- Email: [email protected]
M2054 Correlation of Histological Villous Atrophy with Tissue Transglutaminase Antibody Level in Patients with Diagnosis of Celiac Disease: Retrospective Analysis of a Cohort of 187 Celiac Patients
reasons for referral were: diarrhea (58.1%), dyspepsia (46.2%), anemia (42.7%), and flatulence (29.9%). Malaise, weight loss, depression/anxiety and transaminitis were also common at the initial assessment. TTG levels were assessed in 95.7% of patients and 13.4% of the patients had a false negative (TTG < 20). 43.6% of patients were overweight or obese; the average BMI at diagnosis was 26.5 (range 16.5-51.7). 57.3% patients underwent DEXA testing. Of those, 46.3% were classified as osteopenic, 23.9% as osteoporotic, and 29.9% as having normal bone density. 78.6% of patients were referred to a nutritionist and followup appointments after diagnosis were requested by the physician in 96.6% of cases. 89.9% of the patients that did follow up noted success of therapy either as decrease in symptoms, TTG levels, or both. Contrary to common misconceptions, our patients were often older and overweight. In fact, our data suggests the obesity epidemic extends to celiac patients with nearly half of patients being overweight or obese by BMI at the time of diagnosis. Celiac disease has a diverse clinical spectrum in modern practice with GI complaints frequent, but other non-specific manifestations also noted. The sensitivity of TTG in our study population was lower than expected with a 13.4% false negative rate. 70.2% of patients who were tested had osteoporosis or osteopenia highlighting the importance of screening for coexisting bone disease. Patient noncompliance was a significant problem in the study population and hence patient education and followup are critical.
M2051 Morbidity and Mortality of Undiagnosed Celiac Disease in Adults Over 50 Years of Age. a Population Based Study Jonathan D. Godfrey, Tricia L. Brantner, Waleed Brinjikji, Kevin N. Christensen, Deanna L. Brogan, Carol T. Van Dyke, Brian Lahr, Joseph J. Larson, Lee J. Melton, Alan R. Zinsmeister, Robert A. Kyle, Joseph A. Murray
M2053
BACKGROUND: Celiac disease (CD) is a common digestive disease although many cases are unrecognized. Approximately 1% of the general population may have undiagnosed CD but the impact of undiagnosed CD on morbidity and mortality is unknown. AIM: To quantify the morbidity and mortality of undiagnosed CD. METHODS: Stored sera from a populationbased cohort of 16,592 Olmsted County, Minnesota residents comprising 85% of adults ≧50 years of age was screened for CD using sequential testing. Participants with known CD were excluded. Tissue transglutaminase (tTGA) antibodies were tested in all subjects; those positive for tTGA had a confirmatory endomysial antibody (EMA) assay. Undiagnosed CD was defined by the presence of both tTGA and EMA antibodies. A nested case-control study compared these serologically defined CD subjects 1:2 to age- and gender-matched control subjects with negative serology. Complete medical records were reviewed for comorbid conditions by reviewers unaware of serum status. Conditional logistic regression was used to identify factors associated with positive serology. The Kaplan Meier (KM) method was used to estimate overall survival and survival free of subsequent diagnosed CD. RESULTS: We identified 127 (0.75%) subjects with undiagnosed CD. After a median (range) of 10.1 (0.212.5) years of follow-up after serum draw, 20 cases (10-yr KM rate 15%) were subsequently diagnosed with CD and excluded from the final analysis, along with their matched controls; no controls were subsequently diagnosed. The remaining 107 cases (53% male, mean age 64.9) and 214 matched controls were included in the association analysis. Among the >100 conditions assessed, few were significantly associated with serology status. Undiagnosed CD patients had an increased risk of osteoporosis, worse bone density scores, and lower ferritin levels. They also had less arthritis, lower BMI, and lower cholesterol levels. Overall survival was not affected. CONCLUSION: With the exception of impaired bone density, adults over 50 years of age whose CD remained undiagnosed did not demonstrate excess comorbidity or mortality compared to controls. In patients with undiagnosed CD over the age of 50 years, ~15% will be diagnosed with CD within 10 years.
