Year: 2023
Type: editorial
Abstract: Both type 1 diabetes and coeliac disease are immune-mediated diseases. There are a series of similarities and dissimilarities, between these two diseases. The similarities include a strong association with human leukocyte antigen (HLA) class II genotypes, children accounting for a considerable proportion of affected individuals and the early appearance of disease-associated autoantibodies. When we consider the dissimilarities, a conspicuous feature is that the driving antigen is known in coeliac disease but not in type 1 diabetes. Dietary gluten was identified as an antigen in coeliac disease almost 70 years ago. Dutch paediatrician Willem Dicke observed that the health of some children improved considerably during World War II in the Netherlands, due to the severe bread shortage. However, when the allied planes began dropping bread to the country, the condition of the same children quickly got worse.1 Other dissimilar characteristics are related to the fact that females dominate patients with coeliac disease, in contrast to the higher incidence of type 1 diabetes in males. They are also partly related to different association with HLA class II haplotypes. The HLA DR3-DQ2 haplotype, particularly the homozygosity for this haplotype, has been strongly associated with coeliac disease, whereas the HLA DR4-DQ8 haplotype has been shown to confer the highest risk for type 1 diabetes in Northern Europe.2 It has been well established that patients with type 1 diabetes have an increased risk of coeliac disease and the frequency has been reported to vary from 1.6% to 16% in various countries.3 A Swedish study showed that children and adolescents with coeliac disease had a 2.4-fold risk of presenting with type 1 diabetes before 20 years of age.4 A peculiar feature is that coeliac disease is more commonly diagnosed after the presentation of type 1 diabetes, which indicates that genetic elements are not the only factors to contribute to this comorbidity. It has been suggested that the intake of cereals increases after a diagnosis of type 1 diabetes, which may increase the manifestation of coeliac disease. Another contributing factors is likely to be that children with type 1 diabetes are regularly screened for coeliac disease in most countries. Sweden witnessed an epidemic of coeliac disease in the late 1980s, when the incidence increased four-fold among children under 2 years of age.5 The disease rate remained high until 1995. These trends were related to changes in the national infant feeding recommendations. In 1984, these recommendations suggested that the introduction of gluten should be postponed from 4 to 6 months of age. At the same time, the gluten content in commercially available infant formula increased. The feeding recommendations were modified again in 1996 and these were close to those recommendations before 1985. A gradual introduction of gluten, preferably during breastfeeding, was now recommended. This, combined with reduced gluten content in infant products, resulted in a rapid decrease in the incidence rate of coeliac disease. A Swedish study, published in 2022, had three key findings. First, that there have indeed been waves of coeliac disease, but probably more than one. Second, that the increase in coeliac disease is driven, to a large extent, by increased screening and surveillance. Third, that the incidence of coeliac disease has stabilised and is not increasing anymore.6 In the current issue of Acta Paediatrica, Bybrant et al. present the results of a study that aimed to explore whether children with type 1 diabetes experienced an increased risk of coeliac disease during the coeliac epidemic in Sweden.7 This lasted from 1986 to 1996. The analysis was based on two birth cohorts, one covering children born in 1992–1993 and the other comprising those born in 1997–1998. Information on the diagnosis of type 1 diabetes was retrieved from four different patient registers, while the information on the diagnosis of coeliac disease was collected from three other registries. The results showed that the prevalence of coeliac disease was similar during the epidemic period (10.7%) and the post-epidemic period (11.7%). In addition, the authors found that the children in the epidemic cohort were diagnosed with coeliac disease at an older age than those in the post-epidemic cohort, with a difference of 1.6 years in the mean age at diagnosis. This may have reflected more intensified screening for coeliac disease in the younger cohort. The sex distribution was quite even among the children diagnosed with coeliac disease in both diabetic cohorts. This finding was different from the general population, where there has been a definite excess of females among subjects with coeliac disease,8 as mentioned earlier. The authors emphasise that their study compared the national prevalence of coeliac disease among children and adolescents in two population-based cohorts: one representing the epidemic era and the other the post-epidemic period. They also discuss the limitations of their study. The authors list changes in screening procedures, increased awareness of coeliac disease and the quality of the registries as potentially confounding factors that may have affected the outcomes of their study. However, they conclude, that it is unlikely that such factors would have had any impact on the results, since the maximum time gap between the oldest and youngest children in the study was only 6 years. The data in the Bybrant et al study indicate that the epidemiological pattern of coeliac disease was different in children with type 1 diabetes and the general population, as no signs of the coeliac epidemic could be seen among cases affected by type 1 diabetes. They speculate that this may have been due to a stronger genetic susceptibility in subjects who presented with both diseases.7 That is hardly due to the HLA genotypes, as the HLA DR4-DQ8 haplotype, which confers the highest risk for type 1 diabetes, is less common among children with both diseases than in those with just type 1 diabetes.9 Some non-HLA genes that predispose individuals to type 1 diabetes have also been associated with an increased risk of coeliac disease. Parkkola et al found that a series of non-HLA genes conferred increased risks of other autoimmune diseases among children with type 1 diabetes. These include single nucleotide polymorphisms in the genes encoding RGS1, FUT2 and CD69.8 Such genetic differences may be related to the epidemiological features that characterise the group of patients with both coeliac disease and type 1 diabetes. The paper by Bybrant et al in this issue of Acta Paediatrica adds useful data to the debate about the similarities and dissimilarities between type 1 diabetes and coeliac disease. None.
Source: Acta Paediatrica
DOI: 10.1111/apa.16940
Author: Mikael Knip
Institutions: University of Helsinki, Tampere University Hospital
Language: English
Subfield: Gastroenterology
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
Funder: Academy of Finland
OpenAlex ID: W4385798686