Awareness of celiac disease (CD) has increased during the last few decades, but the majority of patients with CD remain undiagnosed (1,2). CD has a wide range of clinical manifestations, and nongastrointestinal manifestations are common (3–5). Many patients are discovered by screening high-risk groups such as patients with type 1 diabetes mellitus, Mb Down, or first-degree relatives of patients with CD (6–8). CD is at present one of the most common chronic diseases in childhood. It is classified as an autoimmune disease in which dietary gluten induces the disease in genetically susceptible individuals who carry the HLA-DQ2 and/or the HLA-DQ8 haplotype (9,10). An effective treatment is available by adherence to a gluten-free diet (GFD).
CD is characterized by small intestinal mucosal enteropathy that is commonly graded according to the Marsh-Oberhuber criteria and can vary from intraepithelial lymphocytosis (grade 1) to more extensive lesions with crypt hyperplasia (grade 2) and various degrees of villous atrophy (grade 3) (11,12). In children, the mucosal lesions may have a patchy distribution and are sometimes located only in the duodenal bulb (13,14). Until recently, a biopsy of the small intestine with the characteristic gluten-induced enteropathy was mandatory for a CD diagnosis. According to the newly revised guidelines from the European Society for Pediatric Gastroenterology, Hepatology, and Nutrition (ESPGHAN), however, the biopsy is no longer mandatory for diagnosis in symptomatic patients when the serology marker levels exceed 10 times the upper limit of normal in combination with positivity for endomysial antibodies (EMA) and HLA DQ2/DQ8 (15). The ESPGHAN recommendations, however, suggest that all of the asymptomatic patients should be biopsied to confirm the diagnosis.
The first step in screening and diagnosing CD is therefore the evaluation of the serological markers, anti-tissue transglutaminase antibodies of type immunoglobulin A (tTG-IgA), and, as a possible second step, anti-EMA. The sensitivity and specificity of both methods are high (16). It is notable that the EMA test has a slightly higher specificity because low levels of tTG-IgA can occur in conditions unrelated to CD, but this is less likely to be the case for EMA (17,18). Several studies of symptomatic patients with CD have shown a weak correlation between lower levels of tTG-IgA and intestinal enteropathy (19,20). In contrast, other studies indicate that high levels of tTG-IgA (>100 U/mL) have a high specificity for Marsh 3 lesions (19,21–26), which underpins the revision of the ESPGHAN diagnostic guidelines. To the best of our knowledge, however, there are yet no data on the correlation between tTG-IgA levels and the degree of enteropathy in screening-detected patients with CD from the general childhood population.
The primary aim of the present study was to evaluate a potential correlation between the level of tTG-IgA and the degree of gluten-induced enteropathy in screening-detected patients with CD from a general child population and to assess to what extent the revised ESPGHAN guidelines can be extended to cover this group of patients.
The present study is a substudy of a cross-sectional CD screening study entitled “Exploring the Iceberg of Celiacs in Sweden,” which is part of the European Prevent CD project. The screening was school-based and included 2 birth cohorts of 12-year-old children born in 1993 or 1997 with the screening taking place in 2005 to 2006 and 2009 to 2010, respectively. Five pediatric departments collaborated in the present study (Umeå, Norrtälje, Norrköping, Växjö, and Lund), and each of the sites included a major city with municipalities in the surrounding suburbs and countryside. A total of 18,325 children were invited, and 13,279 consented to CD screening (Fig. 1). The proportion of girls was similar in both cohorts (48% and 49%, respectively). Positive serological markers were found in 291 children, and these children were invited to a further investigation with a small intestinal biopsy.
Inclusion criteria for the present substudy were that a small intestinal biopsy had been performed and a second blood sample drawn at the time of the biopsy had been analyzed for tTG-IgA. The Exploring the Iceberg of Celiacs in Sweden study was approved by the regional ethical review board of Umeå University, Umeå, Sweden.
Participating children were asked for a blood sample at their school, and a second serum sample was obtained at the time of biopsy (Fig. 1). The samples were analyzed at the same laboratory for tTG-IgA by enzyme-linked immunosorbent assay (Celikey, Phadia GmbH, Freiburg im Breisgau, Germany). The manufacturer's cutoff for positive tTG-IgA was 5 U/mL, but to increase the sensitivity of the screening we chose to consider all of the values >4 U/mL to be elevated. Serum samples were analyzed in duplicate within the measuring range 0 to 100 U/mL, and the mean value was used. Serum samples with intermediate values of tTG-IgA (2–4 U/mL) were additionally analyzed for EMA-with a 1:5 dilution as a cutoff. EMA were determined by indirect immunofluorescence using monkey esophagus tissue (Binding Site, Birmingham, UK).