AIM Inflammatory bowel disease (IBD) and celiac disease can co-exist. False positive celiac serology (CS) has been reported in IBD. However, the diagnostic utility of CS in patients with IPAA has not been studied. Our hypothesis is that false positive CS may associate with autoimmunity related to chronic antibiotic-refractory pouchitis. Our aims were to evaluate the prevalence of true and false positive CS in IPAA patients and the factors associated with false positive. METHODS Chart review was performed in 725 consecutive patients encountered at the pouchitis Clinic to extract CS result ordered. Pathologic examination of biopsy obtained from pouch endoscopy and esophagogastroduodenoscopy was used to exclude celiac disease. Only ulcerative colitis patients were included. Refractory pouchitis was diagnosed based on persistent symptomatic pouchitis after a 4-week single- or dualantibiotic therapy. RESULTS CS was odered in 132 patients with 26 being positive. 76.9% (20/26) were false positive. 6 patients with confirmed celiac disease were excluded. Pouch diagnoses were refractory pouchitis (19%), acute pouchitis (28.6%), irritable pouch syndrome (18.3%), Crohn's disease of the pouch (16.7%), cuffitis (12.7%), and surgical complications (4.7%). 45% of the patients with a false positive CS were subsequently diagnosed with refractory pouchitis compared to 12.3% with a negative CS (p=0.003). This association remained significant (odds ratio[OR],5.4; 95% confidence interval[CI],1.7-16.7;p=0.004) after adjusting for female gender (OR,0.29; 95%CI, 0.1-0.88; p=0.03), pouch duration(OR,1;95%CI,0.9-1.1;p=0.83), presence of autoimmune condition (OR,1.66;95%CI,0.44-6.33;p=0.46) or extraintestinal manifestation of IBD (OR,1.28; 95%CI,0.47-3.48;p=0.63) in a multivariate logistic regression model. The false positive CS was not found to be associated with the other pouch-related diseases. CONCLUSION False positive CS appeared to be common in patients with IPAA and it may be associated with chronic antibiotic-refractory pouchitis. These findings suggest that autoimmunity may play a role in the pathogenesis of chronic pouchits. Comparison of Clinical Factors Between Patients with and without Refractory pouchitis
False Positive Celiac Serology Is Associated with Refractory Pouchitis in Patients with Ulcerative Colitis Lei Lian, Bo Shen
M2054 Correlation of Histological Villous Atrophy with Tissue Transglutaminase Antibody Level in Patients with Diagnosis of Celiac Disease: Retrospective Analysis of a Cohort of 187 Celiac Patients Barbara Zanini, Stefania Bertolazzi, Alessandra Mora, Daniele Turini, Francesco Lanzarotto, Chiara Ricci, Alessandro Pozzi, Vincenzo Villanacci, Alberto Lanzini
M2052 Celiac Sprue in Practice: Obesity and Bone Disease Common Krzysztof L. Kopec, Edward Feller, Steven E. Reinert, Samir A. Shah
Serum IgA-tissue transglutaminase antibodies (TTGA) are sensitive and specific markers of celiac disease (CD), but small bowel biopsy is regarded as the gold standard for CD diagnosis. Recent evidence suggests that markedly increased TTGA levels occur rarely in absence of CD, but the relationship of serum TTGA titer with abnormal small bowel histopathology, and with the severity of the histopathologic damage is still uncertain. In order to obtain this information we studied the relationship between TTGA levels and duodenal histopatology in a cohort of CD patients identified out of our CD Clinic data-base (n=967 patients) and in a cohort of consecutive non-CD controls. We identified all CD patients diagnosed using a human recombinant commercial ELISA (Eu-tTG™ IgA; Eurospital, Trieste, Italy, upper normal limit 7 U/ml) for TTGA detection, and patients with gastrointestinal complains testing TTGA negative and with Marsh 0 score at duodenal histology. Analyses of receiver-operator curve (ROC) was used to evaluate cut-off points for TTGA as a predictor of CD, and of severity of histopathologic damage (Marsh 1-2 vs 3). A total 232 adult patients, 187 CD and 45 controls, met the selection criteria. Baseline characteristics for CD patients were: age 36 ± 13 (mean ± SD) years; typical gastrointestinal symtoms in 49% of patients; TTGA level
In this study we attempted to characterize the presentation of celiac sprue in a community GI practice and assess the quality of care provided. By reviewing our experience, we hoped to understand how celiac presents in modern clinical practice and facilitate its future prompt diagnosis and appropriate management. We performed a retrospective study of patients with confirmed celiac sprue in a community GI practice in Providence, RI diagnosed during 2000 - 2007. Patients were identified by review of pathology records and their charts were reviewed for demographics, tissue transglutaminase antibody (TTG) level, presenting signs/symptoms, and BMI at diagnosis. We assessed the quality of care provided upon diagnosis by the rate of the following: assessment for underlying bone disease, nutritional referral, and whether a follow-up appointment was made. The study was approved by local IRBs. The study population consisted of 117 patients with biopsy-confirmed celiac sprue. The mean age at diagnosis was 48.6 (range 18-88); 11.1% had a positive family history. The most common
A-475
AGA Abstracts
AGA Abstracts
seem to be associated with improved nutrient intake in this cohort. Improved tools to assess and improve the quality of the GF diet are greatly needed.
