Coeliac disease (CD) is defined as a small bowel enteropathy triggered by the ingestion of wheat gliadin in genetically susceptible individuals. The diagnosis of CD can be made at any age of life but especially during the early childhood, between 9 and 24 months, or in the third or fourth decade of life (1). The prevalence of CD between 2.5 and 15 years in the general population ranges from 3 to 13/1000 children, or approximately 1:300 to 1:80 children in Europe and the United States (2).
CD encompasses a large spectrum of clinical manifestations ranging from serious symptomatic forms to completely asymptomatic ones (2–6). Early childhood CD frequently has gastrointestinal symptoms such as diarrhoea with failure to thrive, abdominal pain, vomiting, constipation and abdominal distension starting between 6 and 24 months of age, and nongastrointestinal manifestations (1,4). Children may have a milder constellation of symptoms such as abdominal discomfort, indigestion, or only nongastrointestinal symptoms. In parallel, histological changes in the small intestinal mucosa can also range from typical severe lesions (villous atrophy, hyperplasia of the crypts) to mild forms characterized by minor degrees of enteropathy with a lymphocytic infiltrate and increased epithelial cell proliferation (7).
Along with the duodenal biopsy, serological and genetic tests have acquired greater importance in CD diagnosis in both symptomatic and nonsymptomatic forms. The finding of immunoglobulin (Ig) A antibodies to gliadin (AGA-IgA), endomysium (EMA-IgA), and tissue transglutaminase (tTG-IgA) and their disappearance on a gluten-free diet are considered as supporting evidence for the diagnosis of CD.
The utility to perform these 3 serological tests simultaneously was reconsidered when it was demonstrated that EMA-IgA and tTG-IgA had a higher diagnostic performance than AGA-IgA (2,8,9). EMA-IgA and tTG-IgA are greatly sensitive (88%–100%) and specific (95%–99%) tests to identify individuals experiencing CD (8,10,11). For this reason, the use of AGA-IgA was no longer recommended (2,12). However, there is still doubt about the interest of detection of AGA-IgA especially in children younger than 2 years with neither EMA-IgA nor tTG-IgA, as it has been shown that AGA-IgA are more accurate in these cases (13–16).
The aim of our study was to evaluate the real usefulness of AGA-IgA in a large children population, younger than 2 years of age, with a suspicion of CD to find out whether the determination of AGA-IgA should be abolished as it was suggested by others (2,12,17) or should still be used in daily practice.
PATIENTS AND METHODS
The present retrospective study analysed 4122 children (median age at diagnostic 15 months, range 3–24 months, with a sex ratio F:M equal to 1:3) experiencing clinical symptoms that could be associated with CD and had been investigated for AGA-IgA along with tTG-IgA and/or EMA-IgA. The patients came from the paediatric departments of several hospitals (Angers, Annonay, Bourg en Bresse, Dijon, Chalon-sur-saône, Saone, Chambery, Lons-le-Saunier, Lyon, Marseille, Montceau les Mines, Montluçon, Montpellier, Paris, and Strasbourg). Clinical data and histological findings of intestinal biopsies were obtained from 62 and 33 children displaying isolated positive AGA-IgA, respectively. The severity of mucosal atrophy was described according to the histological Marsh classification (18). Definitive diagnosis of CD was confirmed according to the consensus statement on CD of the National Institutes of Health (12).
Detection of Antibodies
The serological markers most commonly used in clinical routine were performed in 10 different laboratories (Angers, Dijon, Lyon, Marseille, Montpellier, Paris, and Strasbourg). Detection of EMA-IgA was performed by indirect immunofluorescence technique using tissue sections from monkey oesophagus as substrate (Biorad, Marnes la Coquette, France; The Binding site, Saint Egrève, France). tTG-IgA were detected by an enzyme-linked immunosorbent assay and fluorimetry using as antigen human recombinant transglutaminase, according to the manufacturer's instructions (Eurospital, Trieste, Italy; Phadia, Saint Quentin en Yvelines, France; FIDIS Biomedical, BMD, Marne la Vallée, France). AGA-IgA were identified by enzyme-linked immunosorbent assay using a native gliadin antigen (fraction of gliadin from wheat gluten) according to the manufacturer's instructions (BMD; Menarini, Rungis, France; Phadia).
Three hundred and twelve children of 4122 (8%) displayed at least EMA-IgA and/or tTG-IgA (Table 1). Eighty-five children of 3810 (2,2%) that were both EMA-IgA and tTG-IgA negative were shown to display AGA-IgA.
