Statistics from Altmetric.com
Editor,—We read with interest the Science Alert comment by Mäki (Gut 1997;41:565–6) on Dieterich et al’s paper1 identifying tissue glutamine (tTG) as the antigen for endomysial antibody (EMA). Unfortunately, Dr Mäki’s comments were somewhat speculative and severely biased towards his own view that gliadin somehow (but how?!) reveals neo-epitopes which, by inducing antibodies to connective tissue, apparently provide the key to the central pathogenic mechanism for gluten sensitivity. It is hardly useful to read that “… coeliac disease is indeed self-perpetuating and irreversible if the environmental trigger, gliadin, is not removed …”: that information has been around since Dicke’s era.
That there have been exciting findings from Sollid and colleagues from Oslo regarding the in vitro response of cloned (CD4+) mucosal T lymphocytes to gliadin and its derivative peptides with the production of interferon γ and other Th1-type cytokines,2 seems to have escaped Dr Mäki’s pen.
Moreover, it seems certain that, over the next few years, the Oslo group is set to define the qualitative T lymphocyte responses underlying mucosal damage in gluten sensitivity, and the gliadin peptides which evoke such changes. It is important to stress that these experiments underpin the drift of clinical research over the years which again has led to the inevitable conclusion that gluten sensitivity depends on T lymphocyte responses and not on B (humoral) immunology.3 4 That gluten sensitivity with all its clinical and immunopathological findings can occur without demonstrable antibody5 should amply inform Dr Mäki (and others) that a theory of pathogenesis for gluten sensitivity, based solely on antibodies, will not do5; that idea has already been dismissed by others.6 7
More importantly, at present there is no discussion in the literature about EMA negative patients. It is important to avoid a self-fulfilling prophecy—that is, taking biopsy samples only from EMA positive individuals. A recent editorial (Lancet1991;337:590) notes the disparity between diagnosis and serology. In most studies, the sensitivity of serological markers has been evaluated in terms of severe (flat) mucosal lesions, or alternatively, a biopsy had only been performed when serological markers were positive.8-10
In contrast, we showed when using tTG that sensitivities and specificities for a subgroup of patients fulfilling the ESPGAN criteria with partial villous atrophy at presentation, initially tested by the Berlin group (Dieterich, Schuppan), gave disappointing values of 44% and 88% respectively.
Again, in two independent, prospectively studied groups of coeliac patients,11 12 the overall sensitivity and specificity of EMA was 50%, and 90–95% respectively. Clearly, EMA is not exclusively positive in every gluten sensitised individual. However, when EMA positivity is related to the severity of the proximal mucosal biopsy, then sensitivity for EMA is about 90% for total villous atrophy, but only 30% for the milder infiltrative-hyperplastic lesions with partial villous atrophy.13 Thus whether the EMA test is positive or not depends entirely on the presence of a severe lesion and possibly on the length of intestine involved. This point needs to be remembered in population studies, especially when a flat, severe lesion is taken as sole manifestation of coeliac disease.
Much more needs to be learned about effective screening for gluten sensitised individuals. Endomysial antibodies alone fail to predict all such cases and clearly, therefore, do not constitute the universal panacea for this disease as Dr Mäki wants us to believe. Gluten sensitivity is not due exclusively to endomysial antibody production.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.