AGA Abstracts
21 ± 16.2 U/ml; 10 (5.3%) patients were classified as Marsh 1, 14 (7.5%) Marsh 2 and 163 (87.2%) Marsh 3. Age for the 45 controls was 42 ± 9 years and TTGA level was 0.8 ± 0.1 U/ml. Area under the ROC was 0.98 (95% CI 0.96-0.99 , Std error 0.009, p <0.0001) indicating very high discrimination between CD and non-CD patients (sensitivity 100%specificity 95%). Data of sensitivity, specificity, PPV and NPV for TTGA cut off points for off villous atrophy (Marsh 1-2 vs Marsh 3) are summarized in Table 1. In a population at high risk for CD, TTGA exceeding the upper limit of normal is virtually always associated with CD histopathology, and the level predicts villous atrophy. Because of the highly selected population of our study, results cannot support the concept of avoiding duodenal biopsy for diagnosing CD in average risk patients but this can represent an option in high risk patients or in presence of safety issues precluding biopsy. Sensitivity, specificity, PPV and NPV for TTGA cut-off points for villous atrophy (Marsh 12 vs 3)
presentations. Degree of VA (n=308) was mild in 38%, and severe in 62%. There was no significant difference in the age or mode of presentation with regards to degree of VA. CONCLUSIONS: Classical presentations of CD in children are not as common as previously believed. Atypical presentations predominate, especially growth issues and abdominal pain. Diarrhea or diarrheal presentation is frequent only in the very young age groups. Most children have severe degrees of VA, but the severity of mucosal alteration does not correlate with mode of presentation. The reasons for the changing presentations in pediatric CD are not clear and warrant further study. M2057 Proper Diagnosis and Management of Intrahepatic Biliary Cystic Tumors: Comparision with Atypical Intrahepatic Simple Cysts Jeong Kyun Seo, Su Hyun Kim, Joo Kyung Park, Sang Myung Woo, Ji Bong Jeong, Jin Hyeok Hwang, Ji Kon Ryu, Jin Wook Kim, Sook Hyang Jeong, Yong-Tae Kim, Yong Bum Yoon, Se Hyung Kim, Sang Hyub Lee, Min A Kim, Kuhn Uk Lee Biliary cystadenoma (BCA) and biliary cystadenocarcinoma (BCAC) are often confused with other intrahepatic cystic diseases. The aims of this study were to analyze predictive factors and clinical outcome for biliary cystic tumors (BCTs) and evaluate the clinical charateristics of BCAC. We retrospectively reviewed preoperative diagnoses, overall characteristics and postoperative outcome of 20 BCTs and 19 cystadenoma-mimicking simple cysts that were pathologically confirmed. Comparing with atypical simple cysts, symptoms, left lobe cyst, thick wall, septation, mural nodule, bile duct dilatation and an increase of serum alkaline phosphatase were associated with BCTs. However, on multivariate analysis, only mural nodule, left-lobe cyst and an increase of serum ALP were significantly frequent in BCTs with Odds ratio 75.5, 13.8 and 33.0, respectively. Among the 20 BCTs, 7 BCACs were diagnosed. Cystic fluid analysis provided no significant differences between BCTs and simple cysts. Among charateristics, mural nodule (p<0.01), intrahepatic cyst debris (p<0.01), and bile duct dilation (p=0.04) were associated with BCACs. After fine needle aspiration cytology of BCTs, all except 1 BCACs with atypical cell showed non-specific findings. After surgical excision (97.4% of patients), only 1 patient with BCAC had recurrence during 29 months of follow up period. Intrahepatic cysts with mural nodule, left-lobe cyst or increment of serum ALP seems to be most suggestive of BCT. The malignant potency of BCT may be high in patients with intracystic debris, bile duct dilation and mural nodule. The prognosis of patients with BCT after complete surgical resection is excellent.
M2055 Time-Course of T-Transglutaminases Antibodies and Adherence to Gluten Free Diet (GFD) Among Celiac Patients in the General Population: Results of a “CD-Watch” Population Based Program Alberto Lanzini, Barbara Zanini, Alessandra Mora, Stefania Bertolazzi, Daniele Turini, Francesco Lanzarotto, Chiara Ricci, Francesco Donato According to clinical trials 45-80% only of celiac disease (CD) patients follow a strict GFD, but no information is available on how these results translate at the clinical level in the general population. The aim of our study was to assess the time-course of CD related antibodies and of compliance to GFD in CD patients in the general population. We extracted information from a prospectively maintained data-base of clinical, serological and histopathological characteristics of all CD patients diagnosed in our region since year 1997 (n= 2581 at December 2007) as part of a community based “Brescia CD-Watch” program (1050000 residents) in northern-Italy. Starting on year 2004, all prevalent and subsequent incident CD cases are annually tested for t-transglutaminases (t-TG ), and receive a clinical examination and a 4 point Likert scale questionnaire for adherence to GFD at 3 year intervals. Two-thousand-one-hundred-three, 1735, 1442 and 1076 patients had t-TG checked for 1 ,2 , 3 or 4 years since January 2004. At year 4, 1005 of 1076 complied with all 4 consecutive annual testing, and 838 (83%) had “persistently negative”, 15 (2%) “persistently positive” and 152 (15%) “intermittently negative or positive” t-TG. In the 3 groups female/male ratio was similar at 2/1, mean age at diagnosis was 21.6 years (95% C.I. 20.5-17.2), 29.9 years (19.0-40.9)* and 20.0 years (17.3-22.6); mean length of GFD was 7.5 years (7.1-7.9), 5.1 years (2.7-7.4) and 7.5 years (6.4-8.5), respectively. In “intermittently negative or positive” group annual t-TG serum-conversion positive-to-negative was 63% and negative-to-positive was 37%. By applying these proportions to a model developed to predict outcome of medical therapy of peptic ulcer [Lancet 1981;1(1821):29-30], a steady-state is predictable with equal proportion (5.8%) of “intermittently negative or positive” patients serum-converting either way each year at steady-state. According to this model 96% of CD patients on GFD are predicted t-TG “negative” and 4% “positive” during each yearly interval. Adherence to GFD was reported as “strict” by 1211 patients, “no diet at all” by 35 and with “infrequent (1/ month)” or “frequent (2-4/month)” digressions by 49 and by 27, respectively at the first triannual check. In conclusion our study indicates that a very high proportion of CD patients seroconvert to negative t-TG (85%) and adhere strictly to GFD (91%) within a community based “CD-Watch” follow-up strategy. By theoretical modelling 96% of CD patients are maintained t-TG negative during each calendar year by GFD. * p<0.001 ”persistently negative” vs “intermittently positive or negative” t-TG
M2058 Staining for Intracytoplasmic Lumen Increases the Detection Rate of Malignant Bile Duct Strictures Alexander W. Jahng, David S. Chung, Binh V. Pham, Sofiya Reicher, Brian Yee, Liya K. Abramyan, Samuel W. French, Viktor E. Eysselein Background and study aims: Endoscopic biopsies have a low sensitivity to diagnose malignant bile duct strictures. Molecular and cytologic markers may help to determine the malignancy of a biopsy specimen where histologic evaluation alone is non-diagnostic. Patients and methods: Fifty-seven patients who underwent forceps biopsies were retrospectively identified, yielding a total of 42 and 37 biopsy specimens for benign and malignant strictures, respectively. Biopsy specimens were evaluated for conventional histology, along with the staining for p53, Ki-67, carcinoembryonic antigen (CEA), CA19-9, CAM5.2, or intracytoplasmic lumen (IL). Sensitivity, specificity, positive and negative predictive values (PPV and NPV), and positive and negative likelihood ratios (PLR and NLR) were calculated to evaluate the performance of each test. Results: Histology alone resulted in a sensitivity and specificity of 54% and 100%. Addition of IL had the best performance, with sensitivity, specificity, PPV and PLR of 76%, 98%, 97% and 32, respectively. Intracytoplasmic lumen is a pathologic structure that is thought to occur in the setting of loss of cellular morphology and polarity. CAM5.2 increased the sensitivity only to 59%, although with comparable specificity, PPV and PLR (98%, 96% and 25, respectively). p53, Ki-67, CEA and CA19-9 increased the sensitivity to detect malignancy (59% to 81%), but significantly reduced the specificity, PPV and PLR (76% to 88%, 72% to 82%, and 3 to 5, respectively). All the markers performed poorly as a negative test to rule out malignancy (NPV 70% to 82% and NLR 0.2 to 0.5). Conclusions: The addition of cytologic marker intracytoplasmic lumen can improve the diagnostic value of endoscopic biopsy, and may change the course of management for patients with indeterminate histology.
M2056 Classical Presentation of Diarrhea in Celiac Disease Is Rare in Children Norelle Rizkalla Reilly, Jianfeng Cheng, Amy R. DeFelice, Philip Kazlow, Govind Bhagat, Peter H. Green
M2059
BACKGROUND: There are limited data on the clinical characteristics of pediatric celiac disease (CD) in the United States. AIMS: To describe the characteristics of pediatric CD patients at a major urban referral center in the northeastern United States. METHODS: A database of pediatric patients from 2000 to 2008 diagnosed with CD by small bowel biopsy was created using a disease diagnosis based search. Data were analyzed with regard to age at presentation using categories of 3-year increments (0-3 years; >3-6 years; >6-9 years; >915 years; >15 years), by mode of presentation (classical/diarrhea predominant or atypical), and by severity of villous atrophy (VA), which was classified as mild (< partial VA) and severe (subtotal/total VA). RESULTS: Of 318 patients (females 58%, M:F = 1:1.4) the mean age at diagnosis was 8.29 years (SD 4.91; range 1.1-19.5 yrs). In cases with available data (n=295), the major modes of presentation were classical (9%) and atypical (91%). The latter comprised growth problems (mainly short stature and failure to thrive) (26 %), abdominal pain (22%), screening (first degree relatives,16%, and individuals with autoimmune disorders, 7%), constipation (5%), abdominal distension (4%), anemia (3%), and others (8%). The group with classical presentation was significantly younger (mean age 4.43 yrs vs. 8.58 yrs, p<0.0001), and those with abdominal pain were significantly older (10.4 yrs vs. 7.59 yrs, p<0.0001) compared to those with other types of presentations. No age predilection was observed in patients presenting with growth problems. In the youngest group, (N=62), females predominated (1:1.32) and diarrhea (26%) and growth problems (31%) were the major presentations. The oldest age group, (N=26), also had a female predominance (1:2.11) with screening of first-degree relatives (15.4%) and abdominal pain (50%) as the major
AGA Abstracts
