See “Coeliac Disease: Time for a New Diagnostic Approach in Symptomatic Children” by Volta on page 241.
Strict diagnostic criteria for celiac disease (CD) in children were first established by the European Society of Pediatric Gastroenterology and Nutrition in 1969 (1). The so-called 3-biopsy rule recommended performing at least 3 small bowel biopsies (SBBs): characteristic histological lesions in the first SBB led to the diagnosis of suspected CD, but only after confirmation of histological relapse in a third biopsy related to gluten challenge (GC), a definite diagnosis of CD could be confirmed (1).
These rules were used for almost 20 years until experience in large series of children showed that the GC could probably be avoided in 95% of the cases (2). Thus, diagnostic criteria were modified in 1990, and GC was restricted to infants younger than 2 years at the first biopsy to exclude other causes of enteropathy or whenever the initial diagnosis is uncertain (3). Moreover, the new criteria considered for the first time that presence of CD anti-gliadin antibodies (AGA) at disease onset, followed by antibody vanishing after gluten withdrawal, added support to the diagnosis (3–5); however, further development of anti-endomysial antibodies (EMA) in the late 1980s (6–8), followed by tissue transglutaminase (tTG) being recognized as the autoantigen of CD in the 1990s, represented a true revolution in the field of CD diagnosis (9). It has been shown that both IgA EMA and IgA-anti-tTG recognize the same autoantigen and overall display a sensitivity (S) and specificity (Sp) for CD diagnosis >95% (10–12). More recently, serological tests for antibodies against deamidated gliadin peptides (DGP) (13) have displayed a higher S and Sp than conventional AGA. Correlation between duodenal histopathology and tTG levels in pediatric patients with CD has been repeatedly reported, and higher levels were always associated with characteristic villous atrophy (14–17). Thus, it has been suggested that with further validation of this observation in larger series of patients, strongly positive tTG titer may be considered sufficient for CD diagnosis in children (15–18); however, for less-severe histologic lesions, lower S is observed for EMA and tTG, ranging from 89% to as low as 30%. Besides, positive predictive value (PPV) for biopsy confirmation of CD in asymptomatic individuals remains to be ascertained (18).
Additionally, in the last few years, a strong association of CD with the genetic markers human leukocyte antigen (HLA)-DQ2 and/or HLA-DQ8, which when combined reach an S >96% in most populations, has been established (19,20). This implies that a negative result of HLA-DQ2 and/or DQ8 renders CD diagnosis highly unlikely (20,21).
Notwithstanding this new scenario, European Society of Pediatric Gastroenterology, Hepatology, and Nutrition (ESPGHAN) 1990 CD diagnostic criteria have prevailed for 20 years, and even recent published guidelines still require a characteristic histological lesion of the small bowel mucosa to establish the diagnosis of CD (22,23).
Lately, critical voices have been claiming that, in view of their high diagnostic accuracy, specific antibody testing may replace the SBB in the diagnostic procedure for CD, at least for some selected patients, especially those with high anti-tTG antibody levels (14). This position is concordant with the accumulated experience showing that conventional histological examination is a questionable criterion standard for the diagnosis of CD (24–26). Additionally, a high relapse rate after GC in children younger than 2 years with positive EMA and villous atrophy at diagnosis has been demonstrated, supporting the view that routine GC should not be mandatory in these cases (27,28).
A recent survey was conducted among ESPGHAN members to get insight into the real practices for CD diagnosis and to identify what modifications, if any, were demanded by the majority of members of our society (29). Interestingly, 90% requested a revision and modification of the present criteria; 44% wanted to omit the first SBB in symptomatic cases with positive TTG IgA or EMA IgA in HLA-DQ2/DQ8–positive individuals. Even so, approximately half of the respondents believed that challenge should not be mandatory for all children diagnosed (first biopsy) before the age of 2 (29).
