Objectives: The in vitro gluten challenge test is an important diagnostic modality in celiac disease (CD), especially in patients who begin treatment with a gluten-free diet before adequate diagnostic workup or in cases with atypical CD. Available literature was reviewed regarding the accuracy of the in vitro gluten challenge test for CD diagnosis.
Methods: MEDLINE, Scopus, and Google Scholar were searched, and studies that used serology and bowel biopsy as the criterion standard for diagnosis were included in our study. Data on authors, publication year, characteristics of the patient and control groups, patients' diet, duration of the gluten challenge test, histology findings, endomysial antibody (EMA) and anti-tissue transglutaminase (tTG) levels, CD markers, and intercellular cell adhesion molecule-1, and human leukocyte antigens before and after the gluten challenge test were extracted.
Results: Overall, 15 studies were included in this meta-analysis. Pooled sensitivity %/specificity % was 84/99 for EMA after the challenge, 52/96 for EMA without the challenge, 95.5/98.3 for anti-tTG after the challenge, and 95.1/98.3 for anti-tTG without the challenge test. Sensitivity/specificity for immunological markers were 89/97 for the percentage of CD25⁺-lamina propria lymphocytes, 96/91 for the percentage of CD3⁺-lamina propria lymphocytes, and 96.1/85.7 for the percentage of intercellular cell adhesion molecule-1-lamina propria lymphocytes. The factors that increased the sensitivity of EMA were longer test duration, and the evaluation of patients on a gluten-containing diet or short-term gluten-free diet.
Conclusions: The in vitro gluten challenge test can be a useful part of the diagnostic workup of CD, rather than only a model to evaluate its mechanisms.