Antitumor Effect of Gemcitabine-Eluting Silicone/Polyurethane Membrane in Tumor Animal Model: A Preliminary Study Seok Jeong, Don Haeng Lee, Mina Kim, Soon-Sun Hong, Jung Il Lee, Jin-Woo Lee, Kye Sook Kwon, Hyung Gil Kim, Yong Woon Shin, Young Soo Kim, Chung Hwon Lee Background: The bile duct cancer usually has a biological behavior such that it grows locally with infrequently metastasizing to the distant site. Therefore, need for an efficient local therapy is rising. There were several studies on local application of drugs in bile duct cancer, including studies on drug eluting stent for malignant biliary stricture. However, there was no report about antitumor activity of gemcitabine-eluting membrane as a stent-covering material in animal model. Aims: We evaluated the antitumor effect of gemcitabine-eluting silicone/polyurethane membrane in a mouse model. Methods: The stent-covering membrane was made from gemcitabine-containing polyurethane coated with silicone. For evaluation of antitumor effects, we used a mouse tumor model (Balb/c, female, 7 weeks). The mice received 1×106 cells of CT-26 (murine colon cancer cell line) in a volume of 0.1 mL into the dorsal skin. Membrane with either gemcitabine of various concentration or vehicle (control, 0.5, 1, 2 wt%) was inserted. The tumor size was measured using calipers, and the body weight of the tumor model was monitored until 21 days after the insertion of each membrane. Results: When harvested on day 21 after the insertion, the tumor mass was decreased in a gemcitabine dose dependent manner (5833 ± 1355 mm3 for control, 4207
A-476
M2051 Morbidity and Mortality of Undiagnosed Celiac Disease in Adults Over 50 Years of Age. a Population Based Study Jonathan D. Godfrey, Tricia L. Brantner, Waleed Brinjikji, Kevin N. Christensen, Deanna L. Brogan, Carol T. Van Dyke, Brian Lahr, Joseph J. Larson, Lee J. Melton, Alan R. Zinsmeister, Robert A. Kyle, Joseph A. Murray
M2053
BACKGROUND: Celiac disease (CD) is a common digestive disease although many cases are unrecognized. Approximately 1% of the general population may have undiagnosed CD but the impact of undiagnosed CD on morbidity and mortality is unknown. AIM: To quantify the morbidity and mortality of undiagnosed CD. METHODS: Stored sera from a populationbased cohort of 16,592 Olmsted County, Minnesota residents comprising 85% of adults ≧50 years of age was screened for CD using sequential testing. Participants with known CD were excluded. Tissue transglutaminase (tTGA) antibodies were tested in all subjects; those positive for tTGA had a confirmatory endomysial antibody (EMA) assay. Undiagnosed CD was defined by the presence of both tTGA and EMA antibodies. A nested case-control study compared these serologically defined CD subjects 1:2 to age- and gender-matched control subjects with negative serology. Complete medical records were reviewed for comorbid conditions by reviewers unaware of serum status. Conditional logistic regression was used to identify factors associated with positive serology. The Kaplan Meier (KM) method was used to estimate overall survival and survival free of subsequent diagnosed CD. RESULTS: We identified 127 (0.75%) subjects with undiagnosed CD. After a median (range) of 10.1 (0.212.5) years of follow-up after serum draw, 20 cases (10-yr KM rate 15%) were subsequently diagnosed with CD and excluded from the final analysis, along with their matched controls; no controls were subsequently diagnosed. The remaining 107 cases (53% male, mean age 64.9) and 214 matched controls were included in the association analysis. Among the >100 conditions assessed, few were significantly associated with serology status. Undiagnosed CD patients had an increased risk of osteoporosis, worse bone density scores, and lower ferritin levels. They also had less arthritis, lower BMI, and lower cholesterol levels. Overall survival was not affected. CONCLUSION: With the exception of impaired bone density, adults over 50 years of age whose CD remained undiagnosed did not demonstrate excess comorbidity or mortality compared to controls. In patients with undiagnosed CD over the age of 50 years, ~15% will be diagnosed with CD within 10 years.
AIM Inflammatory bowel disease (IBD) and celiac disease can co-exist. False positive celiac serology (CS) has been reported in IBD. However, the diagnostic utility of CS in patients with IPAA has not been studied. Our hypothesis is that false positive CS may associate with autoimmunity related to chronic antibiotic-refractory pouchitis. Our aims were to evaluate the prevalence of true and false positive CS in IPAA patients and the factors associated with false positive. METHODS Chart review was performed in 725 consecutive patients encountered at the pouchitis Clinic to extract CS result ordered. Pathologic examination of biopsy obtained from pouch endoscopy and esophagogastroduodenoscopy was used to exclude celiac disease. Only ulcerative colitis patients were included. Refractory pouchitis was diagnosed based on persistent symptomatic pouchitis after a 4-week single- or dualantibiotic therapy. RESULTS CS was odered in 132 patients with 26 being positive. 76.9% (20/26) were false positive. 6 patients with confirmed celiac disease were excluded. Pouch diagnoses were refractory pouchitis (19%), acute pouchitis (28.6%), irritable pouch syndrome (18.3%), Crohn's disease of the pouch (16.7%), cuffitis (12.7%), and surgical complications (4.7%). 45% of the patients with a false positive CS were subsequently diagnosed with refractory pouchitis compared to 12.3% with a negative CS (p=0.003). This association remained significant (odds ratio[OR],5.4; 95% confidence interval[CI],1.7-16.7;p=0.004) after adjusting for female gender (OR,0.29; 95%CI, 0.1-0.88; p=0.03), pouch duration(OR,1;95%CI,0.9-1.1;p=0.83), presence of autoimmune condition (OR,1.66;95%CI,0.44-6.33;p=0.46) or extraintestinal manifestation of IBD (OR,1.28; 95%CI,0.47-3.48;p=0.63) in a multivariate logistic regression model. The false positive CS was not found to be associated with the other pouch-related diseases. CONCLUSION False positive CS appeared to be common in patients with IPAA and it may be associated with chronic antibiotic-refractory pouchitis. These findings suggest that autoimmunity may play a role in the pathogenesis of chronic pouchits. Comparison of Clinical Factors Between Patients with and without Refractory pouchitis
