Celiac disease (CD) is an autoimmune-mediated enteropathy triggered by the ingestion of wheat, rye, or barley gluten. The criterion standard for the diagnosis of CD is the histological evaluation of the duodenal mucosal tissue obtained by biopsy, and serologic tests are useful to identify individuals who require a duodenal biopsy. Classic serologic tests include anti-gliadin antibodies (AGAs), anti-endomysial antibodies (EMA), and anti-tissue transglutaminase antibodies (anti-tTG) of the immunoglobulin A (IgA) class. AGA testing has little diagnostic value because of its low sensitivity and specificity, especially in children aged 18 months or older (1). EMA and anti-tTG exhibit better sensitivity and specificity, close to 95% to 100%, and have replaced conventional AGA as the best serologic tools for screening CD (1–3). Because the dominant antigen of endomysial antibodies is transglutaminase, both assays measure the same antibodies (4). Therefore, determination of anti-tTG IgA alone may be used as a screening test because it is accurate and not operator dependent (5). In addition, serologic tests may be used to monitor compliance with gluten-free diet (GFD), but this is still a controversial issue, because anti-tTG IgA levels were reported to be scarcely reliable in detecting compliance when compared with duodenal biopsy (6,7); however, they were reported to be significantly lower in a group of adults with CD strictly adhering to the diet (8).
In the last decade, deeper understanding of the immunopathology of the small bowel mucosal damage in CD increased interest in deamidated gliadin peptides (a-DGPs). Dietary gluten is partially digested into gliadin peptides that pass through the small bowel epithelium and are bound to the human leukocyte antigen HLA-DQ2 or DQ8 on the surface of antigen-presenting cells. This binding is strongly enhanced by deamidation, catalyzed by anti-tTG, thus resulting in a greatly enhanced T-cell (9,10) and antibody response (11). Recent studies have shown that deamidation increases binding of antibodies to gliadin in the sera of patients with CD but not of controls (11–14). The reliability of serum antibodies against a-DGP in detecting adherence to GFD was studied in 4 small groups of adults with CD (15–18) and in only a few children with CD (19,20).
The aim of this study was to assess the usefulness of a-DGP antibodies for the evaluation of compliance with GFD in children with CD both in the first year after the diagnosis of CD and the introduction of GFD and after 1 or more years of GFD. To do so we preliminarily calculated the cutoff of both a-DGP IgA and a-DGP IgA+G (a combined detection of IgA and IgG isotypes of a-DGP) with a receiver operating characteristic (ROC) curve in a group of children with active CD and gastroenterological controls. We also compared a-DGPs diagnostic accuracy with that of anti-tTG IgA and classical AGA IgA.
PATIENTS AND METHODS
For the preliminary analysis of the cutoff of a-DGPs in our pediatric population consecutive serum samples were collected in children the same day of endoscopy performed for failure to thrive or abdominal pain or other gastrointestinal symptoms. Upper gastrointestinal endoscopy with duodenal biopsy confirmed active CD in 95 children (male/female [M/F] = 39/56, median age 6.6, range 0.8–16.8 years) and normal duodenal histology in 106 (M/F = 43/63, median age 7.2, range 0.4–17.1 years), which were considered gastrointestinal controls. Histologic duodenal damage was graded according to the Marsh classification (21). In controls histology showed esophagitis in 49, gastritis in 19, esophagogastritis in 18, and normal mucosa in 20 children.
We then analyzed serum samples of 134 children with CD on GFD: 28 of them (M/F = 11/17, median age 8.1, range 1–16.8 years) were newly diagnosed and studied for the first year of GFD with serum samples to be collected every 3 months, and 106 (M/F = 43/63, median age at endoscopy 6.4, range 0.8–16.2 years) were on GFD for 1 year or longer (range 1–14 years) and their median age at serum collection was 10.6 years (range 2.3–22.1). The 28 children with CD studied in the first year of GFD were part of the 95 with active CD.
