Where Is Celiac Disease Coming From and Why?

Catassi, Carlo

Journal of Pediatric Gastroenterology & Nutrition: March 2005 - Volume 40 - Issue 3 - pp 279-282

Department of Pediatrics, Università Politecnica delle Marche, Ancona, Italy; Center For Celiac Research, University of Maryland, Baltimore, USA

Correspondence should be addressed to: Prof. Carlo Catassi, University Department of Pediatrics, Via F. Corridoni 11, 60123 Ancona, Italy, e-mail catassi@tin.it, Fax +39 071 36281, Tel +39 071 5962364

Article Outline

In 1975, Charlotte Anderson, in her classic textbook on Paediatric Gastroenterology, wrote that “The typical child with celiac disease is usually fair-haired and blue-eyed…” (1). Pioneer epidemiologic studies apparently enforced the concept of the “white face of CD” by reporting the highest world incidence of celiac disease (CD) from Western Ireland, Austria and Sweden (2). Like snow in the sun, these views melted away in the light of the subsequent developments of CD epidemiology.

Sensitive serological tools for CD screening (first the antigliadin antibodies, and then the antiendomysial EMA and antitransglutaminase antibodies) became available during the 1980s and 1990s. Disease frequency could then be mapped in terms of prevalence (ratio of affected individuals to overall population) rather than incidence (new diagnoses per unit population per year), as even atypical and silent cases of CD were disclosed by serological screening. Not only was it demonstrated that CD is one of the most common, lifelong disorders affecting around 1% of the general population in Europe (3-4) and other countries of mostly European origin such as the USA (5), Argentina (6), Brazil (7) and Australia (8), but the celiac condition was also unexpectedly found to be frequent in several developing countries populated by non-Caucasian individuals. The Saharawi is a black-eyed, black-haired African population of Arab-Berber origin living in the Western Sahara. They hold the sad Guinness record of the highest CD frequency in the world: 5.6%, which is almost five to ten times the frequency observed in Europe (9). In Saharawi children the predominant clinical picture of CD is typical and the risk of dying from severe diarrhoea and dehydration is considerable, especially during the summer. CD is being more and more frequently recognized in India where presentation in children is usually “hypertypical” with chronic diarrhoea, anaemia and stunting (10).

In this issue of the Journal, Imanzadeh, et al. from the Mofid Children's Hospital of Tehran, Iran, report that CD is a common finding among Iranian children admitted to the hospital because of chronic diarrhea (11). This disorder was diagnosed in 6.5% of 825 cases investigated during the years 1997-2003. It is noteworthy that the frequency of CD was also elevated (0.8%) in the control group (825 age- and sex-matched children). The mean age of the study-group was high (8.5 years), suggesting that many of these cases presented with a longstanding history. The clinical picture was characterized by the typical pattern of gastrointestinal symptoms (chronic diarrhoea, weight loss, vomiting, etc), laboratory (iron deficiency, sometimes elevated aminotransferase levels plus EMA positivity), and biopsy findings (severe villous atrophy in most cases or, less frequently, a minimal change enteropathy). Most patients (87%) showed an excellent clinical and laboratory response after only 6 months of treatment with the gluten-free diet (GFD).

The interest in CD is currently very high in Iran. Specific sessions were dedicated to this condition at the 2001 and 2003 Meetings of the Iranian Society for Gastroenterology. Other published works have shown that CD is a primary health problem in that country, both in children and in adults. One of the highest world prevalences of CD in blood donors was recently found in 2,000 apparently healthy urban Iranian blood donors (1:166) by Shahbazkhani, et al. (12). The same, very active, “Teheran group” also reported an increased prevalence of CD in 250 patients with type 1 diabetes (2.4%) (13), in adults with irritable bowel syndrome-like symptoms (12%) (14), and again in 100 children with chronic diarrhoea (20%) (15). In a recent review, Rostami et al. underlined that wheat has been a major component of the Iranian diet for many centuries. Therefore the GFD represents a real challenge to both patients and clinicians in this area, especially due to the absence of any supply of GF products in Middle Eastern countries (16).

