The diagnosis of coeliac disease is currently based on the demonstration of small-bowel mucosal villous atrophy and crypt hyperplasia. However, the mucosal damage develops gradually and the patients may suffer from clinical symptoms even before the villous atrophy develops. In addition, there are several extraintestinal forms of coeliac disease in where the intestinal damage can be milder or even absent. These observations suggest that the current diagnostic criteria are insufficient to cover the wide range of gluten-dependent disorders. The early mucosal changes are too unspecific for the diagnosis as such, but there is evidence that accurate coeliac autoantibodies may also appear prior the villous atrophy develops. The primary aim of the present study was to evaluate whether adults and children having positive endomysial antibodies (EmA) would suffer from coeliac-type disorder and benefit of the dietary treatment while still having normal small-bowelmucosal morphology.
The dissertation comprised altogether four separate clinical studies. Study I was a randomized clinical trial performed in 70 EmA positive adults and 34 EmA negative controls. At baseline, the small-bowel mucosal structure and inflammatory markers, serum coeliac antibodies, clinical symptoms, celiac-disease associated genotype and laboratory parameters were evaluated from all. A total of 23 EmA positive subjects were found to have normal mucosal morphology and thus did not fulfil the current diagnostic criteria of coeliac disease. They were randomized either to continue with normal diet or to start a gluten-free diet. The remaining 47 EmA positive subjects had villous atrophy and crypt hyperplasia and started the gluten-free dietary treatment. Finally, after one year on trial all the EmA positive participants were re-evaluated. Study II comprised 27 EmA positive adults with normal mucosal morphology and 46 with villous atrophy. Alongside with the serological and histological evaluations, self-rated gastrointestinal symptoms, health-related quality of life, bone mineral density (BMD) and body mass index were measured both at baseline and after one year on a gluten-free diet. Altogether 110 healthy adults served as non-coeliac controls. Study III comprised 17 EmA positive children having normal villi, 42 children with villous atrophy and 17 EmA negative controls. After baseline eight EmA positive children having normal villi continued with a normal diet and five were placed on a gluten-free diet by their parent's decision. Again, all EmA positive children were re-investigated after one year of trial. Finally, Study IV comprised three cases who had coeliac disease with classical gastrointestinal symptoms diagnosed during childhood, but who at some point re-introduced gluten in their diet and after very long asymptomatic period were remitted to hospital because of a suspicion of dermatitis herpetiformis.
The results of the prospective studies (I-III) were analogous and showed that both EmA positive adults and children can suffer from the clinical symptoms and even decreased BMD despite normal small-bowel mucosal structure. In addition, in the EmA positive subjects who continued on a gluten-containing diet the mucosal changes exacerbated, coeliac antibody levels increased and clinical symptoms remained, whereas in those who started the treatment the antibodies decreased and clinical symptoms alleviated. The gluten-dependency of these EmA positive subjects was further supported by the fact that they all had the HLA genotype required for coeliac disease. In Study II it was shown that gluten-free diet may alleviate depression and is not detrimental for the quality of life of the EmA positive subjects having normal villi. The treatment may also improve the decreased BMD and seems to be harmless to the weight control of these EmA positive subjects. Finally, the skin biopsy of the three cases in Study IV confirmed the diagnosis of dermatitis herpetiformis and showed that the phenotype of coeliac disease may change after a long asymptomatic period. In addition, although they had no abdominal symptoms and only partial villous atrophy, all three cases had coeliac antibodies in the serum and coeliac-type autoantibody deposition in the intestinal mucosa.
The results of the present study demonstrated that EmA positive subjects suffer from similar genetically determinated gluten-dependent disorder as coeliac disease patients and benefit of the dietary treatment despite normal small-bowel mucosal morphology. Furthermore, the clinical and histological presentation of coeliac disease may change over time, showing that the intestinal and extraintestinal forms of the disorder belong to the same category of genetic gluten intolerance. These observations indicate that the current diagnostic criteria of coeliac disease are inadequate and should be revisited. In the future more studies are needed to assess whether EmA positive but asymptomatic patients having normal mucosal morphology should be treated and further, to asses the role of the endoscopic studies in the diagnosis of coeliac disease.
