Article Text

Download PDFPDF

Small bowel posters 326–335

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


A. Beale, R. Crimmins, G.L. Swift, P.M. Smith. Department of Gastroenterology, Llandough Hospital, Cardiff, UK

Introduction: The Coeliac Society is reporting increasing numbers of new members each year, a large proportion of whom are adults at diagnosis. It is unsure whether this is simply due to greater exposure of the `coeliac iceberg' or reflects a genuine increase in disease.

Aims: We assessed all new cases of coeliac disease (CD) diagnosed in South Glamorgan between 1996 and 2000, identified their characteristics and presenting features and compared these to figures obtained between 1981 and 1995 in the same area.

Method: Data was obtained from clinical, pathology, dietetic and GP records in the area. Ethical approval was given Bro Taf LREC.

Results: In our population area of approximately 420 000, 125 new cases of coeliac disease were diagnosed between 1996 and 2000. Of these, 112 were adults (90%) and 13 were children at diagnosis. Whilst the number of children has remained stable there has been a significant increase in the number of adults compared to those in the previous 3 quinquennia (see table). There was a male:female ratio of 1:2. The mean age at diagnosis in adults was 53yrs, with 68 (61%) of the adult patients aged over 50 and 20 (18%) aged over 70. 30 (24%) of all patients had a known affected first degree relative. The single most common presenting complaint was anaemia (50%) and 23 (20%) of the adults had osteoporosis at diagnosis. 14 (11%) had another autoimmune disorder.

Abstract 326

Conclusion: There has been a large increase in new cases of coeliac disease in South Glamorgan over the last quinquennia. This appears to be due to the large number of new adult patients including many aged over 70 at diagnosis. It is vital that all clinicians have increased vigilance for this condition to try to reduce possible associated morbidity.


S.J. Moodie, J.S. Fraser, A.L. King, E. Kondeatis, H.J. Ellis, P.J. Ciclitira. Gastroenterology, St Thomas Hospital, Kings College, London, UK

Coeliac disease is strongly associated with the HLA DQ2 (DQA1*0501 DQB1*0201) heterodimer encoded in cis on a DRB1*03 haplotype or in trans on DRB1*07 and DRB1*11 haplotypes. DQ2 negative coeliacs tend to have the disease associated DRB1*04 DQA1*0301 DQB1*0302 (DQ8) haplotype. DQ8 ancestral haplotypes differ at the DRB1*04 locus and in insulin dependant diabetes mellitus DRB1*04 subtypes have been shown to determine DQ8 associated risk in different populations. We aimed to address this issue in UK coeliacs, and also to describe HLA types of the rare DQ2 and DQ8 negative coeliacs.

Methods: 283 UK coeliacs were typed for the presence of the DQ2 heterodimer in cis or trans by PCR-SSP. All those found to be DQ2 negative were typed at a higher resolution for HLA class II alleles, with DRB1*04 subtyping of DQ8 haplotypes and HLA A, B, C typing of all DQ2 and DQ8 negative patients.

Results: 263/283 (93%) were DQ2 positive (252 in cis, 11 in trans), 20/283 (7%) were DQ2 negative. 15/20 of these were DRB1*04 DQ8 positive. 40% of DQ8 haplotypes in DQ2 negative coeliacs carried the DRB1*0404 allele, 30% the 0401 allele, 15% each 0402 and 0404 respectively. Of the 5 DQ2 and DR4 DQ8 negative coeliacs, 4 carried one half of the DQA1*0501, DQB1*0201 (DQ2) heterodimer, and one patient neither part. 4 of the 5 also had HLA B*44, no other association with class I alleles was found.

Conclusions: We found that DQ2 negative UK coeliacs mostly encode DQ8 in accordance with findings in other populations. Several different DQ8 ancestral haplotypes were associated with disease, suggesting a primary role for DQ8 itself. The frequency of DRB1*0404 carrying DQ8 haplotypes was marginally higher and DRB1*0401 marginally lower than the background population frequencies although this was not statistically significant. An interesting finding is that coeliacs without either DQ2 or DQ8 usually carry one part of the DQ2 heterodimer, suggesting that this may be enough to increase the affinity of a DQ heterodimer for binding and presenting gluten derived peptide fragments in someone genetically predisposed to develop coeliac disease.


N.C. Direkze, N.I. McNeil. Ealing Hospital NHS Trust, Uxbridge Road, Southall UB1 3HW, UK

Background: It is recognised that coeliac disease and ulcerative colitis are associated with hyposplenism. Pneumococcal prophylaxis for these patients is not normal practice at our hospital but despite this personal experience suggests that coeliac patients rarely suffer from severe infections. However, a recent Drugs and Therapeutics Bulletin suggested that these patients should be vaccinated to reduce the risk of pneumococcal sepsis. Therefore we decided to assess the current practice of gastroenterologists in the UK.