In parallel, the ESPGHAN working group performed a revision of scientific and technical developments in an evidence-based approach, producing a detailed evidence report on antibody testing in CD (30), which forms the basis of the new guidelines for CD diagnosis recently published (21). Additionally, the working group developed a new and broader definition of CD. Accordingly, the diagnosis cannot rely on 1 single parameter, but on a combination of clinical symptoms, CD-specific autoantibodies, histology, and genetics (21). A scoring system was proposed, but it has to be formally evaluated in prospective clinical studies before it can be recommended in regular clinical use (21).
The implications of the change of the present criteria are obvious, especially regarding the omission of a biopsy in well-selected cases. Thus, the working group considers that a period of implementation and testing is mandatory before these guidelines definitively replace the existing ESPGHAN 1990 guidelines.
The aim of our study was to evaluate the efficiency of a set of 2 serological markers (SMs) plus HLA haplotype, to establish the diagnosis of CD independent of histology.
We conducted a retrospective study involving all consecutive patients suspicious of having CD who had undergone an SBB for diagnostic purposes at the Pediatric Gastroenterology Service of La Fe Hospital between January 2005 and September 2010.
Only symptomatic patients—with chronic or intermittent diarrhea, failure to thrive, weight loss, stunted growth, iron-deficiency anemia, nausea or vomiting, chronic abdominal pain, cramping or distension, chronic constipation, abnormal liver biochemistry—were included in the present study. At the time of the SBB, they were checked for 3 CD-related tests: HLA-DQ2/DQ8 typing, IgA tissue transglutaminase antibody (tTGA), and IgA EMA. Asymptomatic children, who underwent a biopsy as a consequence of positive serology in screening programs, were excluded. Even so, patients with a time span >2 months in between serology and SBB, or which had started a gluten-free diet (GFD) before evaluation, were not included.
Based on the new proposed ESPGHAN guidelines for the diagnosis of CD in children and adolescents (21), we decided to consider the set of tTGA, EMA, and HLA-DQ2/DQ8 as a triple test (TT) for diagnostic approach. We defined a TT as positive if tTGA was ≥60 IU/mL, that is, 10 times the upper limit of normal (ULN) of the used test, together with a positive EMA and a positive HLA-DQ2/DQ8. TT was considered to be negative whenever these 3 conditions were not fulfilled.
SBBs were obtained using the Crosby capsule, and evaluated in the pathology unit after paraffin embedding and hematoxylin and eosin staining. Samples were categorized by the Marsh-Oberhuber classification. The histology was considered the criterion standard for CD diagnosis according to the 1990 ESPGHAN guidelines.
Antibody tests were performed in our investigation laboratory. IgA tTG was performed by an indirect noncompetitive enzyme-linked immunosorbent assay, EliA TM Celikey (Phadia GmbH, Freiburg, Germany) in an automatic system Phadia 100 (Phadia AB, Uppsala, Sweden). Serum dilution was 1:100; the cutoff, as previously established in our population, was 6 IU/mL. Antibody levels were calculated in units per milliliter using a 6-parameter standard curve, as provided by the manufacturer, the range being between 0 and 128 U/mL. EMA was determined by an immunofluorescence in-house method using monkey esophagus (BioSystems, Barcelona, Spain) as antigen, and a polyclonal rabbit anti-human IgA/fluorescein isothiocyanate conjugate (DAKO A/S, Glostrup, Denmark). Positive cutoff was established at a 1:5 dilution. EMA assays were blindly read by 2 experienced independent observers of our Pediatric Gastroenterology Investigation Laboratory.
Total serum IgA was estimated according to age by nephelometry (BN II System, Siemens, Marburg, Germany) and estimated according to age.
A polymerase chain reaction was performed with specific DQA1*0501/0505 and DQB1*02 primers for the DQ2 heterodimer and DQB1* 0302 primers for the DQB8 molecule.
