Celiac disease: India on the global map

See article in J. Gastroenterol. Hepatol. 2006; 21: 1622–1625. Celiac disease (CD), also called gluten-sensitive enteropathy, is an immune-mediated enteropathy that occurs in genetically susceptible individuals by the ingestion of gluten-containing grains (wheat, barley, and rye). It was previously believed that CD is a disease of Europe and Europeans living outside Europe. It was in 1966 that Walia et al.1 and Mishra et al.2 reported CD for the first time in children and adults, respectively, from India. Thereafter CD reports from India were few, with just three more publications appearing until 1993. This lag period was related to a lack of awareness at primary and secondary levels of healthcare, the existence of very few dedicated pediatric gastroenterology centers, the nonapplicability of cumbersome methods of jejunal biopsies, the stringent three-step ESPGAN diagnostic criteria and a great deal of reluctance on the part of doctors to accept the existence of CD in India. Tropical sprue (TS) and gastrointestinal tuberculosis as the common causes of chronic diarrhea and malabsorption overshadowed diagnosis of CD in children. The majority of cases of TS would show mild to moderate changes on small bowel histology and would therefore tip the balance in favor of a diagnosis of TS rather than CD, particularly following some response to antibiotic and vitamin therapy. Wider application and ease of obtaining endoscopic duodenal biopsies as a step in the algorithm for chronic diarrhea, and modifications in the ESPGAN criteria3 paved the way for diagnosis of CD in a larger number of patients from the northern region of India.4–14 On an epidemiological basis, CD in India is prevalent among predominantly wheat-consuming states such as Punjab, Haryana, Delhi, Rajasthan, Uttar Pradesh, Bihar and Madhya Pradesh (Fig. 1). The reasons behind this regional variation in prevalence within the same country are possibly those of genetic predisposition to CD and consumption of wheat as a staple diet in the affected region. Genetic studies from Uttar Pradesh15 and Delhi16 have shown that CD is associated with multiple DR3-DQ2 haplotypes, and is thus no different from European susceptibility patterns. The DR3 allele frequency of 14.9% from north India (Delhi) is comparable to south India (Tamil Nadu), with 14.34% among Yadhavas and 11.6% among Piramalai Kallars castes.17,18 However, a major difference in DQ2 allele frequency exists between the two regions; in north India it is 31.9%, and in south India it is 12.78% for Piramalai Kallars and 9% for Yadhavas.17,18 Thus the regional difference of CD occurrence in India is possibly linked to genetic differences coupled with variation in staple diet (wheat in north India and rice in south India). Celiac disease belt (dark area) in India, including the states of Punjab, Haryana, Delhi, Rajasthan and Uttar Pradesh. Celiac disease in India occurs more often in children1,4–11 than in adults.2,12–14 Over the past decade awareness of CD has increased substantially in India and, parallel to this, reported cases also show an upward trend both in children and adults. There were 130 CD cases reported from India in 1966–2000 compared to 517 in 2001–2005. One of the major factors leading to increased diagnosis is the discovery and subsequent use of CD serology [immunoglobulin A antiendomysial antibodies (IgA EMA) or IgA tissue transglutaminase antibodies (IgA tTG)] coupled with small bowel histology particularly in cases where the Marsh grades are less than IIIb.19 Partial villous atrophy with positive CD serology and response to gluten-free diet (GFD) would confirm a diagnosis of CD that would have been otherwise confused with a non-CD etiology. It was previously a matter of concern that adult CD was not recognized in India. However, over a period of time CD is increasingly being diagnosed among adults in this country. Factors that resulted in recent reports of adult CD from north India12–14 were existence of CD among Indian children, demonstration of genetic susceptibility (Indian population) in the presence of environmental insult with dietary wheat, reports of adult CD from Western countries, and availability of antibody testing to overcome diagnostic overlap with TS and tuberculosis. This may indicate that many of those so-called TS/unknown etiology/tuberculosis patients were in fact the cases of CD. TS is a major cause (39%) of sporadic malabsorption syndrome among Indian adults.20 The exact pathogenesis of TS is unknown; however, speculation has centered around small bowel bacterial colonization and overgrowth, and altered gut motility.21 Acute infective diarrheas affecting the small bowel may be followed by chronic small bowel diarrhea due to post-infective malabsorption or post-infective irritable bowel syndrome. However, significant villous atrophy that is known to occur in TS is rare following acute infective diarrhea. According to one study, CD constitutes 26% (35/137) of all malabsorption syndrome cases in Indian children.7 Of the 35 children with CD, 9% were younger than 2 years of age. In another study, CD comprised 16.6% of the 246 cases of chronic diarrhea.9 The article by Sood et al.22 in this issue of the Journal has addressed an important epidemiological question of the prevalence of CD among school children in India. The study has been carried out in Ludhiana district of the state of Punjab, which falls within the CD belt in northern India. This is the first CD field study from India and hence is valuable. In this study CD was confirmed in 14 of 4347 screened school children giving a disease frequency of 1 in 310 and a prevalence rate of 0.3%. This rate is comparable to average worldwide prevalence of 0.37% based