Compliance with GFD was ascertained through a structured interview always performed by the same investigator (G.O.). Parents and children were instructed to follow specific dietary restrictions and received additional counseling to reinforce the instructions at each follow-up visit. The same operator always performed a physical examination and assessed growth parameters. Dietary compliance assessment at follow-up visits was based on self-reported dietary habits by the child and the parents, and was categorized as strict or partial compliance with GFD. When a discrepancy between parents and child reports was demonstrated by the structured interview, the final categorization (ie, partially or strictly compliant) was assigned on the basis of the investigator's opinion. The investigator was completely blind to the serologic test results because the interview was done on the same day of the serum samples collection.
Written informed consent for the anonymous use of their data was obtained from the parents of all of the children included in the study, and the study protocol was approved by local ethical committee.
Laboratory Assays for CD-specific Serology
After centrifugation at 3000 rpm for 10 minutes, sera were stored at −20°C until the enzyme-linked immunosorbent assays (ELISAs) were performed. AGA and anti-tTG—both of the IgA class—determinations were performed by EliA Gliadin IgA and EliA Celikey IgA assays, respectively (Phadia AB, Uppsala, Sweden), with a manufacturer's suggested cutoff of 10 U/mL. a-DGP IgA and a-DGP IgA+G (a combined detection of IgA and IgG isotypes of a-DGP) determinations were performed by Quanta Lite Gliadin IgA II and Quanta Lite Celiac DGP Screen, respectively (INOVA Diagnostics, San Diego, CA), with a manufacturer's suggested cutoff of 20 arbitrary units (AU). All of the tests were conducted in accordance with the manufacturer's instructions. Total IgA assay was conducted by an immunonephelometric assay (BN ProSpec; Siemens Healthcare Diagnostics, Marburg, Germany).
To determine in our population the cutoff value of a-DGP IgA and a-DGP IgA+G, we performed ROC curve analysis. Using ROC curves, we identified the cutoff value associated with the highest sensitivity and specificity by plotting for each cutoff value sensitivity versus (1 − specificity). The area under the ROC curve and the 95% confidence interval (95% CI) were assessed. We then compared the sensitivity and specificity of a-DGP IgA, a-DGP IgA+G, anti-tTG IgA, and AGA IgA to detect compliance with GFD in the first year of GFD and after 1 year or longer of GFD by 2-tailed Fisher exact test or χ2 test, when appropriate. For all of the serologic assays sensitivity, specificity, positive likelihood ratio (LR+), and negative likelihood ratio (LR−) were calculated.
Levels of CD-specific antibodies were expressed as median, interquartile range (IQR), and range values. Comparisons among groups were performed by the Mann-Whitney U test and the Kruskal-Wallis ANOVA test, as appropriate. Post-hoc comparisons were performed with Bonferroni correction. The k statistic was used to analyze agreement between a-DGPs and anti-tTG or AGA assays. The level of agreement was classified as k < 0.20: poor; 0.21 to 0.40: fair; 0.41 to 0.60: moderate; 0.61 to 0.80: good; and 0.81 to 1.00: excellent.
Statistical analysis was performed with STATISTICA software version 6.1 (Statsoft Inc, Tulsa, OK) and ROC curve analysis with MedCalc software version 11.0.1 (MedCalc Software, Mariakerke, Belgium).
Cutoff Determination of A-DGP IgA and IgA+G Antibodies by ROC Curves
The ROC curve analysis performed on the a-DGP IgA and a-DGP IgA+G levels measured in 95 children with active CD at diagnosis and 106 gastroenterological controls showed that the cutoff value for a-DGP IgA was 13.1 AU (ie, all children with an a-DGP IgA level 13.1 AU were considered negative). The cutoff value for a-DGP IgA+G was 16.5 AU (ie, all children with an a-DGP IgA+G level 16.5 AU were considered negative). Figure 1 shows the ROC curves for a-DGP IgA and a-DGP IgA+G. Area under the curve was 0.948 (95% CI 0.908–0.975, P = 0.0001) for a-DGP IgA and 0.972 (95% CI 0.939–0.990, P = 0.0001) for a-DGP IgA+G.