A patient was considered positive for HLA-DQ2/DQ8 test if he/she presented a haplotype considered of risk in our specific population: DQA1*05:01-DRB1*03-DQB1*02:01 in homo- or heterozygosity; DQA1*02:01DRB1*07-DQB1*02:02/DQA1*05:05-DRB1*11-DQB1*03:01; DQA1*03-DRB1*04-DQB1*03:02 in homo- or heterozygosity. Otherwise, the patient was considered negative for HLA-DQ2/DQ8.
S, Sp, PPV, and negative predictive value (NPV) of the TT were determined using conventional formulas. In addition, a test for equality of proportions with continuity correction (chi-square approximation) was used.
A total of 153 patients referred to the gastroenterology unit from January 2005 to September 2010 met the inclusion criteria and were enrolled in the study; 65 were boys and 88 were girls. The age range was 9 months to 14.6 years, mean being 4 years ± 3.6. From the 153 included patients, 71 (46%) were younger than 2 years. Three were excluded because no definitive diagnosis has been established yet, and thus 150 patients were considered for final analysis and statistical evaluation.
A total of 116 children were positive for the TT. Of them, 113 (97.4%) had either a Marsh 2 or a Marsh 3 (4 and 109 patients respectively) histological lesion and a definite diagnosis of CD, and were true positive cases for the TT. Thus, the PPV of the TT was 97.4%. The remaining 3 children (2.6%) had Marsh 0 and were diagnosed as non-CD cases and were considered as false-positives for the TT. Two of these children started a GFD immediately after biopsy; subsequently, their symptoms improved and tTGA levels showed a significant decrease. After a variable period with a GFD, it was decided in agreement with the parents to start a GC, and after 7 and 14 months, respectively, both children had clinical and serological relapse. They underwent a second biopsy, which showed a Marsh 3 lesion; thus, CD was confirmed in these 2 cases.
The third child of this group had elevated liver enzymes without other symptoms, and minor histological changes were reported. No dietary restrictions were advised. During follow-up, although the patient persisted asymptomatic and transaminases normalized without any treatment, tTGA remained positive (>10 times ULN), so a second biopsy, after 5 years, was performed that revealed a Marsh 3 lesion.
Thus, at the end, if we reclassify these 3 children with a final diagnosis of CD after follow-up, then the PPV of the TT turns out to be 100% (Table 1).
In addition, 34 patients had a negative TT. In 9 cases, CD was ruled out: 6 had normal histology, and the other 3 cases had another diagnosis (cow's-milk protein allergy), that is, these 9 patients are the true negatives of the TT. The remaining 25 TT-negative patients had different combinations of positive and negative single tests.
In 3 of 25 children, a Marsh 2–3 lesion was found, whereas HLA-DQ2/DQ8, EMA, and tTGA were negative. During follow-up, one of them is receiving GC for >14 months up to now, has not had any clinical or serological relapse, and no postchallenge biopsy has been performed yet.
Another child is 4 years old and challenge is still pending. The last one, a 7-year-old girl, is receiving a GFD and the parents are reluctant to perform GC. Although these 3 children are considered false-negative for TT, their final diagnosis is not yet confirmed (Fig. 1).
Based on these results, Sp for the TT was 100% and NPV 26%. NPV could improve if the 3 mentioned children finally are diagnosed as non-CD patients (Table 2).
In 22 of 25 children, a definite diagnosis of CD had been established. All 22 showed either a Marsh 2 or Marsh 3 lesion, and all had only 1 or 2 positive markers of the 3 considered in the TT. One girl initially having positive EMA plus at-risk HLA but tTGA <10 times ULN and Marsh 1 remained on a gluten-containing diet; during follow-up, she continued slightly symptomatic and, after 7 months, became TT positive with tTGA >10 times ULN and the histology confirmed Marsh 3, so finally CD was diagnosed.
It is remarkable that 12 of these 22 patients with CD (54.5%) had positive EMA with positive HLA-DQ2 and/or DQ8, and tTGA was >5 times ULN (Fig. 2). The remaining 10 patients had different combinations of SMs.
