Coeliac disease in primary care

BMJ 2007; 334 doi: http://dx.doi.org/10.1136/bmj.39170.410347.80 (Published 05 April 2007) Cite this as: BMJ 2007;334:704
  1. Roger Jones, Wolfson professor of general practice (roger.jones{at}kcl.ac.uk)
  1. Department of General Practice and Primary Care, KCL School of Medicine, London SE11 6SP

    Is common, underdiagnosed, and can present with non-specific symptoms

    Coeliac disease affects around 1% of the general population, but most cases are unrecognised and diagnosis is often delayed considerably.1 2 This is surprising, given how common the disease is and how serious its effects can be.1 3 4 Several possible reasons exist for this delay; the most important is that most patients with coeliac disease do not have typical symptoms of malabsorption. Even if these symptoms are present their non-specific nature may not trigger diagnostic suspicion of coeliac disease. Other atypical presentations can occur, especially in older patients,5 and the disease may even be seen in obese people.6

    In this week's BMJ, Hopper and colleagues report a validated clinical prediction rule to determine an effective diagnostic method of detecting all cases of coeliac disease in people referred for gastroscopy. People with positive tissue transglutaminase antibodies and “low risk symptoms,” as well as those with the high risk symptoms of diarrhoea, weight loss, and anaemia, were investigated with duodenal biopsy, while those with a negative antibody result and low risk symptoms were not biopsied.7 They found that pre-endoscopy serological testing combined with biopsy of people with high risk symptoms had a sensitivity of 100%. They also found that a proportion of high risk patients with positive serology turn out not to have coeliac disease on biopsy, which should prompt reconsideration of the need for a lifelong gluten-free diet. An accompanying commentary in this issue by Graber and Kumar discusses the potential application of the decision rule to clinical practice.8

    Coeliac disease is characterised by a lifelong intolerance to certain storage proteins contained in wheat, rye, and barley, collectively known as gluten, and it is an unusual combination of food intolerance and autoimmunity. Chronic inflammation of proximal small intestinal mucosa, with atrophy of small intestinal villi, is associated with impaired absorption of nutrients and increased secretion of water and solids because of abnormal intestinal permeability. The disease has been classified into four phenotypes.2 In classic coeliac disease, patients have intestinal malabsorption and gastrointestinal symptoms, whereas the atypical (most common) form of the disease has few or no gastrointestinal symptoms. Atypical disease has other problems, however, including iron deficiency anaemia, osteoporosis, short stature, infertility, and unfavourable outcomes of pregnancy.9 In silent coeliac disease, asymptomatic patients are found to have gluten induced villous atrophy, and in latent disease, patients may have normal mucosa or may show villous atrophy, which improves after gluten withdrawal.

    The gold standard for the diagnosis of coeliac disease is a proximal small intestinal (duodenal) biopsy, but serum testing for antigliadin and endomysial antibodies is also available and offers reasonable sensitivity and specificity. The American Gastroenterology Association recommends the use, in primary care, of the IgA tissue transglutaminase antibody test as the single diagnostic serological test.2 However, coeliac disease may cause IgA deficiency, so that IgG endomysial antibodies and transglutaminase antibodies should be checked if biopsy is positive but the IgA tests are negative. Furthermore, because the histology of the disease can be patchy, careful quadrantic biopsies from the duodenum are needed. Even then difficulties in the preparation and interpretation of histological material may create diagnostic uncertainty, so that follow-up and reinvestigation may be needed to confirm the diagnosis.10

    Coeliac disease is more common in people whose first degree relatives have the condition and in patients with iron deficiency anaemia, low bone mineral density, and other autoimmune disorders, such as type 1 diabetes mellitus and autoimmune thyroid and liver disease. Associations between coeliac disease and several other conditions including Down's syndrome, Turner's syndrome, and schizophrenia have been reported. The possible excess of coeliac disease in patients with irritable bowel syndrome and the need to test for the condition in the routine investigation of intermittent abdominal pain and bloating remain controversial. In a recent treatment trial of irritable bowel syndrome in the United Kingdom the prevalence of coeliac disease was 0.7%, similar to the population mean.11 Coeliac disease is associated with excess mortality, including an increased risk of non-Hodgkin's lymphoma and other cancers.12

    A gluten-free diet not only protects against non-Hodgkin's lymphoma, corrects anaemia, and restores normal nutritional and biochemical status, but it also substantially improves quality of life, particularly if troublesome gastrointestinal symptoms have been present.13

    What then are the important messages for primary care? Firstly, patients with unexplained diarrhoea, anaemia, weight loss, infertility, recurrent miscarriage and low birthweight babies, or low bone mineral density are at increased risk of coeliac disease and should be investigated in primary care with antibody tests. Secondly, risk of coeliac disease is higher in patients with a first degree relative with the condition and in those with other autoimmune disorders such as type 1 diabetes (with which coeliac disease sometimes shares the HLA markers DQ2 and DQ8) and autoimmune thyroid and liver disease, and serum antibody testing should be considered. Patients with gastrointestinal symptoms strongly suggestive of coeliac disease who test negative should be referred for a specialist opinion. In all patients with positive antibodies the diagnosis of coeliac disease must be confirmed by endoscopic biopsy, for which specialist referral will also usually be required. Thirdly, while a lifelong gluten-free diet can reverse the effects of gluten enteropathy, the diagnosis, once again, must be confirmed by a small intestinal biopsy. Hopper and colleagues' study offers an attractive algorithmic approach to identifying coeliac disease, and evaluation of this algorithm in a primary care setting is now required.


    • Competing interests: None declared.

    • Provenance and peer review: Commissioned; not externally peer reviewed.