Besides Iran, CD seems to be frequent in other “silk road” countries. We are currently performing a screening project on school children in Cairo City, Egypt (17). The preliminary prevalence of CD in a sample of Egyptian students is 12 on 1531, i.e. 0.78% (95% CI = 034-1.22). In Turkey CD is a very common cause of isolated short stature in school-age children (18). In Israel, CD was said to be relatively common among the Jewish population, but considered to be rarer among the Arabs (19). In reality, a recent serological screening indicated that CD is likely to be as frequent in the Arab population as in the Jewish, as the previously reported difference in disease incidence was due to under-diagnosis in Arabs (20). In a study from Jordan the high incidence of CD was related to the high wheat consumption of the population (135 Kg/head/year) (21). In 123 Arabian patients with type I diabetes Al-Ashwal and Co-workers found an 8.1% prevalence of associated CD (22). In Iraq, Saudi Arabia and Kuwait CD accounts for a large proportion of cases with chronic diarrhea in adults and children (23-25).

In the first place, these studies highlight the need for increased awareness of CD in Middle Eastern countries, as well as in other parts of the developing world. This disorder could be responsible for a burden of morbidity and childhood mortality that is potentially amenable to treatment. Furthermore, the impact of CD is likely to increase in these countries, as the Western-type diet, which is rich in gluten-containing foods like pasta, pizza and hot dogs, becomes more and more popular. Coping with CD treatment in a developing country can be frustrating, due to the general scarcity of food, lack of GF products specifically designed for CD treatment, absence of patients' groups and, above all, poor knowledge of the principles of the GFD.

Besides these practical implications, the new “global” epidemiology of CD seriously questions an old theory on the history of gluten intolerance. How can CD be put into the perspective of the evolution of mankind and the history of agriculture? Doubtless CD did not exist during the Paleolithic age, as the diet of hunter-gatherers was “GF by origin” consisting of only meat, vegetables, seeds and fruits. The domestication of wheat and other gluten-containing cereals, the so-called Neolithic revolution, started some 10,000 years (10 KYA) ago in the “Fertile crescent”, a Near East area including regions of current Turkey (the Karacadag region), Iran, Iraq, Israel/Palestine, Syria and Lebanon. The first wheat varieties to be successfully domesticated were einkorn and emmer wheat. According to Ammermann and Cavalli-Sforza (26), the progressive spread of agriculture from the Near East to Europe took place through the migration of farmers mixing with and partially replacing the indigenous European population (the so-called demic diffusion). In this model the agricultural spread was stimulated by population growth (itself a result of the increasing availability of food) and local migratory activity. Based on the overall archaeological picture of the spread of agriculture in Europe, the wave of its advance apparently moved with a migratory rate of 1 Km/y, reaching the Atlantic side of Europe (e.g. the U.K. and Ireland) “only” 4,000 years ago (27).

During the eighties Simoons theorized that this pattern of agriculture spreading explained the higher CD incidence observed at that time in some Western countries, particularly the West of Ireland (28). He hypothesized that the HLA B8 antigen (the first HLA antigen known to be associated with CD) was once prevalent in pre-Neolithic Europe. Mapping the prevalence of the HLA-B8 antigen across Europe he noticed a south east - north west gradient, with a higher antigen frequency in areas reached later by the “farmers”. The prevalence of this antigen shows indeed a greater than three-fold variation across Europe, with lower values (10%) in South-Eastern countries (e.g. Italy and Greece), intermediate frequency (20%) in continental Europe (e.g. Germany and France) and higher values (30% and more) in the North-West part of the continent (U.K. and Ireland). The post-Neolithic HLA-B8 distribution might be the consequence of the negative selection pressure induced by consumption of gluten-containing cereals. The higher HLA-B8 frequency would be responsible for the apparently higher CD frequency in North Western European countries. In other words, possession of CD genes would confer a disadvantage to those living in areas of high cereal consumption, accounting for the inverse relationship between the frequency of CD in an area and the length of time since the introduction of agriculture.

