$1326 A Comparative Study to Assess the Diagnostic Value of Transabdominal Ultrasound and Magnetic Resonance Imaging in Patients with Inflammatory Bowel Diseases Maria Pascu, Arthur B. Roznowski, Hans-Peter Mueller, Andreas Adler, Bertram Wiedenmann, Axel Dignass Background: lleocolonoscopy still represents the diagnostic gold standard in the work-up of patients with inflammatory bowel diseases (IBD). Because of the necessity to clean the colon extensively and the invasive and often painful procedure, patients are often reluctant to be colonoscoped. The purpose of this study was to evaluate the usefulness of transabdominal ultrasound (US) and magnetic resonance imaging (MR1) to assess disease extension and seventy of the terminal ileum and colon in patients with IBD. Methods: Seventy consecutive patients with confirmed IBD (43 with Crohu's disease and 27 with ulcerative colitis) were included. Patients underwent clinical assessment, laboratory testing, ileocolonoscopy, US, and MRI within 5 consecutive days. Involved bowel segments were defined as those with bowel wall thickness > 3 mm assuming an increased Doppler signal on US or moderate contrast enhancement of the bowel wall on MRI. Endoscopy served as diagnostic gold standard. Severity of disease was graded wath newly developed scores for endoscopic, MRI, and US findings. Results: A segment-by-segment analysis revealed an overall accuracy of 89% for US and 73% for MRI in identifying active IBD lesions. The accuracy was higher in patients with ulcerative colitis (UC) compared to patients with Crohn's disease (CD) for both US and MRI (95% vs. 85% and 81% vs. 67%, respectively). The US activity index showed a strong correlation with the endoscopic activity index (EAI) (r = 0.884, p < 0.0009), whereas the MRI activity index did not correlate with the EA1 (r = 0.344, p = 0.007). Correlation of EAI with US and MRI activity index was better in patients with UC (r = 0.974, p = 0.0009 and r = 0.519, p = 0.009, respectively) than in patients with CD (r = 0.830, p < 0.0009 and r = 0.120, p = 0.482, respective). The extension and severity of inflammation as assessed on US and MRI varied according to location of affected bowel segment. All three imaging methods showed a significant correlation with clinical disease activity index in patients with UC but not in patients with CD. In addition, EA1 and US activity index showed a significant correlation with C-reactive protein in patients with UC. Conclusion: Trausabdominal ultrasound provides a reliable, cheap and safe tool to assess disease extension and activity in patients vath IBD and should be considered as a first-choice method for the follow up of patients with confirmed diagnosis of IBD. $1327 Fecal Anti-Saccharomyces Cerevisiae Antibodies (ASCA) - Evaluation of a New Diagnostic Test in Pediatric Patients with Inflammatory Bowel Disease Stephan Buderus, Helma Ringelmann, Michael J. Lentze, James Boone, David Lyerly Background: The detection of anti-saccharomyces cerevisiae antibodies (ASCA) in serum has been shown to be a reliable marker for differentiating Crohn's disease (CD) from ulcerative colitis (UC). Recently, the presence of ASCA including lgG, IgA and secretory IgA was measured in the feces of adult patients with inflammatory bowel disease (IBD) as an indicator of CD (Boone et al., Gastroenterology 2002,97,$253). Sensitivity and specificity was similar to serology. Aim: To evaluate the presence of ASCA in matching serum and fecal samples of pediatric patients with already classified inflammatory bowel disease. Methods: A total of 23 [BD patients comprising 10 Crohn's disease (CD), 12 ulcerative colitis (UC) and 1 indeterminant colitis tiC), were evaluated for fecal and serum ASCA. The gender ratio was 17 male to 7 female and the age range was 4 to 18 years. Matching serum and fecal samples were collected from each patient and stored frozen until testing. ASCA was detected in both serum and feces by ELISA using the ASCA-CHEK~M(TechLab, Inc.) and in serum only using the QUANTA Lite TM ASCA igG and lgA ELlSA tests (INOVA Diagnostics). In the ASCA- CHEK TM fecal and serum samples were diluted 1:20 and absorbance cut-offs of 0.2 and 1.5 were used, respectively. Results: A total of 4 CD and 1 IC had both serum and feces