Method: A questionnaire was posted to 292 UK gastroenterologists, three months later the results were collated and analysed.

Results: There was a 78% response rate (n=229).

Coeliac disease: Only 25 (11%) of respondents prescribe pneumococcal prophylaxis for hyposplenism in coeliac disease. Of these 25, the majority (17) use pneumovax. To select patients for treatment 22 respondents use haematological indices, two others use ultrasound for this purpose. Of the 229 who responded 9% (n=18) had encountered severe infections in coeliac patients and reported a total 36 severe infections. Severe pneumococcal sepsis, refractory pneumonia and coexisting immunocompromise were commented on as distinguishing features.

Ulcerative colitis: 6% (n=13) of respondents assess ulcerative colitis patients for hyposplenism. Five of these do not treat this hyposplenism. If treated these patients receive pneumovax alone (4) or a mixture of pneumovax and long-term penicillin (2)

Conclusion: Despite the suggestions of the Drugs and Therapeutics Bulletin and the British Haematology Guidelines, our findings suggest that current practice is not to assess or treat coeliac patients for hyposplenism in the UK. As severe pneumococcal disease has been encountered, we believe that gastroenterologists should reassess this practice and produce guidance for these patients.


A.L. King, S.J. Moodie, J.S. Fraser, A.M. Dearlove, D. Curtis, P.J. Ciclitira. Gastroenterology, The Rayne Institute, St Thomas' Hospital, London, UK

Background: We previously demonstrated association of coeliac disease (CD) with a locus on chromosome 2q33, containing the cytotoxic T-lymphocyte associated (CTLA-4) gene and the CD28 gene. Association has also been demonstrated in other European populations, however the precise aetiological polymorphism is still not known. CD is associated with other autoimmune disorders including type 1 diabetes (T1D) and Graves' disease (GD). These conditions also demonstrate association with 2q33, and the strongest association has recently been demonstrated with two single nucleotide polymorphisms close to the CTLA-4 gene: MH30 (-23327G>C) and CT60 (6230G>A) (unpublished data).

Aims: To test for association of MH30 and CT60 with CD.

Methods: 149 family trios consisting of an affected individual plus both parents were genotyped for MH30 and CT60. Genotyping was performed using the SNaPshot method (Applied Biosystems), and products were characterised using an ABI 377 sequencer. Data was analysed using transmission disequilibrium testing (TDT). 100 unaffected spouses of individuals with CD were also genotyped as a control group. Absolute allele and genotype counts were compared with the 149 affected individuals using 2x2 and 2x3 contingency tables respectively.

Results: TDT compares allele transmissions from heterozygous parents to unaffected offspring: for MH30 G=70, C=65 (chi-squared=0.185, 1df, p=0.667), for CT60 G=77, A=67 (chi-squared=0.694, 1df, p=0.405). Similar allele frequencies were demonstrated in case and control groups for both MH30 (chi-squared=0.016, 1df, p=0.899) and CT60 (chi-squared=0.1462, 1df, p=0.702). Genotype frequencies were also almost identical in both case and control groups.

Conclusion: Using both a TDT and a case-control design we demonstrated no evidence of association between CD and the MH30 or CT60 polymorphisms. Although CD, T1D and GD have all demonstrated association with the CTLA-4/CD28 gene region, the polymorphism(s) on 2q33 that confer susceptibility to CD may be different from those conferring susceptibility to T1D and GD.


J.R. Butterworth1, W.M.C. Rosenberg2, S. Jobson3, D. Briggs3, W.M. Howell4, T.H. Iqbal1, B.T. Cooper1. 1Gastroenterology Unit, City Hospital, Birmingham, UK; 2Liver Group, Division of Infection, Inflammation and Repair, University of Southampton, UK; 3National Blood Service, Birmingham, UK; 4Division of Human Genetics, University of Southampton, UK

Introduction: Coeliac disease (CD) is a HLA associated disease that has become increasingly recognised among South Asians resident in the UK. No studies have compared the HLA haplotypes amongst Caucasian and South Asian patients with CD.

Aims: To compare the HLA haplotypes of Caucasian and South Asian CD patients.

Methods: Polymerase chain reaction using sequence specific primers capable of identifying the HLA class I and II alleles was used to type 72 Caucasian and 18 South Asian patients with CD.

Results: Significantly more Caucasian patients were DQ2 positive compared to the South Asians (97.2% compared with 83.3%, P=0.02). The haplotype counts (total number of positive chromosomes) for the HLA-A*01, A*03 and DQB1*02 alleles were significantly higher amongst the Caucasians, (P=0.015, 0.015, 0.002 respectively). By contrast, the HLA-A*26, A*32 and Cw*0702 alleles were more frequent amongst the South Asians, (P<0.0001, 0.002, <0.0001 respectively). The haplotype counts for the HLA-B*08 and DRB1*03 alleles were similar between the groups. The HLA-A*01, B*08, DRB1*03 and DQB1*02 haplotype combination was seen in 65.3% of Caucasian compared with 22.2% of South Asian coeliacs (P=0.001).