According to these cutoffs, a-DGP IgA were positive in 83 of 95 children with CD and in 3 of 106 controls; their median value was 95 AU (IQR 24.3–127.8 AU) in children with CD and 3.8 AU (IQR 2.6–5.7 AU) in controls. a-DGP IgA+G were positive in 90 of 95 children with CD and in 11of 106 controls; their median value was 98.3 AU (IQR 46.1–114.5 AU) in children with CD and 6.6 AU (IQR 4.3–9.3 AU) in controls. Table 1 shows sensitivity, specificity, positive, and negative likelihood ratios for a-DGP IgA, a-DGP IgA+G, anti-tTG IgA, and AGA IgA. Anti-tTG IgA were positive in 94 of 95 children with CD and in 1 of 106 controls. AGA IgA were positive in 68 of 95 children with CD and in 6 of 106 controls.
When comparing sensitivity of the 4 serologic assays, we found both a-DGP IgA and a-DGP IgA+G significantly more sensitive than AGA IgA (P = 0.01 and P < 0.0001, respectively) and no difference between a-DGP IgA+G and anti-tTG IgA, whereas we found a-DGP IgA to be significantly less sensitive than anti-tTG IgA (P = 0.002), but no difference was seen between a-DGP IgA and a-DGP IgA+G. When comparing specificity of the 4 assays, we found a-DGP IgA+G to be significantly less specific than anti-tTG IgA (P = 0.005) but found no difference between a-DGP IgA and anti-tTG IgA. Neither a-DGP IgA nor a-DGP IgA+G specificity was different from that of AGA IgA. Specificity of a-DGP IgA was slightly higher than that of a-DGP IgA+G (P = 0.049). Agreement between a-DGPs and anti-tTG assays was excellent (k = 0.85 for a-DGP IgA and k = 0.84 for a-DGP IgA+G) and between a-DGPs and AGA assays was good (k = 0.73 for a-DGP IgA and k = 0.63 for a-DGP IgA+G). In our series we did not detect children with CD or controls with IgA deficiency.
We further analyzed the diagnostic performance of all of the serologic assays according to age. For this purpose, we divided children with CD in 2 subgroups: ≤3-year-olds (n = 24) and >3-year-olds (n = 71). Serum a-DGP IgA were positive in all 24 ≤3-year-old children with CD and in 59 of 71 (83.1%, 95% CI 72–91) >3-year-old children with CD, showing a significantly higher sensitivity in younger children (P = 0.03). a-DGP IgA+G were positive in all of the ≤3-year-old children with CD and in 66 of 71 (93%, 95% CI 84–98) >3-year-old children with CD, with a similar sensitivity in both younger and older children. AGA IgA were positive in all 24 ≤3-year-old children with active CD but only in 44 of 71 (61.9%, 95% CI 49–73) >3-year-old children with CD, showing a significantly higher sensitivity in younger children (P = 0.0001). Anti-tTG IgA were positive in all 24 ≤3-year-old children with active CD and in 70 of 71 (98.6%, 95% CI 92–100) >3-year-old children with CD. No difference was found as regards the specificity in any of the 4 serologic assays in relation to age of the control children.
Serum Levels in the First Year After CD Diagnosis and Introduction of GFD
Of the 95 newly diagnosed children with CD, 28 were studied in the first year after introduction of GFD with a structured interview, physical examination, evaluation of growth parameters, and serology that were to be performed every 3 months. Not all of the children strictly followed the instructions, and timekeeping was not always correct. At 2 to 4 months, data and serum samples were available for 21 children (13 strictly compliant, 8 partially compliant), at 6 to 8 months for 13 children (8 strictly, 5 partially compliant), and at 9 to 12 months for 20 children (11 strictly, 9 partially compliant). Table 2 shows changes in prevalence of positive/negative samples over time. At 2 to 4 months of GFD, the prevalence of positive anti-tTG IgA was significantly higher in partially compliant children, but the difference was no longer significant in later samples. At 6 to 8 months, the prevalence of positive a-DGP IgA and a-DGP IgA+G was significantly higher in partially compliant children, but at 9 to 12 months only the prevalence of positive a-DGP IgA+G was still significantly higher in partially versus strictly compliant children. In particular, none of the partially compliant children was negative for a-DGP IgA+G at 6 months or later (sensitivity 100%). Moreover, at 9 to 12 months, sensitivity to detect transgressions to GFD was significantly higher for a-DGP IgA+G than for a-DGP IgA (100% versus 44%, P = 0.03).