Four children (4/22) were tTGA and EMA positive (3 had tTGA >10 times ULN and 1 tTGA >2 times ULN), and were HLA not DQ2/DQ8, but in fact all of them were DR7 DQA1*02:01-DQB1*02:02 (half DQ2 molecule), which is 2.8% of the total of children having CD.
Finally, by using a sample test for equality of proportions with continuity correction (chi-square approximation), we compared the TT with the biopsy to determine how likely CD can be predicted. No significant difference was obtained when comparing both methods: P = 0.327 (95% confidence interval −0.052 to 0.010).
In the last few years, many publications have emerged focusing the discussion on the role of histology as the reference standard for the diagnosis of CD. There is a high possibility to avoid this procedure in some clinical circumstances, based on the high efficiency of SMs and the high NPV of HLA non-DQ2/DQ8 (14,31,32).
The value of specific antibodies for CD diagnosis has been extensively studied, evidence showing anti-tTGA and EMA to be the most efficient serological tests. In children, tTGA and EMA have reported diagnostic sensitivities and specificities ≥95% (33); however, the HLA-DQ2/DQ8 genotyping is a test with a high NPV for CD, >95%, and its main role is disease exclusion. Recently, deamidated gliadin antibodies have demonstrated a greater value than conventional gliadin in CD diagnosis, although they cannot replace tTG/EMA, their main utility being in diagnosis in children younger than 2 years when the other SMs are negative (21).
The new guidelines from the ESPGHAN are an evidence-based approach to this question, changing the previous 1990 diagnostic criteria (21).
Based on their recommendations, we wanted to prove and confront in clinical symptomatic children with clinical suspicion of CD the accuracy of our previously defined TT in comparison with the histological results of the biopsies and the final diagnosis including a follow-up period. Using the TT, all of the children who have a proven CD would have been correctly diagnosed. Therefore, we would be able to avoid an SBB in approximately 77% of the patients with suspicion of CD, sparing an invasive procedure and avoiding the limitations of histology as well as the consequences of prolonged waiting time for diagnosis confirmation (overtreatment or subtreatment).
In the group of children younger than 2 years, serological testing may be less reliable with more false-negative results for tTGA and EMA antibodies (34), thus reducing or lowering the NPV; however, newer data suggest that the mentioned levels of accuracy for older children are also valid for children younger than 2 years (33,35). One study compared IgA AGA, IgA EMA, and IgA tTG results in infants up to 2 years of age and in children older than 2. Similar results were found in both age groups for IgA tTG and for IgA EMA (36). Clouzeau-Girard (37) recently published a 98.8% frequency of positive IgA tTG and/or EMA in a group of children with a median age of 18 months that have a biopsy-proven CD. Anyhow, the PPV for this age group would still be 100%; however, only 1 celiac infant younger than 2 years was negative for tTG and EMA, and 1 more only for EMA. This is different from a previous study of our group, which found approximately 8% of children with CD younger than 2 years to be negative for tTG and EMA.
Overall, 71 of the 150 patients enrolled in our study were 2 years or younger; 61 had a confirmed CD diagnosis, and 88.5% had a positive TT. In addition, 6 children had either a positive tTGA or EMA. Only 1 patient younger than 2 years with CD had negative tTGA and EMA: a 10-month infant with a Marsh 3b biopsy, HLA-DQ2 positivity, negative tTGA and EMA;AMA were slightly positive 0.64 UA (RV < 0.30 AU), and intestinal IgA deposits against tTG were positive. These results reaffirm the usefulness of both tTG and EMA antibodies in this age group.
