Reliance on serum endomysial antibody testing underestimates the true prevalence of coeliac disease by one fifth

BACKGROUND: Although IgA endomysial antibody (EmA) is currently the serologic test of choice in selecting suspected coeliac patients for duodenal biopsies, false-negative cases have been reported and may be more common than previous studies suggest. We assessed the sensitivity of EmA for patients with biopsy-confirmed villous atrophy (VA). METHODS: We studied 89 patients without IgA deficiency for whom biopsy had not been primarily prompted by a positive EmA result. VA was graded as partial, subtotal, or total (PVA, STVA, TVA). Serum EmA was assayed with indirect immunofluorescence. RESULTS: The sensitivity of EmA for VA was 78% (69 of 89) and was similar for PVA (79%) and ST/TVA (77%). Only 4 of the 20 EmA-negative patients had increased serum IgA-class antigliadin antibody levels as measured with enzyme-linked immunosorbent assay. All seronegative patients who complied with dietary gluten exclusion responded clinically, with histologic improvement after 12 months in 8 (67%) of 12 patients who had follow-up biopsies. CONCLUSIONS: EmA-negative coeliac disease is common. Reliance on EmA testing to select patients for biopsy will result in significant underdiagnosis.