Q J Med 2002; 95: 75-77
© 2002 Association of Physicians
The hunt for coeliac disease in primary care
1 From the Department of Internal Medicine and 2 Research Unit, Tampere University Hospital, and 1Medical School and 2Public Health School, University of Tampere, and 3 Tampere Health Centre, Tampere, Finland
Received 9 May 2001 and in revised form 16 November 2001
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
Background: Serological screening suggests that most coeliac disease goes undetected.
Aim: To determine how often coeliac disease is found in patients referred for gastroscopy by general practitioners.
Design: Testing of 9971 consecutive patients referred to a single health centre for their first open-access gastroscopy over a 10-year period, without previously-diagnosed coeliac disease or dermatitis herpetiformis.
Methods: Endoscopy and routine duodenal biopsy were in use throughout the study period. The occurrence of coeliac disease was evaluated in patients with dyspepsia, in those with regurgitation, and in those with symptoms suggestive of coeliac disease.
Results: Altogether, 147 (1.47%) patients had coeliac disease: 18/2974 (0.61%) with regurgitation, 41/5347 (0.77%) with dyspepsia, and 88/1650 (5.33%) with suspected coeliac disease. Increasing age reduced the risk very slightly (OR 0.98; 95%CI 0.97–0.99). Risk was significantly increased when there was a suspicion of coeliac disease (OR 9.085; 95%CI 5.371–15.369), while no difference was found between patients with dyspepsia and those with reflux disease (OR 1.33; 95%CI 0.75–2.34). The risk in women was not increased (OR 0.97; 95%CI 0.69–1.38).
Discussion: In this primary-care setting, <1% of patients with upper abdominal symptoms suffered from coeliac disease; population-based serological screening is expected to yield a corresponding prevalence. When physicians had suspected coeliac disease, the condition was nine times more common.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Anaemia, diarrhoea and loss of weight are considered the most typical symptoms in adult coeliac disease. However, patients often suffer from subtle and trivial symptoms, or the disease may even be clinically silent. A number of studies indicate that the prevalence of coeliac disease is as high as 0.5–1.0% when serological screening is applied to individuals without any suspicion of the disease.1–3 On the other hand, the observed frequency of the disease is virtually everywhere lower, indicating that the majority of patients still remain undetected. The experience and skills of general practitioners would appear to determine how effectively coeliac disease is uncovered in the area.4
Dyspepsia can be defined as discomfort or pain in the upper abdomen, whereas heartburn and regurgitation are common symptoms in gastro-oesophageal reflux disease (GERD).5 Most of these conditions are managed in primary care. The reported frequency of coeliac disease in such patients has been 0.5–4.0%,6–8 but it must be noted that many studies have been carried out in hospital clinics.
Provided that a significant number of individuals with subtle gastrointestinal symptoms suffer from coeliac disease, further studies should be considered even in the absence of ‘alarm’ symptoms. The aim of the present study was therefore to assess the frequency of coeliac disease in patients complaining of common gastrointestinal symptoms in primary care. The results were compared to those obtained among patients with a suspicion of coeliac disease. This ten-year study was carried out in a health centre, where open-access endoscopy has been easily available to every family physician, and where this service was widely used.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
The study was done at Tampere Health Centre from January 1990 to June 2000. It involved all patients who were referred for upper gastrointestinal endoscopy by family physicians. Open-access endoscopy was in use throughout the study period. A total of 12000 examinations were done during this period, which is 7.5% of the entire adult population in the area. Patients with previously diagnosed coeliac disease or dermatitis herpetiformis were excluded from analysis, as were individuals undergoing repeat endoscopies.