The 3 patients we found having initially a positive TT but with a negative biopsy developed a Marsh 3 lesion during follow-up while maintaining a gluten-containing diet. These patients are a good example of why biopsies are questioned in some way as the main reference pattern for diagnosis. Some sources of pitfalls are the possibility of a patchy pattern of histological lesions (38); histological variability can exist between different duodenal sites as well as within single biopsies (39); interobserver variation and differences in preparation quality of the tissue sample are also possible. Even less frequent but especially remarkable in younger children is the fact that other conditions different than CD can express partial villous atrophy, intraepithelial lymphocytosis, and crypt hyperplasia such as Giardia lamblia infection and cow's milk allergy, mimicking CD, as in 1 of our cases.
However, there is also evidence that the mentioned antibodies may precede the development of structural abnormalities in the small bowel where the atrophy represents the end stage of the disease (10,26,40). So, in early stages of the disease, antibodies could have a prominent role in the diagnosis of CD.
Strongly positive tTG antibodies are associated with higher degrees of villous atrophy in adult and pediatric patients (14,15,17). The correlation between tTGA antibody and duodenal histopathology is especially remarkable in children, as adults tend to have minor histopathology changes together with lower tTGA titers (41).
Different authors used distinct upper level values for IgA tTG antibody to correlate with duodenal histology characteristic of CD, ranging from ≥30 to ≥100 U. Considering the different ULN of the commercial kits used (ranging from 4 to 30 IU/mL), values 3-fold or above the positive cutoff would be a reliable predictor of a Marsh 3 lesion, with PPV for CD histopathology from 96% to 100% in the different studies (15,16,41).
According to the new ESPGHAN recommendations, we used the suggested 10 times the ULN in a concentration-dependent antibody test, based on calibration curves, which has a 100% PPV of CD (30,32). Expressing the results as multiples of the ULN, as proposed by Hill and Holmes (32), allows comparing results between different commercial assays with different cutoff limits. Nonetheless, in our study of the 22 patients with CD who did not met the criteria for a positive TT, 15 had positive EMA and positive DQ2, but tTGA was <10 times the ULN. Twelve of these 15 patients had tTGA ≥5 times ULN. It will be of interest to assess in forthcoming studies with larger series of patients the possibility of lowering this 10 times the threshold value, with the aim to improve the S without decreasing the Sp. It should also be noted that the Hill and Holmes (32) study was done in adults without considering concomitant EMA results.
Using a threshold value ≥5 times the ULN as a single serological CD marker, we would have had diagnosed 128 of our 141 patients with CD (Fig. 2), thus rising S to 90%, without changing the PPV of the TT. Only 1 patient without CD had tTGA >10 times the ULN, but was negative for EMA. He was diagnosed as having cow's-milk protein allergy.
Some authors report little benefit to test both antibodies (tTGA plus EMA) simultaneously because the concordance rate of these tests is high (42), even though some occasional patients remain negative for EMA despite positive tTGA antibodies and vice versa (40,43). The use of 2 antibodies (tTGA, EMA) in the context of these new approaches that aim to select those children not needing an intestinal biopsy minimizes the risk of a false-positive diagnosis, putting them unnecessarily on a lifelong GFD. Isolated positivity for tTGA antibodies can occur in conditions unrelated to CD such as cow's-milk protein allergy as the patient mentioned above, infections (Epstein-Barr, Coxsackie virus), autoimmune diseases, tumors, myocardial damage, liver disorders, inflammatory bowel disease, and psoriasis (44–46). Freeman (47) reported a few cases with tTGA titers ≥100 U with normal biopsy specimens.
The ESPGHAN evidence report on CD serology (30) estimates that EMA has a higher reliability for the diagnosis of CD that reveals almost an absolute Sp.
In the present study, there was 1 patient who did not fulfill the criteria of a positive TT, HLA genotyping was negative for DQ2/DQ8, serological test showed a tTGA ≥10 times the ULN, and surprisingly, he had positive EMA (1:40) antibodies (false-positive). The biopsy result was Marsh 0–1. This patient's diagnosis was cow's milk allergy. The search for mucosal deposits of anti tTG-IgA was negative, ruling out CD (25). On follow-up, after excluding cow's milk protein while maintaining a gluten-containing diet, symptoms resolved completely after a couple of months and by then TTG and EMA were negative. No clinical relapse was ascertained on further follow-up.
