Statistics from Altmetric.com
Editor,—Recent studies have demonstrated that coeliac disease can develop months or years after the clinical onset of insulin dependent diabetes mellitus (IDDM).1 Over the last 10 years, 200 consecutive children with IDDM at the onset have been tested for IgG and IgA antigliadin antibodies (AGA) by indirect immunofluorescence2 and for IgA antiendomysial antibodies (EmA) using monkey oesophagus and, more recently, human umbilical cord as substrate.3 An antibody follow up was also performed in 151 of these patients every 3–6 months. At initial testing six diabetic children (four girls and two boys, median age 9 years, range 2–13) were positive for both IgA EmA and IgA AGA (associated with IgG AGA in five cases). A flat duodenal mucosa consistent with a diagnosis of coeliac disease was found in five (2.5%), whereas the remaining child (a 9 year old girl), who was positive for both IgA EmA and AGA at high titre, showed a normal small intestinal mucosa. One year later, antibody tests were again positive and a second biopsy specimen revealed the appearance of a subtotal villous atrophy (table 1, case 1). During follow up a further four children (three boys and one girl, median age at IDDM onset 3.3 years; range 2.9–3.5) of those initially antibody negative became positive. Two of these patients with antibody appearance within 10 and 16 months respectively showed a flat mucosa (table 1, cases 2 and 3). The late development of small intestinal atrophy in these three IDDM cases added 2% to the prevalence of the disease, which resulted as a whole 4.5%. The other two patients who became antibody positive (one only for IgA EmA at low titre) 18 months and 4.5 years after IDDM onset showed a histological picture of non-specific duodenitis (table 1, cases 4 and 5). It is possible to speculate that flat mucosa will never develop in them (especially in the patient with 4.5 years of follow up), nevertheless a condition of potential coeliac disease which needs to be confirmed by immunohistochemical studies is very likely. From a clinical viewpoint all those with coeliac disease (including latent and potential cases) did not show gastrointestinal symptoms except one, diagnosed 12 months after IDDM onset, who presented a mild malabsorption syndrome (table 1, case 1).
Of the four (2%) IDDM patients, positive at onset only for IgG AGA, associated in one case with IgA AGA, three were biopsied and small intestinal findings were completely normal. Moreover, both IgG and IgA AGA disappeared in the sera of all these patients within six months, supporting the hypothesis that their fleeting positivity as well as that of non-organ specific and organ specific autoantibodies is a facet of the immunological derangement observed at IDDM onset.4
Our results show that the whole prevalence of coeliac disease in IDDM after a 10 year follow up is 1:25 (1:20 including also potential cases) and the documented finding of late developing mucosal atrophy significantly concurs in determining this high rate. Therefore, antibody screening and biopsy of IDDM patients only at disease onset is not enough to exclude gluten sensitive enteropathy. Seroconversion of IgA antibodies after the manifestation of IDDM predicts coeliac disease, but it can occur also years before developing flat mucosa. In this view, IgA EmA should be preferred to AGA (both IgA and IgG) for their higher sensitivity and specificity.3 The increase of IgA EmA titre after their appearance may help in timing rebiopsy.
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.