Inflammatory bowel disease: a comparison of demographic and clinical characteristics between caucasians and ethnic minorities

S254

Abstracts

of intestinal barrier integrity which is a crucial parameter in the pathogenesis of IBD and its flare-up. Further studies are needed to see if antioxidant could prevent this stress induced tissue oxidation and whether antioxidants can benefit subjects with IBD particularly those who experience flare-up after stressful events. 765 INFLAMMATORY BOWEL DISEASE RELATED OSTEONECROSIS Gregory G. Klingenstein, B.A., Asit K. Shah, M.D., Roger N. Levy, M.D., Daniel H. Present, M.D., M.A.C.G.*. Mount Sinai School of Medicine, New York, NY. Purpose: Osteonecrosis is the death of cellular bone components due to occlusion of medullary blood supply, by trauma or disease process. The use of high-dose corticosteroids in the treatment of a variety of diseases has long been implicated as a cause of multi-focal osteonecrosis (avascular necrosis, ischemic necrosis). Corticosteroids have been prescribed in order to reduce inflammation in patients with Crohn’s disease and ulcerative colitis (inflammatory bowel disease). Limited studies in the current literature suggest that IBD predisposes patients to steroid-induced osteonecrosis. Our study examines a large patient population in order to test the hypothesis that IBD patients receiving only moderate doses of corticosteroids may suffer from osteonecrosis. Methods: A total of 23 charts, from the practices of five gastroenterologists and an orthopaedic surgeon were reviewed. We recorded the peak dose, calculated the mean dose, cumulative dose, and the number of courses of medication. Clinical orthopaedic evaluation included number of joints involved, ARCO classification of hip disease, type of procedure performed, and outcome. Results: All IBD patients with osteonecrosis received corticosteroids, the mean dose was 20.6mg, moderately less than a similar SLE study (25mg) or another IBD study (26mg). The mean cumulative lifetime dose of prednisone was 15,571 mg and each patient received treatment for an average of 25 months. The most common joints affected with osteonecrosis were hips (35), followed by knees (10), and shoulders (8). Conclusions: When compared to other conditions requiring steroids, osteonecrois in IBD patients seemed to be induced by lower doses of prednisone for shorter periods of time. Furthermore, patients who were treated with less than 20mg per day of prednisone had milder stages of osteonecrosis of the hip. When corticosteroid therapy at even low doses is selected for the treatment of IBD, a high index of suspicion for early diagnosis and management of osteonecrosis is recommended. 766 IS MUCOSAL MAST CELL IN SUBJECTS WITH INFLAMMATORY BOWEL DISEASE DIFFERENT FROM HEALTHY CONTROLS? Ashkan Farhadi, M.D., Eva U. Sotil, M.D., Maliha Sheikh, M.S., Ali Banan, Ph.D., Ali Keshavarzian, M.D.*. Rush Medical college, Chicago, IL. Purpose: In vitro and animal studies showed that Mast Cells (MC) are critical element of inflammatory cascade. However, the role of MC in IBD is still controversial. One possible reason for this controversy is lack of standard method to assess the size of gastrointestinal mucosal MC population and activation. Membrane bound receptor CD 117 (C-Kit) is a growth factor receptor that bond with Stem Cell Factor (SCF). It appears on MC as it matures from stem cell. Tryptase is one of the MC granule contents, which could be stained within or outside of MC. Using these two targets, we stained and assessed colonic mucosal MC in subjects with IBD and compared it with the value from healthy controls. Methods: Nineteen subjects (6 Controls, 6 with inactive UC 7 with inactive CD) were studied. Sigmoid biopsies were taken during unprepared sigmoidoscopy. Biopsies were stained using two immunohistochemical techniques mentioned above to stain MC granules by mouse monoclonal

AJG – Vol. 98, No. 9, Suppl., 2003

anti-tryptase antibody (ATA) and MC membrane by monoclonal antibody against CD-117. A total of 9 random distinct fields, which were completely covered by transversely cut glands, were selected for MC counting and the mean value was considered as the number of MC per HPF. The number of MC obtained by two methods of staining was compared using paired t-test. Results: In healthy controls the number of MC was the same using either method of staining [mean MC/HPF was 21 and 16 using ATA and CD117 method, respectively (p⫽ns)]. In contrast the number of MC was significantly higher using ATA staining in subjects with inactive IBD [mean MC/HPF was 19 and 5 using ATA and CD117, respectively (p⫽0.001)]. Conclusions: 1) MC membrane receptor and granule staining provide similar result in healthy subjects but dissimilar results in subjects with IBD. Thus methods of MC assessment need to be validated in both health and disease status. 2) This discrepancy in MC count in subjects with IBD might be explained by: over counting of the packs of granules as MC in IBD, due to higher rate of MC degranulation; absence of c-kit receptor on the mucosal MC, as an indicator of increased number of immature MC due to higher turnover of MC in mucosa; diverse population of the MC in IBD subjects (like brain MC which are C-kit negative); internalization of c-kit receptor due to hyperstimulation of this receptor as a negative regulatory mechanism. Whether difference in CD117 receptor has any significance in the pathogenesis of IBD remains to be investigated.