The clinical data were obtained for 62 of 85 children (Fig. 1). Thirty-three of the 62 children underwent a duodenal biopsy; 24 had small intestinal mucosal evaluation without revealing any degree of villous atrophy. Nine duodenal biopsies were classified as total (Marsh IIIc) (n = 1), subtotal (Marsh IIIb) (n = 5), and partial (Marsh IIIa) (n = 3) villous atrophy. Only 5 of those 9 patients were finally diagnosed as having a biopsy-proven CD, whereas the 4 other patients belong to the group of children with a diagnosis of non-CD; these children (57/62) experienced isolated diarrhoea or constipation (n = 38), growth deficiency (n = 10), cow's-milk protein intolerance (n = 2), iron-deficient anaemia (n = 1), Helicobacter pylori–associated gastritis (n = 1), other gastritis associated with recurrent abdominal pain (n = 2), cystic fibrosis (n = 1), schwannoma (n = 1), and myelodysplastic syndrome (n = 1).
Our results suggest that in a population of children younger than 2 years suspected of having CD, there is no need to simultaneously perform the detection of AGA-IgA in addition to EMA-IgA/tTG-IgA. It is well known that a majority of children diagnosed as having CD displayed tTG-IgA and EMA-IgA during their first year of life, but it has been shown that AGA-IgA can precede detectable tTG-IgA levels in serum (13). The presence of AGA-IgA before tTG-IgA is not well understood; AGA-IgA could reach the peripheral circulation, whereas tTG-IgA may stay mainly in the intestine (14,19,20).
Although the last guidelines for the diagnosis of CD, based on the current evidence, practical considerations including accuracy, reliability, and cost, indicated that measurement of AGA-IgA was no longer recommended (2), there are reports in which AGA-IgA seemed to be more sensitive than tTG-IgA and EMA-IgA with a prevalence of 97%, 83%, and 83%, respectively, in the population of children younger than 2 years (14). It was also demonstrated that tTG-IgA and EMA-IgA are less sensitive than AGA-IgA, with a positivity of 87%, 88%, and 89% of the cases, respectively (16); however, these studies only focused on children with biopsy-verified CD and not on a large population of children with a suspicion of CD. Therefore, it was important to analyse a large cohort of children who were investigated for CD but with no confirmed diagnosis at the time the serological tests were performed. Moreover, the present multicentric study was done with different serological tests avoiding the bias of a lack of sensitivity of a particular technique to detect AGA-IgA or difference in the clinical screening of patients.
In our study, only 2.1% (85/4122) of patients displayed isolated AGA-IgA with no tTG-IgA and EMA-IgA. Considering AGA-IgA as a second-line test when tTG-IgA and EMA-IgA were negative, the accuracy of the test was also poor, with 85 of 3811 children (2.2%) being positive. Clinical data were obtained for 62 of 85 children because 23 children were no longer studied by the paediatric departments, certainly because of weak symptoms and recovery. The CD diagnosis was probably excluded for those 23 patients. Thirty-three children with isolated AGA-IgA underwent a duodenal biopsy, and histologically proven CD was established for 5 patients. From a clinical point of view, 8.1% and 0.12% of the children, when considering the population of isolated AGA-IgA or the entire population, experienced CD, respectively. The vast majority of children displaying isolated AGA-IgA experienced non-CDs that were previously shown to be associated with these antibodies, except for schwannoma (21).
Our study is in accordance with previous studies showing that AGA-IgA should no longer be performed because of a lack of specificity or a lack of sensitivity when compared with tTG-IgA and EMA-IgA (11,15,22,23). Another study gave the same results, showing that all of the AGA-IgA–positive results in the presence of negative tTG/EMA were false-positives (17).
In addition, in our experience, isolated AGA-IgA does not really help the choice of stepping up to the biopsy level, this decision remaining clinically based, as recommended in the case of unusual seronegative CD by the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition (2). Although the sensitivity of this test has been described to be higher, we believe it should not be used in daily practice because of its poor specificity. The systematic test of AGA-IgA conferred no additional diagnostic benefit in comparison with tTG-IgA.
Because 5 patients were finally diagnosed as having CD, AGA-IgA could be proposed as a second-line test. We also must take into account that new generation of serological tests using antigen derived from gliadin or synthetic deamidated gliadin peptides are now available. The benefit of these new tests in patients with neither tTG-IgA nor EMA-IgA (5,16,24), as well as other tests such as HLA-DQ2/-DQ8 genotype typing, need to be evaluated in large populations.
In conclusion, although some children with CD younger than 2 years display isolated AGA-IgA, our data confirm in a large cohort of patients that the last guidelines for the diagnosis of CD should be followed with suppression of testing AGA-IgA in daily practice.
