Abstract
Objectives IgA antibodies against tissue transglutaminase (anti-TG2) is a reliable marker of celiac disease (CD). However, IgA-deficient patients are not identified and young children may lack IgA anti-TG2. Combined detection of IgA and IgG (IgA/IgG) against deamidated gliadin peptides (DGP) has shown a high diagnostic performance for untreated CD. Here we examined the utility of IgA/IgG anti-TG2, IgA/IgG anti-DGP and IgA/IgG against a mix of TG2 and DGP (anti-TG2/DGP) in finding CD among children. Methods Serum antibodies against TG2, DGP, and TG2/DGP were determined with ELISA in 242 children referred to a paediatric gastroenterologist. Fifty had untreated CD verified by an intestinal biopsy and 192/242 children had other diseases than CD. Results Forty-eight untreated CD children had increased IgA/IgG anti-TG2, 47/50 had increased IgA/IgG anti-DGP and 46/50 had increased IgA/IgG anti-TG2/DGP. One control subject had increased IgA/IgG anti-TG2 and IgA/IgG anti-TG2/DGP, whereas 7/192 control subjects had increased IgA/IgG anti-DGP. The IgA/IgG anti-TG2 assay had the best performance with a sensitivity of 96%, a specificity of 99.5% and the area under the ROC-curve was 0.996 (95% CI 0.992–1, p < 0.0001). Conclusions Detection of one antibody is not sufficient when screening for untreated CD among children due to cases of IgA deficiency. The inclusion of DGP antigens in the IgA/IgG combination assays seems to affect the sensitivity and specificity negatively, whereas detection of IgA/IgG anti-TG2 has the potential of finding most untreated CD patients, including those with IgA deficiency.
Acknowledgements
The authors thank the late associate professor Anders Dannaeus for his excellent contribution to the present work and for his life-long commitment to the celiac disease research and the clinical care of children with gastrointestinal diseases. The authors thank Helene Lettesjö and Elin Heed for excellent technical support.
Disclosure statement
All the authors fulfil the authorship criteria and the study was designed as a research project between the involved authors, each contributing with expertise necessary for performing the study. No economical support from an entity with financial interests in the subject matter has occurred in the present study. The final manuscript was seen and approved by all authors and they took due care to ensure the integrity of the work. This article or parts of it has never been published or is not currently under consideration elsewhere.
Funding information
The study was supported by the Gillbergska foundation, Uppsala, Sweden, and the Swedish Governmental Agency for Innovation Systems (VINNOVA Sambio 2007-00084).