767 INFLAMMATORY BOWEL DISEASE: A COMPARISON OF DEMOGRAPHIC AND CLINICAL CHARACTERISTICS BETWEEN CAUCASIANS AND ETHNIC MINORITIES Robin B. Forman, D.O., Nicole M. Gordon, M.D., Georgia Panagopoulos, Ph.D., Burton I. Korelitz, M.A.C.G.*. Beth Israel Medical Center, New York, NY and Lenox Hill Hospital, New York, NY. Purpose: Experienced IBD clinicians have been impressed by the relative virulence of Crohn’s disease (CD) and ulcerative colitis (UC) in African Americans. This has never been scientifically confirmed and still less is known about other ethnic groups compared to Caucasians. Methods: We reviewed our IBD database at Lenox Hill Hospital of ⬎1200 patients and identified 37 minorities (36% African Americans, 36% Hispanics, 28% Asians) and matched them with 37 White patients for age, gender, and disease type. Their charts were reviewed and the two groups were also compared for similarity of professional status, medical insurance, disease location and manifestations, number of hospitalizations, treatment, appendectomies and duration of follow-up. Results: Each group was comprised of 12 men and 25 women, 26 with CD and 11 with UC. The distributions of professional status and medical insurance were similar, (p⫽0.68, p⫽0.77, respectively). The two groups were similar with respect to age of symptom onset (p⫽0.45), disease behavior (p⫽0.75) and treatment modalities (p⬎ 0.17). Table 1 presents the results of comparisons on other disease and clinical characteristics. Caucasians with CD had more colitis and ileocolitis whereas ethnic minorities had more upper GI involvement and ileitis alone (p⫽0.037). Disease location in patients with UC approached statistical significance (p⫽0.056), with disease localized to the left colon in minorities and white patients experiencing more universal disease. There was also a strong tendency for minorities to have more appendectomies (13/37) than whites (6/37) (p⫽0.055). Table 1. Caucasians Ethnic Minorities median (range) median (range) p- value Age of symptom onset (years) Lag time to diagnosis (months) Number of hospitalizations Duration of follow-up (months) Extraintestinal manifestations

24 (3-55) 5 (1-192) 2 (0-10) 80 (1-343) 1 (0-4)

25 (1-52) 23 (1-288) 3 (0-16) 57 (1-439) 0 (0-4)

0.45 0.05 0.02 0.64 0.02

AJG – September, Suppl., 2003

Conclusions: The two groups had equivalent sociodemographic characteristics, but were different in location of disease, lag time to diagnosis, number of hospitalizations, extraintestinal manifestations, and incidence of appendectomies. Since these are preliminary analyses, the addition of more patients will allow for the examination of ethnic groups separately in anticipation of finding new genetically controlled features.

768 INTRAVENOUS CYCLOSPORIN IN ULCERATIVE COLITIS: LONG-TERM FOLLOW-UP OF THE UNIVERSITY OF CHICAGO EXPERIENCE Phillip Y. Chung, M.D., Russell D. Cohen, M.D.*, Barbara S. Kirschner, M.D., Stephen B. Hanauer, M.D., FACG. University of Chicago, Chicago, IL. Purpose: We reported and published our initial open-label 5 year experience with intravenous cyclosporin (CSA) in the treatment of steroidrefractory ulcerative colitis (UC) at ACG 1996 (Am J Gastroenterol 1999: 94(6):1587). We now present long-term follow-up on the same 42 patients, to discern whether CSA is truly a surgery-sparing option. Methods: The electronic and paper charts of the original 42 patients were reviewed. Clinical course, medication use, surgeries, and adverse events were recorded. Patients who initially received CSA but eventually went to colectomy were referred to as “surgical” patients; those who retained their colons as “nonsurgical”. Results: The 42 patients [60% male, mean age 34 (range 11– 67)] received a total of one (n⫽35), two (6), three (1), or four (1) courses of IV CSA (4mg/kg/day). The mean duration of IV CSA was 10.3 days (⫾5.4), at a mean level of 379 (⫾126) ng/ml (HPLC). Thirty-one patients were continued on oral CSA for a mean of 17.9 weeks (⫾23.2); mean level 224 (⫾123) ng/ml. We previously reported an initial response in 36/42 (86%), with 10 additional patients undergoing colectomy (median 18 weeks, range 3-79). Of the 26 (62%) whom originally avoided colectomy, 18 (43% of the initial total) still had their colons after a median 6.7 years (0.1–11.8). Fourteen of 18 (78%) non-surgical patients had received 6-MP or AZA (mean 64.4⫾42.1 months), vs. 14/24 (58%) surgical patients (mean 28.3⫾32.5 months) (p⫽0.21 Chi-square). Life-table analysis predicted 12 year “noncolectomy survival” rates of 43% for all patients, 54% for those treated with 6-MP/AZA, and 17% for those who did not receive 6-MP/ AZA, respectively (p⫽0.05 6-MP vs. none; Mantel-Haenszel chi-square). Only 1 of 8 patients requiring ⬎1 course of IV CSA has avoided colectomy. Complications, resulting in CSA discontinuation in 6 patients, were all reversible, with complete recovery.