The authors thank all of the paediatricians and the anatomopathologists (hospitals: Angers, Dijon, Lyon, Marseille, Montpellier, Paris, and Strasbourg) who contributed to the study.
1. Fasano A, Catassi C. Current approaches to diagnosis and treatment of celiac disease: an evolving spectrum. Gastroenterology 2001; 120:636–651.
2. Hill ID, Dirks MH, Liptak GS, et al. Guideline for the diagnosis and treatment of celiac disease in children: recommendations of the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition. J Pediatr Gastroenterol Nutr 2005; 40:1–19.
3. Norris JM, Barriga K, Hoffenberg EJ, et al. Risk of celiac disease autoimmunity and timing of gluten introduction in the diet of infants at increased risk of disease. JAMA 2005; 293:2343–2351.
4. Fasano A, Catassi C. Coeliac disease in children. Best Pract Res Clin Gastroenterol 2005; 19:467–478.
5. van Heel DA, West J. Recent advances in coeliac disease. Gut 2006; 55:1037–1046.
6. Ravikumara M, Tuthill DP, Jenkins HR. The changing clinical presentation of coeliac disease. Arch Dis Child 2006; 91:969–971.
7. Ciclitira PJ, Johnson MW, Dewar DH, et al. The pathogenesis of coeliac disease. Mol Aspects Med 2005; 26:421–458.
8. Stern M. Comparative evaluation of serologic tests for celiac disease: a European initiative toward standardization. J Pediatr Gastroenterol Nutr 2000; 31:513–519.
9. Rostom A, Dube C, Cranney A, et al. The diagnostic accuracy of serologic tests for celiac disease: a systematic review. Gastroenterology 2005; 128:S38–S46.
10. Sulkanen S, Halttunen T, Laurila K, et al. Tissue transglutaminase autoantibody enzyme-linked immunosorbent assay in detecting celiac disease. Gastroenterology 1998; 115:1322–1328.
11. Horvath K, Hill ID. Anti-tissue transglutaminase antibody as the first line screening for celiac disease: good-bye antigliadin tests? Am J Gastroenterol 2002; 97:2702–2704.
13. Ghedira I, Sghiri R, Ayadi A, et al. Anti-endomysium, anti-reticulin and anti-gliadin antibodies, value in the diagnosis of celiac disease in the child. Pathol Biol (Paris) 2001; 49:47–52.
14. Lagerqvist C, Dahlbom I, Hansson T, et al. Antigliadin immunoglobulin A best in finding celiac disease in children younger than 18 months of age. J Pediatr Gastroenterol Nutr 2008; 47:428–435.
15. Burgin-Wolff A, Gaze H, Hadziselimovic F, et al. Antigliadin and antiendomysium antibody determination for coeliac disease. Arch Dis Child 1991; 66:941–947.
16. Maglio M, Tosco A, Paparo F, et al. Serum and intestinal celiac disease-associated antibodies in children with celiac disease younger than 2 years of age. J Pediatr Gastroenterol Nutr 2010; 50:43–48.
17. Holding S, Abuzakouk M, Dore PC. Antigliadin antibody testing for coeliac disease in children under 3 years of age is unhelpful. J Clin Pathol 2009; 62:766–767.
18. Revised criteria for diagnosis of coeliac disease. Report of Working Group of European Society of Paediatric Gastroenterology and Nutrition. Arch Dis Child 1990;65:909–11.
19. Hansson T, Dannaeus A, Kraaz W, et al. Production of antibodies to gliadin by peripheral blood lymphocytes in children with celiac disease: the use of an enzyme-linked immunospot technique for screening and follow-up. Pediatr Res 1997; 41:554–559.
20. Marzari R, Sblattero D, Florian F, et al. Molecular dissection of the tissue transglutaminase autoantibody response in celiac disease. J Immunol 2001; 166:4170–4176.
21. Bonamico M, Ballati G, Mariani P, et al. Screening for coeliac disease: the meaning of low titers of anti-gliadin antibodies (AGA) in non-coeliac children. Eur J Epidemiol 1997; 13:55–59.
22. Baudon JJ, Johanet C, Absalon YB, et al. Diagnosing celiac disease: a comparison of human tissue transglutaminase antibodies with antigliadin and antiendomysium antibodies. Arch Pediatr Adolesc Med 2004; 158:584–588.
23. Simell S, Hoppu S, Hekkala A, et al. Fate of five celiac disease-associated antibodies during normal diet in genetically at-risk children observed from birth in a natural history study. Am J Gastroenterol 2007; 102:2026–2035.
24. Catassi C, Fasano A. Celiac disease diagnosis: simple rules are better than complicated algorithms. Am J Med 2010; 123:691–693.