A large UK study in adults considered the use of both IgA tTG and EMA. The present study identified improvements in PPV and some small differences in S, Sp, and NPV if both tests were used, either in a 2-step process or simultaneously, compared with if tests were completed individually (48).
Regarding HLA, 4 of 141 patients with CD were negative for HLA-DQ2/DQ8, representing 2.8%, which is in keeping with a previous study in our community (49); however, these 4 patients bear a half-DQ2 molecule, which could be responsible for the risk to develop CD.
The high PPV of the TT and the absence of significant differences between TT and biopsy in their likelihood to diagnose CD make the TT an attractive diagnostic tool to avoid an SBB; however, the low NPV of the TT reinforces the new ESPGHAN recommendation that all children with different combinations of serological and genetic markers not fulfilling the criteria of a positive TT should undergo an SBB to confirm CD diagnosis.
Moreover, the presence of positive antibodies, tTGA or EMA, in patients having a normal SBB does not necessarily constitute a false-positive finding because they may be a sign of early disease, predicting forthcoming histological lesions (50).
In conclusion, because of the high PPV and Sp in our TT, this preliminary study supports the view that in selected population of symptomatic and TT-positive children, CD diagnosis could be established without the need of performing an SBB, in agreement with the new ESPGHAN guidelines.
Although the results shown here are promising, they should be validated in a greater population. A further large-scale, prospective, collaborative confirmatory study is required.
1. Meeuwisse GW. Diagnostic criteria in coeliac disease. Acta Paediatr Scand 1970; 59:461–463.
2. Guandalini S, Ventura A, Ansaldi N, et al. Diagnosis of coeliac disease: time for a change? Arch Dis Child 1989; 64:1320–1324.
3. Walker-Smith JA, Guandalini S, Schmitz J, et al. Revised criteria for diagnosis of coeliac disease. Arch Dis Child 1990; 65:909–911.
4. Ribes-Koninckx C, Giliams JP, Polanco I, et al. IgA antigliadin antibodies in celiac and inflammatory bowel disease. J Pediatr Gastroenterol Nutr 1984; 3:676–682.
5. Troncone R, Ferguson A. Anti-gliadin antibodies. J Pediatr Gastroenterol Nutr 1991; 12:150–158.
6. Grodzinsky E, Jansson G, Skogh T, et al. Anti-endomysium and anti-gliadin antibodies as serological markers for coeliac disease in childhood: a clinical study to develop a practical routine. Acta Paediatr 1995; 84:1–7.
7. De Lecea A, Ribes-Koninckx C, Polanco I, et al. Serological screening (antigliadin and antiendomysium antibodies) for nonovert coeliac disease in children of short stature. Acta Paediatr Suppl 1996; 412:54–55.
8. Korponay-Szabo I, Kovacs J, Czinner A, et al. High prevalence of silent celiac disease in preschool children screened with IgA/IgG antiendomysium antibodies. J Pediatr Gastroenterol Nutr 1999; 28:26–30.
9. Dieterich W, Ehnis T, Bauer M, et al. Identification of tissue transglutaminase as the autoantigen of celiac disease. Nat Med 1997; 3:797–801.
10. Troncone R, Maurano F, Rossi M, et al. IgA antibodies to tissue transglutaminase: an effective diagnostic test for celiac disease. J Pediatr 1999; 134:166–171.
11. Rostom A, Dubé C, Cranney A, et al. Celiac disease. In: Summary, Evidence Report/Technology Assessment no. 104. AHRQ Publication no. 04-E029-1. Rockville, MD: Agency for Healthcare Research and Quality; 2004.