Conclusions: We have shown preliminary data of distinct HLA associations amongst Caucasian and South Asian patients with CD. This may explain the observed differential presentation in age and symptoms at diagnosis that we have previously reported1. It suggests that non-HLA regions are likely to be stronger determinants of CD susceptibility in South Asians compared to Caucasians.

1Butterworth JR, Iqbal TH, Cooper BT. Coeliac disease in South Asians resident in Britain. Comparison with White Caucasians. Gut 2001;48 (Suppl I):A79.


S.D. Johnston, J.S.A. Collins, T.C.K. Tham, N.I. McDougall, P. Murphy. Belfast City Hospital, Royal Victoria Hospital, Ulster Hospital Dundonald, Antrim Area Hospital, Craigavon Area Hospital, N. Ireland

Background: Tissue transglutaminase is now recognised as the autoantigen for antiendomysial antibodies (EMA). Antibodies to tissue transglutaminase (tTG) have been proposed as a valuable test for coeliac disease since they have a sensitivity of 85–100% and specificity of 76–98% which compares favourably with the respective values for EMA (sensitivity 86–100%; specificity 94–100%).

Aim: To evaluate the value of tTG antibodies for the diagnosis of coeliac disease in our outpatient population.

Methods: Patients presenting with symptoms suggestive of coeliac disease were evaluated using serological testing and duodenal biopsies. Four endoscopic duodenal biopsies were taken and assessed histologically for features of coeliac disease. Coeliac disease was defined as at least severe partial villous atrophy, subtotal or total villous atrophy. The sensitivity, specificity, negative (NPV) and positive predictive value (PPV) of the three serological tests were compared.

Results: 78 patients (24 male; mean age 53.0 years) were included in the study, of whom 25 (11 male; mean age 54.4 years) were diagnosed as having coeliac disease. Weight loss (7 v 3; p=0.01) was more frequent in coeliacs compared to controls whereas the frequency of anaemia (6 v 10) and diarrhoea (6 v 9) did not differ significantly between the two groups. The sensitivity, specificity, NPV and PPV of tTG (88%; 85%; 94% and 73%) were compared to those for EMA (92%; 98%; 96%; 96%) and antigliadin antibodies (80%; 79%; 89%; 64%) respectively.

Conclusions: The diagnostic value of tTG antibodies was intermediate between that of EMA and AGA. Duodenal biopsy remains the gold standard diagnostic test for coeliac disease.


P. Kelly1,2, I. Menzies4, R. Crane4, I. Zulu1, C. Nickols3, R. Feakins3, V. Mudenda1, M. Katubulushi1, S. Greenwald3, M. Farthing5. 1Digestive Diseases Research Project, University of Zambia School of Medicine, University Teaching Hospital, Lusaka, Zambia; 2Dept Adult and Paediatric Gastroenterology and 3Dept Histopathology & Morbid Anatomy, St Bart's & Royal London School of Medicine & Dentistry, London, UK; 4King's College Hospital, London, UK; 5University of Glasgow School of Medicine, Glasgow, UK

Background: Small intestinal mucosal architecture, absorptive capacity and permeability differ in tropical and temperate populations, but their dependence on environmental factors has not been established in indigenous (as opposed to migrant) adults.

Aims: To assess variability of jejunal architecture and function over time and their responsiveness to environmental influences.

Participants: Adults (n=202) resident in a small part of one impoverished township in Lusaka, Zambia living under conditions of high exposure to enteropathogens.

Methods: Jejunal biopsy and four-sugar absorption/permeability measurement annually in each of 3 consecutive years; morphometry of villous and crypt compartments.

Results: All biopsies showed macroscopic enteropathic changes. Permeability was higher in younger women, and in HIV seropositive adults. Structural and functional measurements were only weakly correlated with each other and did not show parallel correlations with intestinal infection, nutrition, or HIV stage. HIV seropositive participants tended to show a progressive reduction in villous height. There was a high degree of variability over time between individuals, with a majority changing significantly from one year to the next. In the group as a whole there was a 15–20% seasonal variation in villous height.

Conclusions: In this tropical population, the jejunal mucosa is dynamic and responsive to environmental influences. Architectural and functional measurements do not predict each other but are complementary.


