New Data on the Reliability of the No-Biopsy Diagnosis in Pediatric Celiac Disease

Posted on by

Chang D, Wong M, Camila Cardenas M, et al. Pediatrics. 2025;156(3):e2025070897. Positive Predictive Value of Tissue Transglutaminase IgA for Celiac Disease.

Background:For a no-biopsy celiac diagnosis, “the [ESPGHAN] criteria were revised in 2020, to specify that only a highly positive tTG IgA greater than or equal to 10× ULN and a positive EMA on a second blood sample are sufficient to diagnose celiac disease, obviating the need for HLA testing or symptoms.” NASPGHAN “has yet to adopt similar serologic criteria for diagnosing celiac disease. A clinical report in 2016 continued to recommend a confirmatory biopsy in all suspected cases of celiac disease. This recommendation cited concerns of interassay variability.”

Methods: This was a retrospective, multicenter North American cohort study of children (<18 years) with an elevated tTG IgA within 6 months of an esophagogastroduodenoscopy between January 2016 and December 2021. Biopsy-confirmed celiac disease was determined by the presence of intraepithelial lymphocytosis and villous atrophy. The primary outcomes were the PPV of an elevated tTG IgA and tTG IgA greater than or equal to 10 times the upper limit of normal (10× ULN).

The study identified 4019 children with the following characteristics: 63.3% female; 9% type 1 diabetes, and 2% Down syndrome

Key findings:

  • Histologic findings in the entire cohort were consistent with celiac disease for 3321 children (PPV = 82.6%)
  • Among the 1739/4019 (43.2%) children with tTG IgA greater than or equal
    to 10× ULN, 1651 had biopsy-confirmed celiac disease (PPV10× = 94.9%). Five percent (88/1739) of children did not have histologic findings of celiac disease, including 41/1739 (2%) with normal histology
  • EMA only marginally improved specificity as 76% of children without celiac disease and tTG IgA greater than or equal to 10× ULN had a positive EMA, albeit on the same sample. There was variations across providers, practices, and countries, which resulted in EMA not being performed in all cases or tested on the same blood sample as the tTG IgA
  • Assays performed worse in children with type 1 diabetes (PPV10× 89%) than that in those without T1DM (95.7%)
  • Of those with esophageal biopsies, 175/2980 (6%) had at least 15 eos/hpf in at least 1 esophageal biopsy
  • Of those with gastric biopsies, 912/3534 (25.8%) had histologic gastritis, including 1.4% (51/3534) with evidence of Helicobacter infection (n (49 cases of H pylori and 2 cases of H heilmannii)

My take: This study indicates that the no-biopsy diagnostic approach for celiac disease may need to be reconsidered. Though, it is likely that the no-biopsy approach would have had a higher PPV if the guidelines were actually followed (eg. separate EMA confirmatory sample). The values in this study show a much lower PPV than prior studies (96% in those without T1DM). In addition, among the 4% without celiac disease (potential celiac disease), about half of them would likely progress to celiac disease and could potentially benefit from a gluten free diet. Thus, up to 2%of patients, based on this study, would probably be harmed by being placed on a gluten free diet.

Another point to reiterate based on this study, diagnostic confirmation by a specialist prior to dietary changes is recommended. Increasingly, patients are referred after starting a gluten free diet.

Related blog posts:

Brooklyn Botanic Gardens