature and NAT clinical data. Patients entering the nonresponse health state at any point are assumed to remain in that health state for the duration of the model. Resource use such as concomitant steroid use and hospitalization is estimated for patients in each health state. Costs associated with each comparator agent are composed of pharmacy and medical costs derived from published price lists and an analysis of CD claims from a database assembled by Health Benchmarks International. The drug costs for INF and ADA were weighted based upon the distribution observed in phase 4 trials. RESULTS: Over the 2-year maintenance period, it was estimated that NAT patients would be in remission for 0.40 years versus 0.21 and 0.23 for those receiving INF and ADA, respectively. The combined drug and medical cost during induction and 2 years of maintenance were predicted to be $69,977 for NAT, $62,597 for INF, and $60,277 for ADA. NAT was associated with a 12% increase in total cost compared to INF, but resulted in 95%, 85%, and 62% increases in time in remission, time in steroid-free remission, and time in remission or response. NAT was associated with a 16% increase in total costs, but was more effective than ADA in improving these outcomes by 73%, 56%, and 50%, respect- ively. CE ratios were insensitive to increases in NAT-related costs (up to $36,000) or decreases in NAT efficacy (up to -25%). CONCLUSIONS: This model, based on estimates from the available published literature, projected NAT to be the most cost-effective treatment alternat- ive for patients who had failed prior anti-TNFα therapy.