Prior to endoscopy, patients were categorized into three study groups. The first comprised patients with regurgitation or heartburn (GERD group). We thought (although the issue is debatable) that patients with GERD would represent the normal population in terms of the risk of coeliac disease. The risk of coeliac disease was surmised to be moderate in the second group, i.e. subjects suffering from upper gastrointestinal symptoms such as abdominal pain or discomfort (dyspepsia group). The third group consisted of subjects with a great likelihood of coeliac disease, namely individuals suffering from diarrhoea, weight loss or anaemia, and those referred specifically for duodenal biopsy due to coeliac suspicion, i.e. those with a positive family history or known positive coeliac disease serology (coeliac suspicion group). Upon endoscopy, two to four biopsy specimens were taken from the distal part of the duodenum in each patient. The diagnosis of coeliac disease was based on the subtotal or severe partial villous atrophy and crypt hyperplasia, and on clinical or histological recovery on a gluten-free diet. This means that all coeliac patients fulfilled the current ESPGAN criteria.9
The study plan was approved by the Ethical Committee of Tampere University Hospital.
Statistical methods
As a multivariate technique, we used logistic regression analysis. The results are summarized using odds ratios (OR) with 95% confidence intervals (95%CI). The computation was carried out using SPSS/Win (Version 10.0) software.
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
The study cohort comprised 9971 patients: 6297 women and 3674 men. Their median age was 58 years (range 12–93). Altogether, 147 new coeliac cases were found, i.e. 1.47% of the entire study population. In group 1 (GERD), coeliac disease was found in 18 (0.61%) of 2974, in group 2 (dyspepsia) in 41 (0.77%) of 5347. In group 3 (patients with suspected coeliac disease), the diagnosis was made in 88 (5.33%) of 1650.
Coeliac disease was found equally often in men (1.47%) and women (1.48%). Female gender was thus not a risk factor for coeliac disease (OR 0.973; 95%CI 0.688–1.376). Increasing age reduced the risk slightly (OR 0.982; 95%CI 0.973–0.991). There was no significant difference between group 2 (dyspepsia) and group 1 (GERD) (OR 1.326; 95%CI 0.752–2.339). The risk was significantly increased in group 3 (suspicion of coeliac disease), (OR 9.085; 95%CI 5.371–15.369).
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
The purpose of this study was to assess the necessity of looking rigorously for coeliac disease in individuals presenting with non-specific and often vague upper gastrointestinal complaints. Since both serological screening and visible duodenal markers in endoscopy fail to detect 20% or even more coeliac patients,7,8,10 small-bowel biopsy as a gold standard was taken for each patient undergoing endoscopy. Our unique long-term endoscopy facilities in general practice enabled us to study patients who all were enrolled by family physicians. It has recently been stressed that such an approach often provides a better basis for daily practice.11
A confounding factor was that serological screening tests for coeliac disease were available throughout the study period. This will undoubtedly produce some selection bias in group 3 (coeliac suspicion). We do not know exactly how frequently family physicians were prescribing these tests, but they were told to use them only when considering coeliac disease. On the other hand, the relatively low detection rate for coeliac disease (5%) in this group suggests that only a minority of patients underwent antibody screening prior to endoscopy.
The main attention should be focused on groups 1 and 2, where there was no apparent suspicion of coeliac disease. It would be reasonable to assume that in patients with dyspepsia, coeliac disease would be more common than in an unselected population. Instead, and to our surprise, the risk was only 0.7%, and virtually the same as in GERD. Dyspepsia and GERD are a considerable burden on health-care resources. The rate of detection of coeliac disease in these two conditions did not exceed that yielded in screening studies. It is also notable that, contrary to the common view, women were not at increased risk of coeliac disease.
By comparison, in more or less selected hospital series, coeliac prevalences of 4.7% (7/150)7 and 1.2% (6/517)8 have been reported. Heikkinen et al.6 found a coeliac prevalence similar to ours (0.5%) in primary-care patients. However, the number of participants was lower (400), and only two coeliac cases were found.
