The relationship between anti-endomysium antibodies and villous atrophy in coeliac disease using both monkey and human substrate.

BACKGROUND/OBJECTIVE: Circulating antibodies offer a noninvasive diagnostic screening test in patients with coeliac disease with severe histopathological abnormalities. This study assesses for the first time the sensitivity and reliability of anti-endomysium in screening for coeliac disease in patients with milder forms of villous atrophy using human umbilical cord and monkey ileum. MATERIALS AND METHODS: Serum from 124 adults and children > 2 years old including 33 patients with coeliac disease on a gluten-free diet and 91 patients referred to the laboratory for screening was studied. The presence of IgA-anti-gliadin (AGA) (ELISA) and IgA-anti-endomysium (EMA) was detected in the serum using monkey ileum and human umbilical cord (HUC) substrates. Patients with abnormal serology results or severe clinical complaints were invited to attend for a small-bowel biopsy. The prevalence of EMA detected on monkey ileum and HUC was compared with the histopathological features of coeliac disease at presentation. Fifty-three of the 91 patients screened for coeliac disease underwent a small intestinal biopsy. RESULTS: Twenty-three of the 91 patients suspected of having coeliac disease had coeliac disease. The EMA test was positive in 18 of 23 using both monkey ileum and HUC (sensitivity 78%). Partial villous atrophy (PVA) was seen in four of the five EMA-negative patients, and subtotal/total villous atrophy (SVA/TVA) was demonstrated in 18 of the 23 cases with positive EMA. Both substrates detected identical positive cases. There was an excellent concordance between EMA sensitivity evaluated on HUC and those on monkey ileum. One patient was EMA-negative on monkey ileum but positive on HUC and one patient who was EMA-positive on monkey ileum was EMA-negative on HUC. Only one of 33 coeliac disease patients on gluten-free diet for more than one year with persisting TVA had positive EMA. The rest of the cases had a negative EMA on both HUC and monkey ileum. CONCLUSION: A negative result for EMA in coeliac disease patients with a normal IgA value does not exclude the diagnosis of coeliac disease. A positive EMA is seen mostly in those coeliac disease patients with severe tissue damage (SVA/TVA). EMA has a low sensitivity in coeliac disease patients with PVA in spite of use of different substrates.