12. Li M, Yu L, Tiberti C, et al. A report on the International Transglutaminase Autoantibody Workshop for Celiac Disease. Am J Gastroenterol 2009; 104:154–163.
13. 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.
14. Barker CC, Mitton C, Jevon G, et al. Can tissue transglutaminase antibody titers replace small-bowel biopsy to diagnose celiac disease in select pediatric populations? Pediatrics 2005; 115:1341–1346.
15. Donaldson MR, Firth SD, Wimpee H, et al. Correlation of duodenal histology with tissue transglutaminase and endomysial antibody levels in pediatric celiac disease. Clin Gastroenterol Hepatol 2007; 5:567–573.
16. Donaldson MR, Book LS, Leiferman KM, et al. Strongly positive tissue transglutaminase antibodies are associated with Marsh 3 histopathology in adult and pediatric celiac disease. J Clin Gastroenterol 2008; 42:256–260.
17. Dahlbom I, Korponay-Szabó IR, Kovács JB, et al. Prediction of clinical and mucosal severity of coeliac disease and dermatitis herpetiformis by quantification of IgA/IgG serum antibodies to tissue transglutaminase. J Pediatr Gastroenterol Nutr 2010; 50:140–146.
18. Tursi A, Brandimarte G, Giorgetti GM. Prevalence of antitissue transglutaminase antibodies in different degrees of intestinal damage in celiac disease. J Clin Gastroenterol 2003; 36:219–221.
19. Mearin ML, Ribes-Koninckx C, Biemond I, et al. Influence of genetic factors on the serum levels of antigliadin antibodies in celiac disease. J Pediatr Gastroenterol Nutr 1984; 3:373–377.
20. Margaritte-Jeannin P, Babron MC, Bourgey M, et al. HLA-DQ relative risks for coeliac disease in European populations: a study of the European Genetics Cluster on Coeliac Disease. Tissue Antigens 2004; 63:562–567.
21. Husby S, Koletzko S, Korponay-Szabo I, et al. European Society for Pediatric Gastroenterology, Hepatology, and Nutrition guidelines for the diagnosis for coeliac disease. J Pediatr Gastroenterol Nutr 2012; 54:136–160.
22. 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.
23. Rostom A, Murray JA, Kagnoff MF, et al. American Gastroenterological Association (AGA) Institute Technical Review on the Diagnosis and Management of Celiac Disease. Gastroenterology 2006; 131:1981–2002.
24. Kaukinen K, Collin P, Maki M, et al. Latent celiac disease or celiac disease beyond mucosal atrophy. Gut 2007; 56:1339–1340.
25. Koskinen O, Collin P, Korponay-Szabo I, et al. Gluten-dependent small bowel mucosal transglutaminase 2-specific IgA deposits in overt and mild enteropathy coeliac disease. J Pediatr Gastroenterol Nutr 2008; 47:436–442.
26. Kurppa K, Ashorn M, Iltanen S, et al. Celiac disease without villous atrophy in children: a prospective study. J Pediatr 2010; 157:373–380.
27. Killander A, Arnell H, Hagenas L, et al. Omitting control biopsy in paediatric coeliac disease: a follow-up study. Acta Paediatr 2007; 96:1190–1194.
28. Wolters VM, van de Nadort C, Gerritsen SA, et al. Is gluten challenge really necessary for the diagnosis of coeliac disease in children younger than age 2 years? J Pediatr Gastroenterol Nutr 2009; 48:566–570.
29. Ribes-Koninckx C, Mearin Mª L, Korponay-Szabó IR, et al. Coeliac disease diagnosis: ESPGHAN criteria or need for a change? Results of a questionnaire. J Pediatr Gastroenterol Nutr 2011; 77:5316–5323.
30. Giersiepen K, Lelgemann M, Stuhldreher N, et al. ESPGHAN Working Group on Coeliac Disease Diagnosis. Accuracy of diagnostic antibody tests for coeliac disease in children: summary of an evidence report. J Pediatr Gastroenterol Nutr 2012; 54:229–241.
