Coeliac disease (CD) develops almost exclusively in individuals with the HLA-DQ2 or HLA-DQ8 haplotypes. However, these haplotypes occur in about 40% of the general population. As most children have exposure to gluten, this indicates there are additional genetic and environmental triggers involved in disease development. Previous research, based on parental reporting of illness, suggest a role for infections, particularly those of the gastrointestinal (GI) tract, in the development of CD. The only prospective study in this area suggested that frequent rotavirus infections might increase the risk of developing CD antibodies in a high-risk cohort. Other retrospective studies looking at adenovirus, enterovirus and orthoreovirus have shown inconclusive or conflicting results.
This study considered the question of potential GI triggers by using a longitudinal birth cohort analysis of the most frequently occurring viruses: enterovirus and adenovirus. The study aimed to test whether the presence of human enterovirus and adenovirus in monthly faecal samples was more common before the development of CD antibodies in cases subsequently diagnosed with CD compared with children who did not develop CD.
Between 2001-2007, 46,939 newborns in Norway were screened for the HLA-DQ2/DQ8 genotype conferring an increased risk of type 1 diabetes and CD. 912 (1.9%) of the children screened were found to have the risk genotype and were followed with repeated blood and faecal samples from the age of 3 months. Plasma samples were collected at 3, 6, 9 and 12 months and then annually thereafter. Monthly stool samples were collected between 3-36 months of age and stored. Those children actively contributing blood samples during 2014-2016 (n=501) were invited for CD screening, of whom 220 agreed to participate. The European Society for Paediatric Gastroenterology Hepatology & Nutrition (ESPGHAN) 2012 diagnostic criteria were used for CD diagnosis. 27 cases of CD were identified and following evaluation for eligibility, 25 cases were matched to two controls each.
The time interval when cases seroconverted for CD markers was then determined by retrospectively analysing biobanked samples that had been collected longitudinally. Of the 25 case-control groups, 15 had this seroconversion period covered by monthly stool sampling, whereas the remaining 10 seroconverted after the collection of the stools had finished.
The study found a significant association between exposure to enterovirus and subsequent risk of CD. Adenovirus was not found to be associated with CD. Although the effect sizes are relatively small, this study suggests that enterovirus infections in early life could be one of several key risk factors for the development of CD. The observations by this group suggest that several types of enterovirus, high titre and long duration of infections in the period after gluten introduction were involved. In addition, these results work with the theory that viral infections may disrupt the mucosal barrier resulting in increased translocation of gluten peptides into the mucosa as an initial step in loss of tolerance.
It may be that CD patients have enteric barrier disruption prior to the development of CD autoantibodies and as a result a susceptibility to enterovirus. However, the authors believe a more plausible explanation to be that enterovirus causes impaired barrier function which in turn increases the risk of CD. A challenge in examining the temporal association is the difference in time between the triggering event and disease onset as measured serologically in this study. The study design is also less sensitive to detection of infections occurring with lower frequency and shorter duration of viral shedding so a non-specific response to several viral infections needs further investigation.
Further corroboration of results from this study is needed, in larger sample sets, to reach conclusions about causality, however, if enterovirus is confirmed as a trigger factor, the use of vaccination could reduce the risk of CD development.
Kahrs CR, Chuda K, Tapia G et al. BMJ 2019;364:1231. Doi :10.1136/bmj.1231
Link to original article: https://www.bmj.com/content/364/bmj.l231