Abstracts

S255

769 EFFECTS OF TOBACCO AND ALCOHOL ON THE CLINICAL COURSE OF CROHN’S DISEASE Gordon F. Williams, M.D., Mahesh Tipirneni, M.D., John O’Brien, M.D.*. Southern Illinois University, Springfield, IL and Michigan State University/Kalamazoo Center for Medical Studies, Kalamazoo, MI. Purpose: Smoking is a well-known risk factor for Crohn’s disease (CD). In the aggregate, the negative effects of smoking on CD include more relapses, increased complications, and the need for more surgeries. However the role that tobacco plays on various behaviors or patterns of CD (luminal, fistulizing, stricturing) is not known. In addition, to date the role of alchohol (ETOH) on the relationship of smoking and CD is confounding. This ongoing study examines the role of tobacco and ETOH on various patterns (luminal, fistulizing, stricturing) of CD. Methods: In this ongoing retrospective analysis (chart review) of 80 CD patients (patients), each case was categorized by CD pattern (luminal, fistulizing, or structuring behaviors) and tobacco (& ETOH) useage. Twenty-nine of the 80 patients had a non-luminal CD pattern (i.e. stricturing & / or fistulizing behaviors) and were termed “progressive disease”. The average duration of their illness was determined. Patients who had never used ETOH or tobacco were used as the control group. Results: The table shows the number of patients who have fistulizing and/or stricturing type Crohn’s disease and the average disease duration broken down by Tobacco/Alcohol Usage. Effects of Alcohol and Tobacco on Progression of Crohn’s Disease

Alcohol/Tobacco Usage

Number of Patients

Average Duration of Illness

Not Current Smokers/Current ETOH drinkers Current Smokers/Current ETOH drinkers Current Smokers/Not Current ETOH drinkers Not Current Smokers/Not Current ETOH drinkers

8 3 5 13

13 years 10 years 9 years 13 years

Conclusions: Although the number of patients to date studied has been limited, the data suggests that smoking results in an accelerated development of progressive CD patterns (9 –10 yrs) than in non-smokers (13 yrs). In addition, ETOH has little if any effect on the progression of CD in smokers or non-smokers. Larger clinical studies will be necessary to confirm these trends. It is also recommended that future studies look specifically at the respective outcomes of fistulizing vs structuring behaviors of CD (with regard to tobacco and ETOH usage) and not condense these two categories together.

770 ANTI-TUMOR NECROSIS FACTOR THERAPY FOR PATIENTS WITH REFRACTORY ULCERATIVE COLITIS Camron Kiafar, D.O., Francisco C. Ramirez, M.D., Neil Shernoff, M.D.*. Carl T Hayden VA Medical Center, Phoenix, AZ and CIGNA Healthcare, Phoenix, AZ.

Conclusions: Short-term CSA followed by maintenance therapy with 6MP/AZA allows more than 50% of patients with severe, steroid-resistant UC to avoid colectomy long-term. Retreatment with CSA is rarely successful.

Purpose: To determine the clinical efficacy of infiximab in patients with UC not responding to conventional therapy. The monoclonal antibody to TNF-alpha, infiximab is widely and effectively used for the treatment of Crohn’s disease. However, data regarding its clinical utility in ulcerative colitis (UC) is limited. Methods: Medical records of patients with refractory UC treated with infliximab in the outpatient setting as last medical resort were reviewed and analyzed. Diagnosis of UC was made by clinical, endoscopic and histologic findings. Response to treatment was defined as loss of diarrhea, hematochezia, abdominal pain and decrease or discontinuation of corticosteroids. Patients received infiximab at 5 mg/kg as single or multiple infusions. Setting: CIGNA heatlh care.