positive for ASCA by ASCA-CHEK TM, showing a mean absorbance of 1.659 and 0.391, respectively. In serum, 7 CD were positive by the ASCA-CHEKrMand QUANTA Lite TM IgG and IgA ELISA tests. The single IC patient was positive by all tests except the QUANTA LiterMIgA ELISA Test. Conclusions: A) These first results in pediatric patients with IBD indicate that the determination of fecal ASCA is possible and yields comparable results to serum testing. Fecal determination of ASCA is a non-invasive test that may prove to be especially useful in children who are screened for IBD. B) Although designed for use with feces, the ASCA- CHEK TM also works with serum, showing similar results as an established serum test. However, optimal dilutions and cut offvalues in the serum system will have to be determined. S1328 Incidence and Phenotypic Associations of Novel Antibody Markers in Crohn's Disease fan D. R. Arnott, Carol J. Landers, Stephan R. Targan, Jack Satsangi Introduction: The diagnostic value and pathogenic importance of serological markers in patients with inflammatory bowel disease remain under evaluation. Antibodies directed against Saccharomyces cerevisiae (ASCA) and the perinuclear component of neutrophils (pANCA) have been the most widely studied and may be of diagnostic importance. OmpC and I2, antibodies directed against luminal bacterial components (E. Coli and pseudomonas respectively'), have recently been described in Crohn's disease (CD). The data reported in the Los Angeles population suggest a selective loss of tolerance to microbial antigens in CD. Phenotypic associations of these antibodies are unknown. We aimed to assess the frequency of ASCA, pANCA, OmpC and 12 in an independent CD cohort and establish phenotypic associations. Methods: 141 well-charactensed CD patients (76 females, median age 39years (17-88)) and 78 healthy controls (HC) were studied. 36 had ileal disease, 58 had colonic disease and 57 had both ileal and colonic disease. 42 had associated perianal disease. 73 had undergone previous surgery. ELISAassayed ASCA, ANCA, OmpC and 12. The pattern of ANCA posiuvity was assessed by immunoflouresence. Results: All antibodies were more prevalent in CD than HC (ASCA 39% v 4%, ANCA 14% v 1%, 12 52% v 10%, OmpC 37% v 1%, all p<0.001). A positive serum ASCA was associated with ileal disease (76% v 42%, p<0.001) and previous surgical resection (73% v 37%, p<0.001) where as pANCA was a marker for colonic disease (63% v 30%, p<0.05). OmpC and 12 were associated with increasing age, increasing disease duration and the need for intestinal resection (p<0.001) OmpC is also associated with ileal disease (p<0.01). Patterns of antibody positiwty were identified by cluster analysis. Conclusion: We have confirmed the previous association of ASCA and ileal disease and ANCA and colonic disease. OmpC is also associated with ileal disease and OmpC and 12 are associated with increasing disease duration $1329 Faecal Calprotectin In Steroid Dependent Colitis: An Indicator of Steroid Response? Robert J. Atkinson, John Hunter Introduction: Corticosteroids are widely used in inflammatory bowel disease to achieve remission. Whilst fraught with conrplicatinns, a significant number of patients require long term therapy to maintain a symptomatic response. Faecal calprotectin (CPT), a leukocyte derived protein, has been shown to correlate with disease activity and predict the likelihood of future relapse. We present data from a study of Curcuma extract (PYMSO014) as a potential steroid sparing agent, assessing the effect of PYM50014 on CPT concentrations and the potential of CPT as a disease marker. Methods: 27 patients with steroid dependent colitis were randomly allocated to receive PYM50014 3x220mg b.i.d. (Phytopharm plc), or placebo for 16 weeks. All patients were in remission as determined by the Ulcerative Colitis Index (UCI) or Crohn's Disease Activity Index (CDAI) and on stable doses of prednisolone. CPT excretion was measured (Calprest, Eurospital) at randomisation, 8 and 16 weeks. Each patient's steroid dose was reduced every two weeks until relapse or cessation of the drug. All patients were monitored to determine whether or not their disease relapsed for 6 months after the start of the 16 week treatment period. Results: Median CPT excretion at randomisation was significantly lower in those patients who were to remain in remission (median 107.8mcg/g, (9.16 - 437.51) compared to those who were to relapse (median 590.54mcg/g, (22626 - 903.18) p = 0.009; Mann Whitney U test). Kaplan Meier time to relapse curve showed that those patients with excretion greater than 250mcg/g were signifi- cantly more likely to relapse (p = 0.0082, log rank). There were no differences in UCI, CDAI or serological parameters between the two groups. Discussion: CPT may differentiate between patients with a merely symptomatic response to corticosteroids and those with genuine mucosal healing. Failure to reduce CPT sufficiently may indicate the need for a trial of a different therapy. CPT excretion deserves further investigation as an index of therapeutic response in inflammatory bowel disease $1330 Growth Parameters Do Not Improve in the First Six Months following Infliximab Infusion Jonathan E. Markowitz, Petar Mamula~ Melissa A. Shepanski, Robert N. Baldassano Background: Crohn's disease (CD) in children is characterized by global derangements in growth that are multi-factorial. Monitoring a therapy's effects on growth in pediatric patients gives insight to its overall effectiveness. We sought to evaluate growth in pediatric patients with CD before and after receiving infliximab. Methods: Height and weight measurements were obtained 6 months before a patient's first lnfliximab infusion, at the time of the first infusion, and approximately 6 months after the first infusion Further follow-up points were obtained when possible. Age-adjusted Z-scores were calculated for height (HAZ), and weight (WAZ). Data were compiled through assessment at the time of infliximab infusion and through chart review. Results: Data were available on 60 patients (37 male/23 female, median age 15.7 years). Calculations were made usnig the non-parametric Wilcoxon Signed-rank test for paired data. Patients demonstrated abnormal HAZ (-0.93) and WAZ (-0.70) at the basehne measurement. Both parameters (HAZ -0.99, WAZ -0.76) worsened by the time of the first infliximab infusion (median ]62 days later). Approximately 6 months following infliximab (median 183.5 days), HAZ (-1.00) and WAZ (-0.75) were essentially unchanged, still below the scores 6 months prior to inflixamab. A subsequent measurement was available in 48 patients approximately one year after the initial infliximab infusion (median 375 days), with HAZ (-0.92) remaining below baseline while WAZ (-0.61) returned above baseline levels. None of the changes met statistical significance. Conclusions: Treatment with nifliximab did not result in positive linear growth up to 1 year after instituting therapy. Weight did not improve within 6 months of instituting therapy, and showed minimal (not statisucally significant) improvement 1 year after the initial infusion. Further evaluation of infl/xlmab's effects on CD, with particular attention to growth parameters, is necessary in the pediat- ric population. $1331 Endoscopic Healing of Crohn's Ileo-colitis with Azathioprine Jean F. Colombel, Marc Lemann, Yoram Bouhnik, Bernard Duclos, Jean C. Soule, Eric Lerebours, Jean Y. Mary, The Getaid Background/Aim: Data on mucosal healing with immunomodulator agents in Crohn's disease (CD) are scarce. Complete absence of ulcers has been reported in 73% of patients with colonic lesions and 58% of patients with ileal lesions after one-year maintenance therapy with azathioprine (AZA) (1). We report GETAID experience at 5 years. Patients/methods : The study was part of a previously reported randomized AZA withdrawal trial in CD (2). CD patients who were in clinical remission induced by AZA with 0-10 mg of steroids over the last 42 months were randomized to receive either AZA at the same dose as previously, or an equivalent number of placebo tablets for 18 months. Colonoseopy was performed at inclusion in 45 patients (24 F, median age 37 yrs). Disease location was small bowel (n = 3), colon (n = 23) or both (n = 19). Median duration of remission on AZA was 61 months (42-197) with a median AZA dose of 1.7 mg/kg/day (0.4-2.7). Correlation between endo- scopic lesions and CDAI or biological parameters was assessed using Spearman correlation AGA Abstra'cts A-196