False Positive Celiac Serology Is Associated with Refractory Pouchitis in Patients with Ulcerative Colitis Lei Lian, Bo Shen
M2054 Correlation of Histological Villous Atrophy with Tissue Transglutaminase Antibody Level in Patients with Diagnosis of Celiac Disease: Retrospective Analysis of a Cohort of 187 Celiac Patients Barbara Zanini, Stefania Bertolazzi, Alessandra Mora, Daniele Turini, Francesco Lanzarotto, Chiara Ricci, Alessandro Pozzi, Vincenzo Villanacci, Alberto Lanzini
M2052 Celiac Sprue in Practice: Obesity and Bone Disease Common Krzysztof L. Kopec, Edward Feller, Steven E. Reinert, Samir A. Shah
Serum IgA-tissue transglutaminase antibodies (TTGA) are sensitive and specific markers of celiac disease (CD), but small bowel biopsy is regarded as the gold standard for CD diagnosis. Recent evidence suggests that markedly increased TTGA levels occur rarely in absence of CD, but the relationship of serum TTGA titer with abnormal small bowel histopathology, and with the severity of the histopathologic damage is still uncertain. In order to obtain this information we studied the relationship between TTGA levels and duodenal histopatology in a cohort of CD patients identified out of our CD Clinic data-base (n=967 patients) and in a cohort of consecutive non-CD controls. We identified all CD patients diagnosed using a human recombinant commercial ELISA (Eu-tTG™ IgA; Eurospital, Trieste, Italy, upper normal limit 7 U/ml) for TTGA detection, and patients with gastrointestinal complains testing TTGA negative and with Marsh 0 score at duodenal histology. Analyses of receiver-operator curve (ROC) was used to evaluate cut-off points for TTGA as a predictor of CD, and of severity of histopathologic damage (Marsh 1-2 vs 3). A total 232 adult patients, 187 CD and 45 controls, met the selection criteria. Baseline characteristics for CD patients were: age 36 ± 13 (mean ± SD) years; typical gastrointestinal symtoms in 49% of patients; TTGA level
In this study we attempted to characterize the presentation of celiac sprue in a community GI practice and assess the quality of care provided. By reviewing our experience, we hoped to understand how celiac presents in modern clinical practice and facilitate its future prompt diagnosis and appropriate management. We performed a retrospective study of patients with confirmed celiac sprue in a community GI practice in Providence, RI diagnosed during 2000 - 2007. Patients were identified by review of pathology records and their charts were reviewed for demographics, tissue transglutaminase antibody (TTG) level, presenting signs/symptoms, and BMI at diagnosis. We assessed the quality of care provided upon diagnosis by the rate of the following: assessment for underlying bone disease, nutritional referral, and whether a follow-up appointment was made. The study was approved by local IRBs. The study population consisted of 117 patients with biopsy-confirmed celiac sprue. The mean age at diagnosis was 48.6 (range 18-88); 11.1% had a positive family history. The most common
A-475
AGA Abstracts
AGA Abstracts
seem to be associated with improved nutrient intake in this cohort. Improved tools to assess and improve the quality of the GF diet are greatly needed.
AGA Abstracts
21 ± 16.2 U/ml; 10 (5.3%) patients were classified as Marsh 1, 14 (7.5%) Marsh 2 and 163 (87.2%) Marsh 3. Age for the 45 controls was 42 ± 9 years and TTGA level was 0.8 ± 0.1 U/ml. Area under the ROC was 0.98 (95% CI 0.96-0.99 , Std error 0.009, p <0.0001) indicating very high discrimination between CD and non-CD patients (sensitivity 100%specificity 95%). Data of sensitivity, specificity, PPV and NPV for TTGA cut off points for off villous atrophy (Marsh 1-2 vs Marsh 3) are summarized in Table 1. In a population at high risk for CD, TTGA exceeding the upper limit of normal is virtually always associated with CD histopathology, and the level predicts villous atrophy. Because of the highly selected population of our study, results cannot support the concept of avoiding duodenal biopsy for diagnosing CD in average risk patients but this can represent an option in high risk patients or in presence of safety issues precluding biopsy. Sensitivity, specificity, PPV and NPV for TTGA cut-off points for villous atrophy (Marsh 12 vs 3)