Our findings further imply that routine small-bowel biopsy upon endoscopy is of limited benefit; however, 59 coeliac patients would have remained undetected had routine duodenal biopsy not been done. Moreover, the information that coeliac disease has been excluded by biopsy is often valuable. We therefore recommend duodenal biopsy, provided its additional cost is low. An alternative approach is to biopsy only patients with visible loss of duodenal folds; this method was, however, not evaluated in the present study.
Coeliac disease is relatively rare in patients contacting family physicians because of upper abdominal symptoms. Small-bowel biopsy seems not to be necessary in these subjects unless there is additional evidence of coeliac disease.
|
Acknowledgments |
|---|
The study was supported by the Medical Research Fund of Tampere University Hospital, the Emil Aaltonen Foundation, and the Finnish Medical Society Duodecim.
|
Notes |
|---|
Address correspondence to Dr P. Collin, Medical School, FIN 33014 University of Tampere, Finland. e–mail: pekka.collin{at}uta.fi
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
1. Johnston SD, Watson RGP, McMillan SA, Sloan J, Love AHG. Prevalence of coeliac disease in Northern Ireland. Lancet1997; 350:1370.[Web of Science][Medline]
2. Kolho K-L, Färkkilä MA, Savilahti E. Undiagnosed coeliac disease is common in Finnish adults. Scand J Gastroenterol1998; 33:1280–3.[Web of Science][Medline]
3. Ivarsson A, Persson LÅ, Juto P, Peltonen M, Suhr O, Hernell O. High prevalence of undiagnosed coeliac disease in adults: a Swedish population-based study. J Intern Med1999; 245:63–8.[Web of Science][Medline]
4.
Hin H, Bird G, Fisher P, Mahy N, Jewell D. Coeliac disease in primary care, case finding study. Br Med J1999; 318:164–7.
5.
Talley NJ, Stranghellini V, Heading RC, Kock KI, Malagelada JR, Tytgat GNJ. Functional gastroduodenal disorders. Gut1999; 45(Suppl. 2):II 37–42.
6. Heikkinen M, Pikkarainen P, Takala J, Räsänen H, Julkunen R. Etiology of dyspepsia: four hundred unselected consecutive patients in general practice. Scand J Gastroenterol1995; 30:519–23.[Web of Science][Medline]
7. Dickey W, Hughes D. Prevalence of celiac disease and its endoscopic markers among patients having routine upper gastrointestinal endoscopy. Am J Gastroenterol1999; 94:2182–6.[Web of Science][Medline]
8.
Bardella MT, Minoli G, Ravizza D, Radaelli F, Velio P, Quatrini M, et al. Increased prevalence of celiac disease in patients with dyspepsia. Arch Intern Med2000; 160:1489–91.
9.
Walker-Smith JA, Guandalini S, Schmitz J, Shmerling DH, Visakorpi JK. Revised criteria for diagnosis of coeliac disease. Arch Dis Child1990; 65:909–11.
10. Rostami K, Kerckhaert J, Tiemessen R, Meijer JWR, Mulder CJJ. The relationship between anti-endomysium antibodies and villous atrophy in coeliac disease using both monkey and human substrate. Eur J Gastroenterol Hepatol1999; 11:439–42.[Web of Science][Medline]
11.
Green LA, Dovey SM. Practice based primary care research networks. Br Med J2001; 322:567–8.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. D Hopper, S. S Cross, D. P Hurlstone, M. E McAlindon, A. J Lobo, M. Hadjivassiliou, M. E Sloan, S. Dixon, and D. S Sanders Pre-endoscopy serological testing for coeliac disease: evaluation of a clinical decision tool BMJ, April 7, 2007; 334(7596): 729 - 729. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. R. Locke III, J. A. Murray, A. R. Zinsmeister, L. J. Melton III, and N. J. Talley Celiac Disease Serology in Irritable Bowel Syndrome and Dyspepsia: A Population-Based Case-Control Study Mayo Clin. Proc., April 1, 2004; 79(4): 476 - 482. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||