S. Sen, M. Mullan, T.J. Parker, J. Woolner, S.A. Tarry, J.O. Hunter. Department of Gastroenterology, Addenbrooke's Hospital, Hills Road, Cambridge CB2 2QQ, UK

Background: Malabsorbed food residues in coeliac disease pass into the caecum and undergo bacterial fermentation. We aimed to assess the effects of a gluten-free diet on colonic fermentation in coeliac disease.

Methods: Five patients with newly-diagnosed coeliac disease underwent colonic fermentation studies after two weeks on a standardised `normal western' diet provided from a metabolic kitchen . The studies were repeated after a further three months on a gluten free diet, the diet in the last two weeks being carefully matched for calorie, protein, fat and carbohydrate content, and for substrates of fermentation with the standard diet. A validated composite symptom score was completed daily. Colonic fermentation was assessed by continuous measurement of gaseous exchange in a 1.4m3 canopy for 24 hours, followed by end-expiratory breath hydrogen determination every 30 minutes for 3 hours after 20ml lactulose. Hydrogen was measured by electro-chemical cell (GMI, Renfrew, UK)

Results: See table. There was a statistically significant reduction in total hydrogen production (p=0.043), although this was not reflected in the breath hydrogen levels.

Abstract 333

Conclusion: Patterns of colonic fermentation improved in all five patients possibly because of reduced malabsorption on a gluten-free diet. Colonic fermentation may be an additional factor producing symptoms in coeliac disease.


G. Monteleone, G. Mazzarella1, V.M. Salvati2, P. Lionetti3, R. Troncone2, T.T. MacDonald. Division of Infection, Inflammation and Repair, University of Southampton, School of Medicine, Southampton General Hospital, Southampton, U K; 1Istituto di Scienze dell'Alimentazione, CNR, Avellino, Italy; 2Department of Pediatrics and European Laboratory for the Investigation of Food-Induced Diseases, University Federico II, Naples, Italy; 3Dipartimento di Pediatria, Universita' di Firenze, Florence, Italy

Background and Aim: Interferon (IFN)-γ is a key mediator of the immunopathology in celiac disease (CD). The aim of this study was to examine the involvement of STAT1, a transcription factor activated by IFN-γ,_ and SOCS-1, a protein which negatively regulate the IFN-γ /STAT1 pathway, in CD.

Methods: Duodenal biopsies, taken from CD patients and normal controls, were analysed for STAT1 by Western blotting, EMSA, and immunoistochemistry, whereas SOCS-1 was analysed by Southern and Western blotting. In an ex vivo organ culture of treated CD biopsies the effect of a JAK/STAT1 inhibitor on the gliadin-mediated induction of costimulatory molecules was examined.

Results: High IFN-γ and a more pronounced phosphorylation and DNA-binding activity of STAT1 were seen in CD in comparison to controls. By immunoistochemistry, STAT1 was localised within the nucleus of epithelial and lamina propria cells. Staining was more intense and diffuse in CD compared to controls. Despite CD samples contained high SOCS-1 RNA, SOCS-1 protein was undetectable. In cultured treated CD biopsies, gliadin induced activation of STAT1 but not SOCS-1. Furthermore, inhibition of STAT1 prevented the gliadin-mediated induction of ICAM-1 and B7–2.

Conclusions: Data suggest that exaggerated IFN-γ__and defective SOCS-1 protein expression can result in a persistent STAT1 activation, thereby contributing to maintain and expand the local inflammatory response in CD.


R. Dor, M. McStay, C. Blanshard. Academic Department of Gastroenterology, Homerton Hospital; St Bartholomew's and the Royal London School of Medicine and Dentistry, UK

Background: It has long been recognised that differences in small bowel morphometry are present in healthy immigrants from Africa and Asia compared to Caucasians. This may be due to increased inflammation in the small bowel mucosa .

Aims: To establish whether morphometric abnormalities persist in second generation immigrants and whether this is attributable to small intestinal (SI) inflammation.

Patients and Methods: 62 dyspeptic patients with no evidence of malabsorption had jumbo biopsies obtained from the distal duodenum. Demographic details were obtained by a questionnaire. Small bowel morphometry was carried out using an image analysis system and the intraepithelial lymphocytes counted. Further biopsies were snap frozen in liquid nitrogen and the levels of IFNgamma measured by RT-PCR

Results: 28 native Caucasians, 27 first gen. immigrants from Africa, the Caribbean and Asia and 7 second gen. immigrants were studied. 3 of these had travelled to their parent's country of origin in the last 2 years. There was a significant increase in crypt depth and decrease in villus height in both immigrant generations compared to Caucasians and a trend towards more marked changes in the first generation group (see table). In both immigrant groups there was a significant increase in intraepithelial lymphocytes and IFN gamma.

Abstract 335

Discussion: Relative villus atrophy and crypt hyperplasia is present in first and second gen immigrants, which may be due to SI inflammation in response to an environmental antigen from their country of origin.