Figure 2 shows the trends of all of the serologic tests. At 2 to 4 and 6 to 8 months, median anti-tTG IgA levels decreased in strictly compliant children only, whereas at 9 to 12 months, the levels decreased also in partially compliant children and were not significantly different in strictly versus partially compliant children. Median levels of a-DGP IgA and a-DGP IgA+G were decreased in all of the children already at 2 to 4 months; however, the difference between strictly and partially compliant children became significant only after 6 months or longer (P = 0.008 for a-DGP IgA and P = 0.04 for a-DGP IgA+G) and 9 to 12 months (P = 0.03 for a-DGP IgA and P = 0.02 for a-DGP IgA+G).
Serology According to Compliance With Diet After ≥1 Year of GFD
According to interview, physical examination, and evaluation of growth parameters, strict compliance was found in 81 (76.4%, strictly compliant) of the 106 children with CD whose serum samples were collected ≥1 year (range 1–14 years) after the introduction of GFD, whereas transgressions were found in 25 (23.6%, partially compliant). None of the children was fully noncompliant. Of the 25 partially compliant children, a-DGP IgA was positive in 15 (60%) and negative in 10, a-DGP IgA+G was positive in 19 (76%) and negative in 6, anti-tTG IgA was positive in 6 (24%) and negative in 19, and AGA IgA was positive in 1 (4%) and negative in 24. Of the 81 strictly compliant children, a-DGP IgA was positive in 11 and negative in 70 (86.4%), a-DGP IgA+G was positive in 7 and negative in 74 (91.4%), anti-tTG IgA was positive in 3 and negative in 78 (96.3%), and AGA IgA was positive in 1 and negative in 80 (98.8%).
When comparing performance of the 4 serologic tests to detect partial compliance (Table 3), sensitivity of a-DGP IgA was significantly higher than that of both anti-tTG IgA (P < 0.0001) and AGA IgA (P < 0.0001) but not significantly different from that of a-DGP IgA+G. Similarly, sensitivity of a-DGP IgA+G was significantly higher than that of both anti-tTG IgA (P < 0.0001) and AGA IgA (P < 0.0001). Specificity of a-DGP IgA in detecting compliance with GFD was lower than that of both anti-tTG IgA (P = 0.047) and AGA IgA (P = 0.005) but similar to that of a-DGP IgA+G, whereas specificity of a-DGP IgA+G was not significantly different from that of anti-tTG IgA and AGA IgA. We also found the 11 strictly compliant children with positive a-DGP IgA (false-positive, median 13.9, range 13.3–17.3 AU) had significantly lower levels than the 25 partially compliant (true-positive, median 20.5, range 7.5–28.5 AU, P = 0.0003).
No other differences were found between strictly compliant and partially compliant children as regards age at diagnosis, sex, duration of GFD, or age at sample collection for serology determination.
In this study we ascertained whether the new serological tests measuring a-DGPs were sensitive enough to detect transgression to GFD in children with CD. To do this we had to preliminarily determine the cutoff of a-DGP IgA and a-DGP IgA+G in our population of children with active CD and gastroenterological controls by a ROC curve analysis.