Simoons' hypothesis, still quoted in recent publications (29), apparently does not survive recent genetic and epidemiologic developments, for the following reasons.

1. The HLA-B8 gene is not directly involved in CD predisposition, as it is simply an “innocent bystander” associated with the “culprit” HLA-D genes because of linkage disequilibrium (the so-called “extended haplotype” B8-DR3-DQ2). Although the geographic distribution of the D locus antigens is less well characterized, it appears that there is only a milder South-East to North-West gradient of the DR3 antigen frequency with a two-fold variation in prevalence. The map of the HLA-DQ2 and -DQ8 antigens, the strongest CD-associated genes (30), is even less clear. No South-East to North-West gradient has consistently been found in their cumulative frequency. For example, the DQ2 genotype, which is found in almost 90% of celiacs, is more common in Turkey (34%) (31) than in the U.K (23%) (32). The HLA-DQ2 and DQ8 genotypes are more common in genetically isolated populations, such as the Saharawi (41.6% DQ2 and/or DQ8 positive in the general population) and the Sardinians (9).

2. The overall prevalence of CD does not show any decreasing trend from European to Middle Eastern countries, which was instead to be expected according to Simoons' theory. The majority of epidemiologic data nowadays available indicates that CD prevalence is mostly stable across European and Near Eastern countries (0.5-0.8% of the general population), with some exceptions. The previously reported differences in the incidence were spurious and caused by the variable clinical presentation and awareness of the disease. CD is only more common in genetically isolated populations, such as the Saharawis (5.6%), the Finnish (1.5%), and the Sardinians (1%) (4,9).

By comparing the maps of HLA CD-predisposing genes and CD prevalence on a worldwide basis, it is evident that these two variables do not diverge but rather run parallel. The HLA-DR3 gene (in linkage disequilibrium with the DQ2 in cis position - the most common CD predisposing genotype) is more common where wheat is a staple food, e.g. Europe, North Africa and Middle East countries. On the other hand, it is virtually absent in populations traditionally consuming GF cereals, such as the Japanese, the Cayapa and Kogi (Native Americans) and the Polynesians (F. Cucca, personal communication). This is a puzzling finding that could be explained by the dynamic interplay between genetic and environmental factors, as speculated below (9).

There is no doubt that following the Neolithic revolution, gluten-containing cereals progressively become an important food for farmers, originally in the Middle East and later on in Europe. However, there are some major dietary differences between the “Neolithic” and the “current” scenario that need to be taken into account. First, the overall gluten intake was probably different in ancient times. Furthermore, infant feeding was characterized by prolonged breast feeding and delayed introduction of gluten-containing cereals. Since the degree of mucosal damage is related to the amount and the timing of gluten ingestion, the enteropathy of the “ancestral” CD-prone individuals (HLA-DQ2 and DQ8 positive) was presumably milder. In terms of genetic fitness this mild enteropathy did not represent a major selective disadvantage or might even have been favorable, therefore leading to the positive selection of CD-predisposing genes. It could be protective against intestinal infections because a moderately atrophic jejunal mucosa partially lacks the membrane receptors required for microorganism adhesion (33). The inflammation accompanying the “minimal change” celiac enteropathy could provide a source of reactive lymphomonocytes in the intestinal mucosa, further increasing the protective effect against intestinal infections. Thousands of years later, gluten consumption has dramatically changed. In this modified context the CD-prone genotype is no longer neutral or protective, but is instead harmful. The higher gluten intake causes severe damage to the small intestinal mucosa which is responsible in turn for intestinal malabsorption and extra-intestinal complications.