31. Murdock AM, Johnston SD. Diagnostic criteria for celiac disease: time for change? Eur J Gastroenterol Hepatol 2005; 17:41–43.
32. Hill PG, Holmes GK. Coeliac disease: a biopsy is not always necessary for diagnosis. Aliment Pharmacol Ther 2008; 27:72–577.
33. Hill ID. What are the sensitivity and specificity of serologic tests for celiac disease? Do sensitivity and specificity vary in different populations? Gastroenterology 2005; 128:25–32.
34. 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.
35. Jones HJ, Warner JT. NICE clinical guidelines 86. Celiac disease: recognition and assessment of celiac disease. Arch Dis Child 2010; 95:312–313.
36. Viola L, Lugli L, Melegari A, et al. Antitransglutaminase enzyme-linked immune-adsorbed assay in coeliac disease diagnosis: evaluation of a diagnostic algorithm. Pediatr Med Chir 2004; 26:126–131.
37. Clouzeau-Girard H, Rebouissoux L, Taupin JL, et al. HLA-DQ genotyping combined with serological markers for the diagnosis of celiac disease: is intestinal biopsy still mandatory? J Pediatr Gastroenterol Nutr 2011; 52:729–733.
38. Bonamico M, Mariani P, Thanasi E, et al. Patchy villous atrophy of the duodenum in childhood celiac disease. J Pediatr Gastroenterol Nutr 2004; 38:204–207.
39. Ravelli A, Villanacci V, Monfredini CH, et al. How patchy is patchy villous atrophy? Distribution of children with celiac disease. Am J Gastroenterol 2010; 105:2103–2110.
40. Salmi TT, Collin P, Jarvinen O, et al. Immunoglobulin A autoantibodies against transglutaminase 2 in the small intestinal mucosa predict forthcoming celiac disease. Aliment Pharmacol Ther 2006; 24:541–552.
41. Vivas S, Ruiz de Morales JG, Riestra S, et al. Duodenal biopsy may be avoided when high transglutaminase antibody titers are present. World J Gastroenterol 2009; 15:4775–4780.
42. Maki M, Mustalahti K, Kokkonen J, et al. Prevalence of celiac disease among children in Finland. N Engl J Med 2003; 348:2517–2524.
43. Tesei N, Sugai E, Vásquez H, et al. Antibodies to human recombinant tissue transglutaminase may detect coeliac disease patients undiagnosed by endomysial antibodies. Aliment Pharmacol Ther 2003; 17:1415–1423.
44. Clemente MG, Musu MP, Frau F, et al. Antitissue transglutaminase antibodies outside celiac disease. J Pediatr Gastroenterol Nutr 2002; 34:31–34.
45. Di Tola M, Sabbatella L, Anania MC, et al. Anti-tissue transglutaminase antibodies in inflammatory bowel disease: new evidence. Clin Chem Lab Med 2004; 42:1092–1097.
46. Ferrara F, Quaglia S, Caputo I, et al. Antitransglutaminase antibodies in non-coeliac children suffering from infectious diseases. Clin Exp Immunol 2010; 159:217–223.
47. Freeman HJ. Strongly positive tissue transglutaminase antibody assays without celiac disease. Can J Gastroenterol 2004; 18:25–28.
48. Hopper AD, Hadjivassiliou M, Hurlstone DP, et al. What is the role of serologic testing in celiac disease? A prospective, biopsy-confirmed study with economic analysis. Clin Gastroenterol Hepatol 2008; 6:314–320.
49. Donat E, Planelles D, Capilla A, et al. Allelic distribution and the effect of haplotype combination for HLA type II loci in the celiac disease population of the Valencian community (Spain). Tissue Antigens 2009; 73:255–261.
50. Holmes GKT. Potential and latent celiac disease. Eur J Gastroenterol 2001; 13:1057–1060.