The cutoff values we calculated (ie, 13.1 AU for a-DGP IgA and 16.5 AU for a-DGP IgA+G) were lower than the manufacturer's suggested 20 AU. This is in accordance with some previous studies that suggested lowering a-DGP cutoff. In a group of children, the best cutoff was 9.7 for IgA and 9.8 for IgG (22). In other studies including both children and adults, the best cutoffs were 14.8 for IgA and 9.6 for IgG (16) or 11.4 for IgA and 9.6 for IgA+G (18) or 17.1 for IgG (23). Other studies, however, confirmed the manufacturer's cutoff of 20 as the optimal one in children both for screening and for follow-up after GFD (20), even in those with total IgA deficiency (24). On the contrary, some authors suggested raising the cutoff to 30 for a-DGP IgA, IgG, and IgA+G (25) or even to 45 for a-DGP IgA+G when used for CD mass screening (26). According to our cutoff, both a-DGP IgA and a-DGP IgA+G showed good sensitivity as screening tests for CD, but sensitivity was still lower than that of anti-tTG IgA. Therefore, neither of them could replace anti-tTG to screen for CD before endoscopy, thus confirming conclusions of a recent meta-analysis, according to which anti-tTG perform better than a-DGP IgA as a serologic test for CD screening (27). Both a-DGP IgA and a-DGP IgA+G showed higher sensitivity and specificity than that of AGA IgA. Moreover, in children younger than 3 years of age, the sensitivity of a-DGP was 100% with behavior similar to that of AGA. Indeed, in younger children AGA testing is still recommended because of the high prevalence of children with false-negative anti-tTG reported in many series (1); thus, a-DGP could replace AGA completely even in very young children.
The main aim of our study was to ascertain whether a-DGPs were accurate in monitoring compliance with GFD. To assess their changes in time and evaluate when the difference between strictly and partially compliant became significant, we studied a group of 28 newly diagnosed children with CD during the first year of GFD. It was not possible to longitudinally follow every child, because any of them could be strictly or partially compliant at different time points, so prevalence of negative and positive samples—and not children—at each time point was assessed. Moreover, not all of the samples of all 28 children were available at each time point because not all of the children strictly followed visit timekeeping and some missed ≥1 appointments. Despite these limits, we showed that all of the serologic levels decreased over time in the first year after the introduction of GFD, even in partially compliant children, as already reported in 20 children with CD in whom compliance was not controlled for (25). The general decrease we have seen in all of the serologic tests was probably the result of all of the children being on GFD, and the partially compliant having only occasional dietary transgressions. The most interesting finding in our series was that the prevalence of positive a-DGP levels (both IgA and IgA+G) was significantly higher in partially compliant than in strictly compliant children with CD only at 6 to 8 months, and remained so at 9 to 12 months only for a-DGP IgA+G. In particular, none of the partially compliant children was negative for a-DGP IgA+G at 6 months or longer, showing 100% sensitivity. This finding confirms the improved performance of serology testing reported in adults on GFD for 12 months or longer (28), possibly because it takes time for subjects with CD to reach a thorough comprehension of the intricacy of GFD and a complete adherence to it.
Neither anti-tTG IgA nor AGA IgA seems to be reliable in monitoring compliance with GFD in children because of their low sensitivity. The nonreliability of anti-tTG is still controversial in adults. Their levels were reported to be negative even in subjects with persistent duodenal atrophy (6,7) but were reported to be significantly lower in adults with CD strictly compliant with GFD when compared with those reporting dietary transgressions (8). When comparing anti-tTG and a-DGP accuracy in detecting GFD compliance it was found to be similar in predicting the status of duodenal mucosa in 53 adults with CD after 1 year of GFD (17), whereas a-DGPs were reported to be more reliable than anti-tTG in a small group of 9 adults with CD after 1 year of GFD (15). In children treated with GFD both anti-tTG and a-DGP decreased significantly; however, because of more rapid resolution of a-DGP (19) and a higher prevalence of negative values in children compliant with GFD (20), this antibody measurement was suggested as more useful than anti-tTG in monitoring dietary compliance. Our children studied in the first year after the introduction of GFD a-DGP IgA and a-DGP IgA+G were found to be reliable for monitoring compliance with GFD, when the diet had been consumed for at least 6 months or longer, and this finding was confirmed in our series of 106 children with CD on GFD for ≥1 year, where both a-DGP IgA and a-DGP IgA+G showed higher sensitivity than anti-tTG IgA and AGA IgA. This is at variance from data reported in adults (29), in which the prevalence of positive levels of a-DGP was still high in strictly compliant subjects after 1 year of GFD and not significantly different from partially compliant ones. The more rapid decrease of a-DGP seen in our population of children should be attributed to the younger age at CD diagnosis and GFD introduction. Moreover, in strictly compliant children with positive a-DGP IgA (false-positive) median a-DGP IgA levels were significantly lower than they were in partially compliant children (true-positive) and levels were all below the manufacturer's cutoff of 20 AU (range 13.3–17.3), so they could be considered weakly positive. We did not have any fully noncompliant child, probably because they were clinically supervised and repeatedly given dietary counseling.
