Background and aims It has been reported that occult gastrointestinal bleeding as detected by faecal occult blood (FOB) testing can occur in coeliac disease. This study examines whether a positive FOB is a feature of coeliac disease and whether FOB-positive subjects need investigation for coeliac disease. Methods First, the records of patients on the Nottingham Register for Coeliac Disease were reviewed for positive FOB testing. Second, the Nottingham colorectal cancer screening trial database was also reviewed to examine how many coeliac patients on the Register had participated and to examine their FOB results. Finally, sera from 309 screening trial participants who were FOB-positive but had no colonic abnormality were screened for immunoglobulin A (IgA) gliadin and IgA endomysial and human tissue transglutaminase (tTG) IgA antibodies. Results Five of 590 patients on the Register had had FOB tests at the time of diagnosis; four had positive tests during investigation of diarrhoea and/or anaemia. Of 21 patients on the Register who had participated in the colorectal cancer screening trial, one had a positive FOB test and was found to have a rectal tubulo-villous adenoma. Of the 309 FOB-positive patients, 7% (22 subjects) were positive for IgA gliadin antibodies, but none had IgA endomysial antibodies detected and two subjects had positive human tTG antibody assays for coeliac disease. Conclusions Occult gastrointestinal bleeding occurs in a small number of symptomatic coeliac disease patients before diagnosis, but is no more frequent in treated and undetected coeliac disease patients than in the general population. Unless there are other indications, coeliac disease does not need to be considered in the investigation of a positive FOB test.

OBJECTIVES:Inflammatory bowel disease (IBD) is usually diagnosed as a result of symptoms but occasionally is found during investigation for other conditions. An earlier report from Nottingham had found a high prevalence of previously undetected "asymptomatic" IBD detected as a result of colorectal cancer screening, and the aim of this study was to reassess the prevalence, symptoms, and outcome in these patients. METHODS:We investigated subjects found to be fecal occult blood (FOB) positive in a randomized trial of FOB screening for colorectal cancer. All FOB-positive subjects were investigated by colonoscopy or flexible sigmoidoscopy and barium enema. Subjects with IBD were referred back to their general practitioner for any further investigation and treatment. RESULTS:Seventy-five thousand two hundred fifty-three subjects (aged 45-74) were sent FOB tests and 44,838 (60%) completed a series of tests on one or more occasions. Of 133,000 test series, 1.5% were positive. During investigation 53 cases of previously undetected IBD (52 of ulcerative colitis) were found; 52% (27/52) had proctosigmoiditis only, whereas 25% (13/52) had pancolitis. Only 17% (9/52) were completely asymptomatic, with a half or more reporting some rectal bleeding (54%) or diarrhea (50%). The overall prevalence of undetected ulcerative colitis was 69/10(5) (95% CI = 50-88/10(5)) in people offered screening and 116/10(5) (95% CI = 85-147/10(5)) in people accepting screening and was higher in men. Of 32 subjects followed up 2-12 yr after diagnosis, 91% (29) continued to have few or no symptoms, with only 12 currently receiving any treatment for their colitis. CONCLUSIONS:In comparison with detected disease, undetected ulcerative colitis is relatively common but does usually cause some symptoms. It generally appears to follow a benign course, but a significant proportion have extensive colitis and may therefore be at an increased risk of colorectal cancer.

Tropomyosins (TMs) are microfilamental proteins present in all eukaryotic cells with organ specific isoforms and distinct functions.We demonstrated that non-muscle human TM isoform 5 (hTM5) is the predominant colon epithelial hTM and UC patients show significant autoantibody response against hTM5 (Gastroenterology 1998;Il4:9I2).In an attempt to identify colon epithelial hTM isoform, we used a common TM probe, REN 29, to screen a cDNA library prepared from poly (A)+ RNAs from T-84 cells.We obtained a novel clone called TC-22 whose nucleotide sequence of coding region is identical to that of hTM5 except the last exon 9 (aa222-247).Using the peptide specific to TC-22, we prepared a murine monoclonal antibody (mAb) termed TC-22-4.TC-22-4 mAb was utilized to examine 25 colonic mucosal specimens (surgical and biopsy) for the expression of TC-22 isoform by western blot analysis (WB).Of the 25 specimens, 5 were from primary colon cancer (PCC), and 4 from normal segments of PCc.Recto-sigmoid biopsy was taken from 9 UC, 2 Crohn s disease (CD) and 5 normal spastic colon.Mucosal extract was enriched for TMs and examined by WB using lOug proteins for each sample.Positive control included LCI mAb that detects hTM5.TC-22-4 reacted with TC-22TM isoform at Mr 30K in 4 of 5 PCC, and I of 4 normal area of PeC.None of 5 normal and 2 CD colon mucosa reacted.However, 2 of 9 UC (I with primary sclerosing cholangitis, I with 14 year history of UC with an adenomatous polyp) reacted at Mr 30K with TC-22-4.Each of the UC specimens from this area showed active colitis without severe dysplasia.All 25 specimens reacted with LCl.Conclusion: TC-22, a novel hTM isoform, is strongly associated with colon carcinoma.Further studies are needed to examine if the expression of TC-22 may be a clinically useful biomarker for cancer surveillance in Uc.