In conclusion, it is nowadays clear that CD is a primary health problem in Iran as well as in many other Near and Middle Eastern countries. An increased awareness of this condition is urgently required in that area of the world, as CD is potentially responsible for significant morbidity and childhood mortality. The high prevalence of CD in areas belonging to the Fertile Crescent, like Iran, is at variance with the proposed inverse relationship between CD frequency and the length of time that has elapsed since the introduction of agriculture. A detailed analysis of the geographical distribution of genes that predispose to CD, particularly the HLA-DQ2 alleles, might help to clarify the evolutionary history of this complex disorder.

Back to Top | Article Outline


The evolutionary hypothesis has been drafted by the author and Prof. F. Cucca, from the Dipartimento di Scienze Biomediche e Biotecnologie, Universita' di Cagliari, Italy. The author thanks Mrs. Susan Phillips for kindly revising the manuscript.

Back to Top | Article Outline


1. Anderson CM, Burke V. Paediatric Gastroenterology. Blackwell Scientific Publications, Oxford: 1975; 175.
2. Logan RFA. Epidemiology of celiac disease. In: Marsh MN (Ed). Coeliac Disease. Blackwell Scientific Publications: Oxford 1992; 192-214.
3. Catassi C, Fabiani E, Rätsch IM, Coppa GV, Giorgi PL, Pierdomenico R, et al. The coeliac iceberg in Italy. A multicentre antigliadin antibodies screening for coeliac disease in school-age subjects. Acta Paediatr Suppl 1996;412:29-35.
4. Mäki M, Mustalahti K, Kokkonen J, Kulmala P, Haapalahti M, Karttunen T, et al. Prevalence of celiac disease among children in Finland. N Eng J Med 2003;348:2517-24.
5. Fasano A, Berti I, Gerarduzzi T, Not T, Colletti RB, Drago S, et al. Prevalence of celiac disease in at-risk and non at-risk groups. A large, multicentre study. Arch Intern Med 2003;163:286-92.
6. Gomez JC, Selvaggio GS, Viola M, Pizarro B, la Motta G, de Barrio S. Prevalence of celiac disease in Argentina: screening of an adult population in the La Plata area. Am J Gastroenterol 2001;96:2700-4.
7. Gandolfi L, Pratesi R, Cordoba JC, Tauil PL, Gasparin M, Catassi C. Prevalence of celiac disease among blood donors in Brazil. Am J Gastroenterol 2000;95:689-92.
8. Hovell CJ, Collett JA, Vautier G, Cheng AJP, Sutanto E, Mallon DF, et al. High prevalence of coeliac disease in a population-based study from Western Australia: a case for screening? MJA 2001;175:247-50.
9. Catassi C, Doloretta Macis M, Rätsch IM, De Virgilis S, Cucca F. The distribution of DQ genes in the Saharawi population provides only a partial explanation for the high celiac disease prevalence. Tissue Antigens 2001;58:402-5.
10. Patwari AK, Anand VK, Kapur G, Narayan S. Clinical and nutritional profile of children with celiac disease. Indian Pediatr 2003;40:337-42.
11. Imanzadeh F, Sayyari AA, Yaghoobi M, Akbari MR, Shafagh H, Farsar AR. Celiac disease in children with diarrhea is more frequent than previously suspected. J Pediatr Gastroenterol 2005;40:309-311.
12. Shahbazkhani B, Malekzadeh R, Sotoudeh M, Fayaz Moghadam K, Farhadi M, Akbari R, et al. High prevalence of celiac disease in apparently healthy Iranian blood donors. Eur J Gastroenterol Hepatol 2003;15:475-8.
13. Shahbazkhani B, Faezi T, Akbari MR, Mohamadnejad M, Sotoudeh M, Rajab A, et al. Coeliac disease in Iranian type I diabetic patients. Dig Liver Dis 2004;36:191-4.
