Our understanding of the clinical features and importance of coeliac disease (CD) has been completely transformed in the last decade, thanks to three key developments. Firstly, new serological screening studies, using general populations instead of less representative blood donor groups, suggest that the true prevalence of CD in Europe is 1: 120-1:200 ’ 2: as current serological tests lack 100% sensitivity, even these figures may be underestimates. Secondly, more and more data confirm that only a minority of coeliacs fit the stereotype underweight young patient with diarrhoea and that many lacking this “typical” presentation go undiagnosed. Many patients, particularly males, are overweight 3. First presentation in the elderly is common, frequently with delay in diagnosis 4. Fewer than half of our patients report diarrhoea as a symptom 3. Upper gastrointestinal (GI) symptoms may predominate 5. Following the prototype of dermatitis herpetiformis, presentations involving organ systems other than the gut, with absent or trivial GI symptoms, are increasingly recognised. Anaemia as the sole manifestation of CD is well described, and duodenal biopsy should be part of the routine investigation of all patients with iron deficiency. Although CD is a disease of the proximal small bowel, vitamin B 12 deficiency is common and should be considered as a possible presentation 6. The neurological syndrome of partial seizures, intracerebral calcification and CD is well described in Italy, though curiously little reported elsewhere. Osteoporosis and arthralgia, infertility and miscarriage, and abnormalities in liver biochemistry are all well documented. Finally, Marsh 7 proposes that we should extend the histological spectrum of gluten sensitivity beyond even partial villous atrophy (PVA) - themselves contentious until recently - to include crypt hyperplasia and excess intraepithelial lymphocytes without villous atrophy WA). In Northern Ireland, we estimate that around 90% of coeliac patients currently go undiagnosed *. How can we improve diagnostic pickup of CD? Many coeliac patients have upper GI symptoms of dyspepsia, reflux, dysphagia, nausea/vomiting and noncardiac chest pain. The probable mechanism is gut dysmotility, which is gluten-sensitive 5, and these patients are likely to have endoscopy as the initial investigation to search for peptic disease rather than enteropathy. Routine duodenal biopsy in all patients having endoscopy would have major resource implications for histopathology services and is probably not practical. Endoscopic recognition of VA, studied by Brocchi et al. in this issue of Digest Liver Dis (2002;34: 177-82) 9, allows selection of patients for biopsy. Reduction in, or absence of duodenal folds, scalloping of duodenal folds, and mucosal grooves or mosaic pattern mucosa are recognised markers of VA and have very high specificity. While most authors have focused on appearances in the second part of the duodenum, Brocchi et al. suggest that inspection of the duodenal bulb is also worthwhile and may improve diagnosis, although sensitivity remains well short of 100%. Their classification of VA differed from that used by other researchers, with their “severe PVA” equivalent to subtotal (ST) VA in the Marsh classification and their “STVA” corresponding to total (T)VA. All their coeliac patients studied fell into these two categories, with an overall sensitivity for one or more endoscopic markers of 94%. A recent Northern Ireland study lo identified markers in only 82% of patients within these categories, falling to 58% of patients with milder PVA. The results in both studies were obtained from a series of patients undergoing endoscopy specifically for duodenal biopsy and, therefore, with a high index of suspicion for CD: the sensitivity during routine endoscopy for dyspepsia is likely to be smaller, and one would not expect to identify Marsh grade I or II enteropathy in this way. Nevertheless, as duodenal biopsy, at all endoscopies, is not practical, endoscopic recognition of VA is a pragmatic way to improve diagnosis. Around one-fifth of our coeliac patients are identified in this way lo, and 1:63 of patients referred for endoscopy in primary care had endoscopic markers for VA ‘. We do not, at present, know how many patients with upper GI symptoms have CD as the cause, and there may be a role for serological screening in this context: should dyspeptic patients have coeliac as well as Helicobacter pylori antibodies checked? Serological testing for CD has facilitated diagnosis particularly among patients with mild, non-specific forms or with no GI symptoms, though it too has limitations. Antigliadin antibody testing has limited sensitivity and specificity in adults, and IgA class en
Read full abstract
Aug 1, 2005
Stephan Hellmig + 8
Open Access