Specificity of all of the tests was satisfactory, but we believe that the most clinically relevant characteristic of such tests is sensitivity, because it is better to strengthen dietary instruction in a strictly compliant child than to miss a partially compliant one. Therefore, high sensitivity is essential in detecting dietary transgressions that were reported to enhance the risk of long-term complications (30) and worsen quality of life (31–33).
A limit of our study could be that we did not calculate the optimal cutoff for anti-tTG IgA and AGA IgA because this was not the objective of our study. Manufacturers' cutoffs for AGA and anti-tTG are routinely used as such in most laboratories; however, this could be a source of bias.
In conclusion, both a-DGP IgA and a-DGP IgA+G showed good sensitivity in monitoring compliance with GFD, higher than that of anti-tTG IgA and of AGA IgA, at comparable high specificity, but a-DGP IgA+G seemed to perform better than a-DGP IgA in children on GFD for 9 to 12 months or >1 year. Both a-DGP IgA and a-DGP IgA+G showed good sensitivity and specificity as screening tests for CD, but specificity was higher for a-DGP IgA; however, they did not outperform anti-tTG IgA, which remains the best screening test for CD. In very young children sensitivity of both a-DGPs was high and therefore traditional AGA IgA could be substituted in this age group for their comparable performance.
1. Green PH, Cellier C. Celiac disease. N Engl J Med 2007; 357:1731–1743.
2. Hill ID. What are the sensitivity and specificity of serologic tests for celiac disease? Gastroenterology 2005; 128:S25–S32.
3. 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.
4. Dieterich W, Ehnis T, Bauer M, et al. Identification of tissue transglutaminase as the autoantigen of celiac disease. Nat Med 1997; 3:797–801.
5. Collin P, Kaukinen K, Vogelsang H, et al. Antiendomysial and antihuman recombinant tissue transglutaminase antibodies in the diagnosis of celiac disease: a biopsy-proven European multicentre study. Eur J Gastroenterol Hepatol 2005; 17:85–91.
6. Vahedi K, Mascart F, Mary JY, et al. Reliability of antitransglutaminase antibodies as predictors of gluten-free diet compliance in adult celiac disease. Am J Gastroenterol 2003; 98:1079–1087.
7. Kaukinen K, Sulkanen S, Maki M, et al. IgA class transglutaminase antibodies in evaluating the efficacy of gluten-free diet in celiac disease. Eur J Gastroenterol Hepatol 2002; 14:311–315.
8. Fabiani E, Catassi C, International Working Group. The serum IgA class anti-tissue transglutaminase antibodies in the diagnosis and follow up of coeliac disease. Results of an international multi-centre study. International Working Group on Eu-anti-tTG. Eur J Gastroenterol Hepatol 2001;13:659–65.
9. Molberg O, McAdam SN, Korner R. Tissue transglutaminase selectively modifies gliadin peptides that are recognized by gut-derived T cells in celiac disease. Nat Med 1998; 14:713–717.
10. van de Wal Y, Kooy Y, Van Veelen P, et al. Selective deamidation by tissue transglutaminase strongly enhances gliadin-specific T cell reactivity. J Immunol 1998; 161:1585–1588.
11. Aleanzi M, Demonte AM, Esper C, et al. Celiac disease: antibody recognition against native and selectively deamidated gliadin peptides. Clin Chem 2001; 47:2023–2028.
12. Osman AA, Gunnel T, Dietl A, et al. B cell epitopes of gliadin. Clin Exp Immunol 2000; 121:248–254.
13. Schwertz E, Kahlenberg F, Sack U, et al. Serologic assay based on gliadin-related nonapeptides as a highly sensitive and specific diagnostic aid in celiac disease. Clin Chem 2004; 50:2370–2375.
14. Dieterich W, Esslinger B, Trapp D, et al. Cross linking to tissue transglutaminase and collagen favours gliadin toxicity in coeliac disease. Gut 2006; 55:478–484.