14. Shahbazkhani B, Forootan M, Merat S, Akbari MR, Nasserimoghadam S, Vahedi H, et al. Coeliac disease presenting with symptoms of irritable bowel syndrome. Aliment Pharmacol Ther 2003;18:231-5.
15. Shahbazkhani B, Mohamadnejad M, Malekzadeh R, Akbari MR, Esfahani MM, Nasseri-Moghaddam S, et al. Coeliac disease is the most common cause of chronic diarrhoea in Iran. Eur J Gastroenterol Hepatol 2004;16:665-8.
16. Rostami K, Malekzadeh R, Shahbazkhani B, Akbari MR, Catassi C. Celiac disease in Middle East countries: a challenge for the evolutionary history of this complex disorder ? Dig Liver Dis 2004;36:694-7.
17. Fasano A, Catassi C, Kryszak D, Abou-Zekri ME. Prevalence of celiac disease among school age children in Egypt. Preliminary results of a pilot study (abstract). J Pediatr Gastroenterol Nutr 2005;40 Suppl 1.
18. Altuntas B, Kansu A, Ensari A, Girgin N. Celiac disease in Turkish short-statured children and the value of antigliadin antibody in diagnosis. Acta Paediatr Jpn 1998;40:457-60.
19. Granot E, Korman SM, Sallon S, Deckelbaum RJ. “early” vs. “late” diagnosis of celiac diusease in two ethnic groups living in the same geographic area. Isr J Med Sci 1994;30:271-5.
20. Shamir R, Lerner A, Shinar E, Lahat N, Sobel E, Bar-or R, et al. The use of a single serological marker underestimates the prevalence of celiac disease in Israel: a study of blood donors. Am J Gastroenterol 2002;97:2589-94.
21. Rawashdeh MO, Khalil B, Raweily E. Celiac disease in Arabs. J Pediatr Gastroenterol Nutr 1996;23:415-8.
22. Al-Ashwal AA, Shabib SM, Sakati NA, Attia NA. Prevalence and characteristics of celiac disease in type I diabetes mellitus in Saudi Arabia. Saudi Med J 2003;24:1113-5.
23. Al-Bayatti SM. Etiology of chronic diarrhea. Saudi Med J 2002;23:675-9.
24. Shaltout AA, Khuffash FA, Hilal AA, el Ghanem MM. Pattern of protracted diarrhoea among children in Kuwait. Ann Trop Pediatr 1989;9:30-2.
25. Rastogi A, Malhotra V, Uppal B, Aggarwal V, Kalra KK, Mittal SK. Aetiology of chronic diarrhoea in tropical children. Trop Gastroenterol 1999;20:45-9.
26. Ammerman AJ, Cavalli-Sforza LL. Neolithic transition and the genetics of populations in Europe. Princeton University Press: Princeton (New Jersey) 1984.
27. Jobling MA, Hurles ME, Tyler-Smith C. Human evolutionary genetics. Garland Sciences: New York 2004.
28. Simoons FJ. Celiac disease as a geographic problem. In: Food, nutrition and evolution. Walcher DN and Kretchmer N (Eds). Masson: New York 1981; pp 179-99.
29. Cronin CC. Shanahan F. Why is celiac disease so common in Ireland ? Perspect Biol Med 2001;44:342-52.
30. Louka AS, Sollid LM. HLA in coeliac disease: unravelling the complex genetics of a complex disorder. Tissue Antigens 61:105-17.
31. Tumer L, Altuntas B, Hasanoglu A, Soylemezoglu O, Arinsoy T. Pattern of human leukocyte antigens in Turkish children with celiac disease. Pediatr Int 2000;42:678-81.
32. Howell WM, Leung ST, Jones DB, Nakshabendi I, Hall MA, Lanchbury JS, et al. HLA-DRB, -DQA, and -DQB polymorphism in celiac disease and enteropathy-associated T-cell lymphoma. Common features and additional risk factors for malignancy. Hum Immunol 1995;43:29-37.
33. Kerneis S, Chauviere G, Darfeuille-Michaud A, Aubel D, Coconnier MH, Joly B, Servin AL. Expression of receptors for enterotoxigenic Escherichia coli during enterocytic differentiation of human polarized intestinal epithelial cells in culture. Infect Immun 1992;60:2572-80.
© 2005 Lippincott Williams & Wilkins, Inc.