http://informahealthcare.com/gas ISSN: 0036-5521 (print), 1502-7708 (electronic) Scand J Gastroenterol, 2015; 50(7): 824–831 DOI: 10.3109/00365521.2015.1010571

ORIGINAL ARTICLE

Screening for celiac disease in Danish adults Anna Horwitz1,2, Tea Skaaby1, Line Lund Kårhus1, Peter Schwarz3,4, Torben Jørgensen1,4,5, Ju€ri J. Rumessen6 & Allan Linneberg1,4,7 1

Research Centre for Prevention and Health, The Capital Region, University of Copenhagen, Copenhagen, Denmark, 2Department of Neuroscience and Pharmacology, Center for Healthy Ageing, University of Copenhagen, Copenhagen, Denmark, 3Research Centre for Ageing and Osteoporosis, Departments of Medicine M and Diagnostics, Glostrup Hospital, University of Copenhagen, Copenhagen, Denmark, 4Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark, 5Faculty of Medicine, Aalborg University, Aalborg, Denmark, 6HR-Research Unit and Department of Gastroenterology, Gentofte Hospital, University of Copenhagen, Copenhagen, Denmark, and 7Department of Clinical Experimental Research, Glostrup University Hospital, Glostrup, Denmark Abstract

Key Words:

Objective. The prevalence of celiac disease (CD) as recorded in the Danish National Patient Registry is ~50/100,000 persons. This is much lower than the reported prevalence of CD in other Nordic countries and underdiagnosis is suspected. Our aim was to estimate the prevalence of CD in a population-based study of Danish adults. Methods. A total of 2297 adults aged 24–76 years living in the southwestern part of Copenhagen were screened for CD by immunoglobulin (Ig)A and IgG antibodies to transglutaminases and deamidated gliadin. IgA/IgG-positive participants were invited to a clinical evaluation, including biopsies, by a gastroenterologist. Results. Of the invited 56 participants, 40 underwent a full clinical evaluation and 8 persons were diagnosed with CD; 2 of the 16 persons, who did not complete the clinical evaluation, were considered by experts to have probable CD. None of the above 56 participants had a known history of CD or a recorded diagnosis of CD in National Patient Registry. By combining cases of biopsy-proven CD (n = 8), probable CD (n = 2), and registry-recorded CD (n = 1), the prevalence of CD was estimated to be 479/100,000 (11/2297) persons (95% CI: 197–761). Conclusion. In this general adult population, the prevalence of CD as estimated by screening and clinical evaluation was 10 times higher than the registry-based prevalence of CD. Of 11 participants diagnosed with CD in our screening study, 10 were unaware of the diagnosis prior to the study. Thus, our study suggests that CD is markedly underdiagnosed in Danish adults.

Celiac disease, epidemiology, prevalence, screening

Introduction Celiac disease (CD) is an autoimmune disease triggered by gluten in genetically susceptible individuals. CD is characterized by gastrointestinal symptoms, macroscopic and microscopic changes in the small bowel mucosa, malabsorption, and a wide range of extraintestinal manifestations [1]. The diagnosis of CD is based on determination of CD-specific biomarkers (antibodies), histological examination of duodenal biopsies, and improvement following initiation of gluten-free diet [2]. The reported prevalence of CD varies substantially and large differences have been observed even within short geographical distances [3]. Scandinavia is regarded as a high-prevalence area. In Sweden, CD affects ~1000–3000/ 100,000 children [4,5]; 530/100,000 adults; and 270/100,000 blood donors [6,7]. One study from Norway

History Received 19 December 2014 Revised 12 January 2015 Accepted 19 January 2015 Online 17 February 2015

screened healthy blood donors and found a prevalence of 290/100,000 persons [8]. A Finnish screening study in adults found that the prevalence was 1740 and 1240/100,000 persons as assessed by CD-specific antibodies and biopsy, respectively [9]. No previous study has screened for CD by using CD antibodies and clinical evaluation, including endoscopy, in a Danish general adult population. A Danish registry-based study including all Danish inhabitants (both children and adults) found that the recorded prevalence of diagnosed CD was 55/100,000 persons [10]. A similar nationwide registry-based study of all Danish children found that the prevalence of CD was 80/100,000 children [11]. A Danish registry-based study (based on hospital records) including all Danish adults in Copenhagen during the period between 1976 and 1991 reported a prevalence of 46/

Correspondence: Allan Linneberg, MD, PhD, Research Centre for Prevention and Health, Glostrup Hospital, University of Copenhagen, Ndr. Ringvej 57, building 84/85, DK-2600 Glostrup, Denmark. Tel: +45 38633260. E-mail: [email protected]  2015 The Author(s). Published by Taylor & Francis. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Screening for CD in adults

DOI: 10.3109/00365521.2015.1010571

825

Table I. Baseline characteristics of participants in the 5-year follow up of the Health2006 cohort study.

Gender Female Male Age at baseline 15–34 35–54 55+ Employment status Employed or self-employed Have been employed Have never been employed Smoking status Current daily smoker Occasional smoker Past smoker Never-smoker Anthropometry Waist circumference in cm BMI in kg/m2 Alcohol consumption Units per week in past 12 months Non-drinkers, past 12 months

Participants in follow up (n = 2308)

Non-participants in follow up (n = 1163)

% (n/n total) 54.2 (1250/2308) 45.8 (1058/2308) % (n/n total) 12.7 (294/2308) 47.8 (1104/2308) 39.4 (910/2308) % (n/n total) 76.6 (1745/2279) 22.3 (509/2279) 1.1 (25/2279) % (n/n total) 17.4 (399/2290) 3.6 (82/2290) 34.2 (784/2290) 44.8 (1025/2290) Mean (95% CI) 88.0 (87.5–88.5) 25.7 (25.5–25.9) Median (IQR) 7 (3–14) % (n/n total) 3.9 (88/2285)

% (n/n total) 57.4 (668/1163) 42.6 (495/1163) % (n/n total) 20.4 (237/1163) 43.8 (509/1163) 35.9 (417/1163) % (n/n total) 68.2 (775/1137) 29.1 (331/1137) 2.7 (31/1137) % (n/n total) 32.6 (374/1147) 2.7 (31/1147) 29.0 (332/1147) 35.8 (410/1147) Mean (95% CI) 90.1 (89.2–90.9) 26.4 (26.1–26.7) Median (IQR) 6 (2–13) % (n/n total) 8.2 (93/1141)

p-Value p = 0.067 # p < 0.001 #

p < 0.001 #

p < 0.001 #

p