15. Kaukinen K, Collin P, Laurila K, et al. Resurrection of gliadin antibodies in coeliac disease. Deamidated gliadin peptide antibody test provides additional diagnostic benefit. Scand J Gastroenterol 2007; 42:1428–1433.
16. Tonutti E, Visentini D, Picierno A, et al. Diagnostic efficacy of the ELISA test for the detection of deaminated anti-gliadin peptide antibodies in the diagnosis and monitoring of celiac disease. J Clin Lab Anal 2009; 23:165–171.
17. Volta U, Granito A, Fiorini E, et al. Usefulness of antibodies to deaminated gliadin peptides in celiac disease diagnosis and follow-up. Dig Dis Sci 2008; 53:1582–1588.
18. Rashtak S, Ettore MW, Homburger HA, et al. Comparative usefulness of deamidated gliadin antibodies in the diagnosis of celiac disease. Clin Gastroenterol Hepatol 2008; 6:426–432.
19. Liu E, Li M, Emery L, et al. Natural history of antibodies to deamidated gliadin peptides and transglutaminase in early childhood celiac disease. J Pediatr Gastroenterol Nutr 2007; 45:293–300.
20. Basso D, Guariso G, Fogar P, et al. Antibodies against synthetic deamidated gliadin peptides for celiac disease diagnosis and follow-up in children. Clin Chem 2009; 55:150–157.
21. Marsh MN. Gluten, major histocompatibility complex, and the small intestine: a molecular and immunobiologic approach to the spectrum of gluten sensitivity (‘celiac sprue’). Gastroenterology 1992; 102:330–354.
22. Prause C, Ritter M, Probst C, et al. Antibodies against deamidated gliadin as new and accurate biomarkers of childhood coeliac disease. J Pediatr Gastroenterol Nutr 2009; 49:52–58.
23. Vermeersch P, Geboes K, Mariën G, et al. Diagnostic performance of IgG anti-deamidated gliadin peptide antibody assays is comparable to IgA anti-tTG in celiac disease. Clin Chim Acta 2010; 411:931–935.
24. Jaskowski TD, Donaldson MR, Hull CM, et al. Novel screening assay performance in pediatric celiac disease and adult dermatitis herpetiformis. J Pediatr Gastroenterol Nutr 2010; 51:19–23.
25. Agardh D. Antibodies against synthetic deamidated gliadin peptides and tissue transglutaminase for the identification of childhood celiac disease. Clin Gastroenterol Hepatol 2007; 5:1276–1281.
26. Hogen Esch CE, Csizmadia GD, van Hoogstraten IM, et al. Childhood coeliac disease: towards an improved serological mass screening strategy. Aliment Pharmacol Ther 2010; 31:760–766.
27. Lewis NR, Scott BB. Meta-analysis: deamidated gliadin peptide antibody and tissue transglutaminase antibody compared as screening tests for coeliac disease. Aliment Pharmacol Ther 2010; 31:73–81.
28. Leffler DA, Edwards George JB, Dennis M, et al. A prospective comparative study of five measures of gluten-free diet adherence in adults with coeliac disease. Aliment Pharmacol Ther 2007; 26:1227–1235.
29. Sugai E, Nachman F, Váquez H, et al. Dynamics of celiac disease-specific serology after initiation of a gluten-free diet and use in the assessment of compliance with treatment. Dig Liver Dis 2010; 42:352–358.
30. Holmes GK, Prior P, Lane MR, et al. Malignancy in coeliac disease: effect of a gluten free diet. Gut 1989; 30:333–338.
31. Kolsteren MM, Koopman HM, Schalekamp G, et al. Health related quality of life in children with celiac disease. J Pediatr 2001; 138:593–595.
32. Wagner G, Berger G, Sinnreich U, et al. Quality of life in adolescents with treated coeliac disease: influence of compliance and age at diagnosis. J Pediatr Gastroenterol Nutr 2008; 47:555–561.
33. Nachman F, Del Campo MP, González A, et al. Long-term deterioration of quality of life in adult patients with celiac disease is associated with treatment noncompliance. Dig Liv Dis 2010; 42:685–691.