presentations. Degree of VA (n=308) was mild in 38%, and severe in 62%. There was no significant difference in the age or mode of presentation with regards to degree of VA. CONCLUSIONS: Classical presentations of CD in children are not as common as previously believed. Atypical presentations predominate, especially growth issues and abdominal pain. Diarrhea or diarrheal presentation is frequent only in the very young age groups. Most children have severe degrees of VA, but the severity of mucosal alteration does not correlate with mode of presentation. The reasons for the changing presentations in pediatric CD are not clear and warrant further study. M2057 Proper Diagnosis and Management of Intrahepatic Biliary Cystic Tumors: Comparision with Atypical Intrahepatic Simple Cysts Jeong Kyun Seo, Su Hyun Kim, Joo Kyung Park, Sang Myung Woo, Ji Bong Jeong, Jin Hyeok Hwang, Ji Kon Ryu, Jin Wook Kim, Sook Hyang Jeong, Yong-Tae Kim, Yong Bum Yoon, Se Hyung Kim, Sang Hyub Lee, Min A Kim, Kuhn Uk Lee Biliary cystadenoma (BCA) and biliary cystadenocarcinoma (BCAC) are often confused with other intrahepatic cystic diseases. The aims of this study were to analyze predictive factors and clinical outcome for biliary cystic tumors (BCTs) and evaluate the clinical charateristics of BCAC. We retrospectively reviewed preoperative diagnoses, overall characteristics and postoperative outcome of 20 BCTs and 19 cystadenoma-mimicking simple cysts that were pathologically confirmed. Comparing with atypical simple cysts, symptoms, left lobe cyst, thick wall, septation, mural nodule, bile duct dilatation and an increase of serum alkaline phosphatase were associated with BCTs. However, on multivariate analysis, only mural nodule, left-lobe cyst and an increase of serum ALP were significantly frequent in BCTs with Odds ratio 75.5, 13.8 and 33.0, respectively. Among the 20 BCTs, 7 BCACs were diagnosed. Cystic fluid analysis provided no significant differences between BCTs and simple cysts. Among charateristics, mural nodule (p<0.01), intrahepatic cyst debris (p<0.01), and bile duct dilation (p=0.04) were associated with BCACs. After fine needle aspiration cytology of BCTs, all except 1 BCACs with atypical cell showed non-specific findings. After surgical excision (97.4% of patients), only 1 patient with BCAC had recurrence during 29 months of follow up period. Intrahepatic cysts with mural nodule, left-lobe cyst or increment of serum ALP seems to be most suggestive of BCT. The malignant potency of BCT may be high in patients with intracystic debris, bile duct dilation and mural nodule. The prognosis of patients with BCT after complete surgical resection is excellent.
M2055 Time-Course of T-Transglutaminases Antibodies and Adherence to Gluten Free Diet (GFD) Among Celiac Patients in the General Population: Results of a “CD-Watch” Population Based Program Alberto Lanzini, Barbara Zanini, Alessandra Mora, Stefania Bertolazzi, Daniele Turini, Francesco Lanzarotto, Chiara Ricci, Francesco Donato According to clinical trials 45-80% only of celiac disease (CD) patients follow a strict GFD, but no information is available on how these results translate at the clinical level in the general population. The aim of our study was to assess the time-course of CD related antibodies and of compliance to GFD in CD patients in the general population. We extracted information from a prospectively maintained data-base of clinical, serological and histopathological characteristics of all CD patients diagnosed in our region since year 1997 (n= 2581 at December 2007) as part of a community based “Brescia CD-Watch” program (1050000 residents) in northern-Italy. Starting on year 2004, all prevalent and subsequent incident CD cases are annually tested for t-transglutaminases (t-TG ), and receive a clinical examination and a 4 point Likert scale questionnaire for adherence to GFD at 3 year intervals. Two-thousand-one-hundred-three, 1735, 1442 and 1076 patients had t-TG checked for 1 ,2 , 3 or 4 years since January 2004. At year 4, 1005 of 1076 complied with all 4 consecutive annual testing, and 838 (83%) had “persistently negative”, 15 (2%) “persistently positive” and 152 (15%) “intermittently negative or positive” t-TG. In the 3 groups female/male ratio was similar at 2/1, mean age at diagnosis was 21.6 years (95% C.I. 20.5-17.2), 29.9 years (19.0-40.9)* and 20.0 years (17.3-22.6); mean length of GFD was 7.5 years (7.1-7.9), 5.1 years (2.7-7.4) and 7.5 years (6.4-8.5), respectively. In “intermittently negative or positive” group annual t-TG serum-conversion positive-to-negative was 63% and negative-to-positive was 37%. By applying these proportions to a model developed to predict outcome of medical therapy of peptic ulcer [Lancet 1981;1(1821):29-30], a steady-state is predictable with equal proportion (5.8%) of “intermittently negative or positive” patients serum-converting either way each year at steady-state. According to this model 96% of CD patients on GFD are predicted t-TG “negative” and 4% “positive” during each yearly interval. Adherence to GFD was reported as “strict” by 1211 patients, “no diet at all” by 35 and with “infrequent (1/ month)” or “frequent (2-4/month)” digressions by 49 and by 27, respectively at the first triannual check. In conclusion our study indicates that a very high proportion of CD patients seroconvert to negative t-TG (85%) and adhere strictly to GFD (91%) within a community based “CD-Watch” follow-up strategy. By theoretical modelling 96% of CD patients are maintained t-TG negative during each calendar year by GFD. * p<0.001 ”persistently negative” vs “intermittently positive or negative” t-TG
