Accuracy in diagnosis of celiac disease without biopsies in clinical practice

BACKGROUND & AIMS: The guidelines of the European Society of Pediatric Gastroenterology, Hepatology, and Nutrition allow for diagnosis of celiac disease without biopsies in children with symptoms and levels of immunoglobulin A against tissue-transglutaminase (TGA-IgA) 10-fold or more the upper limit of normal (ULN), confirmed by detection of endomysium antibodies (EMA) and positivity for HLA-DQ2/DQ8. We performed a large, international prospective study to validate this approach. METHODS: We collected data from consecutive pediatric patients (18 years or younger) on a gluten-containing diet who tested positive for TGA-IgA from November 2011 through May 2014, seen at 33 pediatric gastroenterology units in 21 countries. Local centers recorded symptoms; measurements of total IgA, TGA, and EMA; and histopathology findings from duodenal biopsies. Children were considered to have malabsorption if they had chronic diarrhea, weight loss (or insufficient gain), growth failure, or anemia. We directly compared central findings from 16 antibody tests (8 for TGA-IgA, 1 for TGA-IgG, 6 for IgG against deamidated gliadin peptides, and 1 for EMA, from 5 different manufacturers), 2 HLA-DQ2/DQ8 tests from 2 manufacturers, and histopathology findings from the reference pathologist. Final diagnoses were based on local and central results. If all local and central results were concordant for celiac disease, cases were classified as proven celiac disease. Patients with only a low level of TGA-IgA (threefold or less the ULN) but no other results indicating celiac disease were classified as no celiac disease. Central histomorphometry analyses were performed on all other biopsies and cases were carefully reviewed in a blinded manner. Inconclusive cases were regarded as not having celiac disease for calculation of diagnostic accuracy. The primary aim was to determine whether the nonbiopsy approach identifies children with celiac disease with a positive predictive value (PPV) above 99% in clinical practice. Secondary aims included comparing performance of different serological tests and to determine whether the suggested criteria can be simplified. RESULTS: Of 803 children recruited for the study, 96 were excluded due to incomplete data, low level of IgA, or poor-quality biopsies. In the remaining 707 children (65.1% girls; median age, 6.2 years), 645 were diagnosed with celiac disease, 46 were found not to have celiac disease, and 16 had inconclusive results. Findings from local laboratories of TGA-IgA 10-fold or more the ULN, a positive result from the test for EMA, and any symptom identified children with celiac disease (n 399) with a PPV of 99.75 (95% confidence interval [ CI], 98.61-99.99); the PPV was 100.00 (95% CI, 98.68-100.00) when only malabsorption symptoms were used instead of any symptom (n 278). Inclusion of HLA analyses did not increase accuracy. Findings from central laboratories differed greatly for patients with lower levels of antibodies, but when levels of TGA-IgA were 10-fold or more the ULN, PPVs ranged from 99.63 (95% CI, 98.67-99.96) to 100.00 (95% CI, 99.23-100.00). CONCLUSIONS: Children can be accurately diagnosed with celiac disease without biopsy analysis. Diagnosis based on level of TGA-IgA 10-fold or more the ULN, a positive result from the EMA tests in a second blood sample, and the presence of at least 1 symptom could avoid risks and costs of endoscopy for more than half the children with celiac disease worldwide. HLA analysis is not required for accurate diagnosis