on screening.23 Prevalence rates of 0.3–0.7% from a European population and 0.9% from the USA have been reported.23 The authors have rightly highlighted several limitations of their study. Only the high risk group (n = 198; diarrhea, short stature, failure to thrive, pallor and other miscellaneous conditions) was studied. Limiting the study only to the suspect group and not to all subjects possibly resulted in other cases being missed, such as those with asymptomatic disease/occult clinical signs and also a subset of children with probable atypical manifestations. Doing small bowel histology in all children with positive IgA tTG antibody as per the authors’ study design would have been helpful in further categorization of CD. Thus the prevalence rate calculated by the authors is clearly an underestimation of CD in school children. Another interesting point made by Sood et al.22 is that the disease was not suspected previously. This is well documented by previous studies from India (a mean delay in diagnosis by 3.4–5.9 years).9–11 Delayed presentation reflects the lack of CD awareness amongst Asian parents and pediatricians. In developing countries, illnesses such as recurrent gastrointestinal infections causing diarrhea, protein-calorie malnutrition and nutritional anemia occur frequently, and their symptoms resemble those of CD in the initial phase. In my clinical experience, most of the children with CD in India receive multiple courses of antimicrobial agents and even antitubercular treatment before reaching tertiary care centers. Therefore, general practitioners and pediatricians substantially contribute to delay in diagnosis. The authors also emphasized that GFD was instituted in some of their subjects without confirmation of CD by their previously treating doctors. The incorrect practice of starting GFD in suspected CD patients without proper confirmation has surfaced at some focal points in India. This has to be stopped in the larger interest of the community. The authors have also highlighted the point of dietary non-compliance. In our experience strict dietary compliance is observed in the majority (95%) of patients. This can be ensured only by providing proper explanation of the disease, imparting knowledge of the principles of GFD, using locally designed dietary charts containing food items with clear demarcation of segments of consumption and non-consumption columns, and reinforcing dietary compliance at frequent follow-up visits. Non-compliance is observed when children consume toffees or share the lunch box of other schoolmates, or in large joint families where supervision of meals is often by family members other than parents. We educate parents that compliance can be improved if all family members consume GFD may be once a week. On each follow-up visit we emphasize adherence to GFD. Among the other Asian countries, CD is reported from the west Asian countries such as Jordan, Kuwait, Iraq, Iran, Saudi Arabia, Oman and Pakistan.24–31 It is interesting to note that the disease is unreported from other South-East Asian countries, including China, Japan, Malaysia, Thailand and Bangladesh, where rice is the staple diet. It seems that India (especially north) has divided Asia into two parts: CD prone west and non-CD east. It is not clear whether this regional divide in CD prevalence is due to differences in consumption of staple diet (wheat/rice) or due to differences in genetic predisposition. However, on the basis of historical events it can be postulated that genetic mixing had taken place in the west Asian region due to successive invasions from Scandinavian countries and that could probably explain the high prevalence of CD. The gold standard of CD diagnosis is the modified ESPGHAN criteria.3 According to these criteria, only intestinal biopsy changes and response to GFD are sufficient to make diagnosis of CD. However, in developing countries many other conditions other than CD can give rise to villous atrophy. In developing countries, villous atrophy is not always synonymous with CD. However, in the experience of the author, subtotal villous atrophy (Marsh grade IIIb) is usually synonymous with CD in children. We need to have an additional criterion over and above the modified ESPGHAN criteria to distinguish CD from other causes of villous atrophy. A practical option is to use CD serology. In the presence of villous atrophy, a positive CD serology at initial diagnosis augments distinction of CD from other causes of villous atrophy. As of now tTG antibody or EMA are useful markers for the diagnosis of CD. In Asian countries, we would always suggest doing CD serology at diagnosis. In a given clinical setting of CD, positive antibody along with typical small intestinal mucosal changes and response to GFD would establish the diagnosis. Therefore, in Asia serological testing should be added to the modified ESPGHAN criteria to make a definitive diagnosis of CD.32 In conclusion, the myth that CD is the disease of only Europe and other Western countries no longer holds true. Among developing countries there is a greater need of awareness, focusing on recognition of all the subtypes of CD including typical, atypical and silent subjects, and all age groups including infants following supplementation with wheat-containing diet. The extent of CD in Asian countries seems to be huge. In fact the prevalence rate of CD may be higher than in any Western country. Further population studies addressing both symptomatic and asymptomatic groups along with serum IgA studies in both children and adults would help explore other dimensions of the disease in Asian countries including India. Early recognition of CD will definitely have a direct impact on the burden of childhood morbidity and mortality.