M2058 Staining for Intracytoplasmic Lumen Increases the Detection Rate of Malignant Bile Duct Strictures Alexander W. Jahng, David S. Chung, Binh V. Pham, Sofiya Reicher, Brian Yee, Liya K. Abramyan, Samuel W. French, Viktor E. Eysselein Background and study aims: Endoscopic biopsies have a low sensitivity to diagnose malignant bile duct strictures. Molecular and cytologic markers may help to determine the malignancy of a biopsy specimen where histologic evaluation alone is non-diagnostic. Patients and methods: Fifty-seven patients who underwent forceps biopsies were retrospectively identified, yielding a total of 42 and 37 biopsy specimens for benign and malignant strictures, respectively. Biopsy specimens were evaluated for conventional histology, along with the staining for p53, Ki-67, carcinoembryonic antigen (CEA), CA19-9, CAM5.2, or intracytoplasmic lumen (IL). Sensitivity, specificity, positive and negative predictive values (PPV and NPV), and positive and negative likelihood ratios (PLR and NLR) were calculated to evaluate the performance of each test. Results: Histology alone resulted in a sensitivity and specificity of 54% and 100%. Addition of IL had the best performance, with sensitivity, specificity, PPV and PLR of 76%, 98%, 97% and 32, respectively. Intracytoplasmic lumen is a pathologic structure that is thought to occur in the setting of loss of cellular morphology and polarity. CAM5.2 increased the sensitivity only to 59%, although with comparable specificity, PPV and PLR (98%, 96% and 25, respectively). p53, Ki-67, CEA and CA19-9 increased the sensitivity to detect malignancy (59% to 81%), but significantly reduced the specificity, PPV and PLR (76% to 88%, 72% to 82%, and 3 to 5, respectively). All the markers performed poorly as a negative test to rule out malignancy (NPV 70% to 82% and NLR 0.2 to 0.5). Conclusions: The addition of cytologic marker intracytoplasmic lumen can improve the diagnostic value of endoscopic biopsy, and may change the course of management for patients with indeterminate histology.
M2056 Classical Presentation of Diarrhea in Celiac Disease Is Rare in Children Norelle Rizkalla Reilly, Jianfeng Cheng, Amy R. DeFelice, Philip Kazlow, Govind Bhagat, Peter H. Green
M2059
BACKGROUND: There are limited data on the clinical characteristics of pediatric celiac disease (CD) in the United States. AIMS: To describe the characteristics of pediatric CD patients at a major urban referral center in the northeastern United States. METHODS: A database of pediatric patients from 2000 to 2008 diagnosed with CD by small bowel biopsy was created using a disease diagnosis based search. Data were analyzed with regard to age at presentation using categories of 3-year increments (0-3 years; >3-6 years; >6-9 years; >915 years; >15 years), by mode of presentation (classical/diarrhea predominant or atypical), and by severity of villous atrophy (VA), which was classified as mild (< partial VA) and severe (subtotal/total VA). RESULTS: Of 318 patients (females 58%, M:F = 1:1.4) the mean age at diagnosis was 8.29 years (SD 4.91; range 1.1-19.5 yrs). In cases with available data (n=295), the major modes of presentation were classical (9%) and atypical (91%). The latter comprised growth problems (mainly short stature and failure to thrive) (26 %), abdominal pain (22%), screening (first degree relatives,16%, and individuals with autoimmune disorders, 7%), constipation (5%), abdominal distension (4%), anemia (3%), and others (8%). The group with classical presentation was significantly younger (mean age 4.43 yrs vs. 8.58 yrs, p<0.0001), and those with abdominal pain were significantly older (10.4 yrs vs. 7.59 yrs, p<0.0001) compared to those with other types of presentations. No age predilection was observed in patients presenting with growth problems. In the youngest group, (N=62), females predominated (1:1.32) and diarrhea (26%) and growth problems (31%) were the major presentations. The oldest age group, (N=26), also had a female predominance (1:2.11) with screening of first-degree relatives (15.4%) and abdominal pain (50%) as the major
AGA Abstracts
Antitumor Effect of Gemcitabine-Eluting Silicone/Polyurethane Membrane in Tumor Animal Model: A Preliminary Study Seok Jeong, Don Haeng Lee, Mina Kim, Soon-Sun Hong, Jung Il Lee, Jin-Woo Lee, Kye Sook Kwon, Hyung Gil Kim, Yong Woon Shin, Young Soo Kim, Chung Hwon Lee Background: The bile duct cancer usually has a biological behavior such that it grows locally with infrequently metastasizing to the distant site. Therefore, need for an efficient local therapy is rising. There were several studies on local application of drugs in bile duct cancer, including studies on drug eluting stent for malignant biliary stricture. However, there was no report about antitumor activity of gemcitabine-eluting membrane as a stent-covering material in animal model. Aims: We evaluated the antitumor effect of gemcitabine-eluting silicone/polyurethane membrane in a mouse model. Methods: The stent-covering membrane was made from gemcitabine-containing polyurethane coated with silicone. For evaluation of antitumor effects, we used a mouse tumor model (Balb/c, female, 7 weeks). The mice received 1×106 cells of CT-26 (murine colon cancer cell line) in a volume of 0.1 mL into the dorsal skin. Membrane with either gemcitabine of various concentration or vehicle (control, 0.5, 1, 2 wt%) was inserted. The tumor size was measured using calipers, and the body weight of the tumor model was monitored until 21 days after the insertion of each membrane. Results: When harvested on day 21 after the insertion, the tumor mass was decreased in a gemcitabine dose dependent manner (5833 ± 1355 mm3 for control, 4207
A-476