All of the children (n = 291) with elevated serological markers at the screening occasion were referred to a pediatric clinic for a follow-up with a second celiac serology analysis and a small intestinal biopsy. A total of 278 children accepted the follow-up screening procedure, which has previously been described in detail (27,28).
Of the 278 children who had a biopsy performed, 267 children met the inclusion criteria for the study. Five children in the 1993 cohort had IgA deficiency and tTG-IgG >6 U/mL and were excluded, and another 5 children were excluded because no serum samples were taken at the time of the biopsy. One additional patient was also excluded because of an inconclusive biopsy. Of the 267 children in the present study (156 girls and 111 boys), 230 were diagnosed as having CD. There were no significant differences between the 2 birth cohorts with respect to age or sex, so the data for the 2 birth cohorts were combined.
In the 1993 cohort, the small intestinal biopsy was performed either with a suction capsule (n = 54) or by endoscopy (n = 125) and has previously been described in detail (29). In the 1997 cohort, all of the biopsies were performed using endoscopy (n = 99). Biopsies were taken both proximally and more distally in the duodenum, including the bulb, following a recommendation of 4 to 6 specimens per procedure. The mucosal specimens were classified according to the revised Marsh-Oberhuber classification (11). All of the biopsies were subjected to a second histopathological evaluation by a pathologist blinded to the previous results. In case of disagreement, a third pathologist evaluated the biopsy. In the first phase, this diagnostic procedure led to a changed diagnosis in 6 children, of whom5 additional patients with CD were found and 1 child was cleared of the CD diagnosis. In the second phase, there were no additional patients found, even though there were disagreements in the evaluation between the local and one of the expert pathologists. CD diagnosis was achieved through consensus of the 3 pathologists.
The criteria for CD diagnosis included elevated levels of celiac serological markers and small bowel enteropathy consistent with CD. Enteropathy was defined as an increased number of intraepithelial lymphocytes (>30 lymphocytes/100 enterocytes), crypt hyperplasia, and villous atrophy. In patients with Marsh 1 lesions (intraepithelial lymphocytes), symptoms and/or signs compatible with CD were required in combination with an HLA-DQ2/DQ8 haplotype and clinical response to a GFD (27,29).
In the present substudy, we evaluated the results of the serological markers at the time of biopsy. In the 2012 ESPGHAN guidelines, biopsy is no longer mandatory when the tTG-IgA levels exceed 10 times the upper limit of normal. The cutoff for positivity recommended by the manufacturer was >5 U/mL (30). Studies have indicated that a cutoff of 3 U/mL would be more correct, and therefore the results were additionally analyzed at a cutoff of 3 U/mL (31–33). Four children with tTG-IgA >30 U/mL in combination with Marsh 1 lesions and 1 child with tTG-IgA >50 U/mL who had Marsh 0 lesions were additionally tested for EMA.
Frequency tables, cross-tabulations, medians, and quartiles were used to present our data. Student t test was used to test whether there was a difference in the proportion of boys and girls with a high small intestinal damage. Statistical significance was defined as P < 0.05. Microsoft Access was used for data handling, and Stata 11.2 (StataCorp LP, College Station, TX) was used for statistical analyses and figures.
Level of tTG-IgA and the Degree of Enteropathy and CD Diagnosis
We found a wide variation in tTG-IgA levels and the degree of enteropathy among the 267 children who were participating in the present study (Fig. 2). There was, however, a tendency toward a gradual increase in the tTG-IgA levels predicting mucosal damage (Table 1), and Marsh 3 lesions were seen in 56%, 53%, 86%, 87%, and 98% of the children with tTG-IgA levels <3, 3 to 5, 5 to 30, 30 to 50, and >50 U/mL, respectively. There was no significant difference in Marsh 3 lesions between boys and girls—85% versus 79%, respectively (P = 0.23).
There were 2 interesting subgroups of children when comparing the results from the serology and the biopsy. The first group consisted of children with low tTG-IgA levels (<5 U/mL) of whom 55% (37/67) had Marsh 3 lesions and 6% (4/67) had Marsh 1 lesions.
The second group consisted of children with high tTG-IgA levels (>50 U/mL). There were 64 children in this group and the majority of them had Marsh 3 lesions, but there was 1 child with a biopsy graded as Marsh 1 who was EMA positive. In the group of children with tTG-IgA levels 30 to 50 U/mL, there was 1 child with Marsh 0 lesions, who consequently was not given a CD diagnosis, and 3 children who had Marsh 1 lesions. All of the children in this group were EMA positive. All of the patients with CD were HLA DQ2 and/or DQ8 positive. All of the children with Marsh 1 lesions and high tTG-IgA levels (>30 U/mL) were considered to have CD based on additional criteria according to the study design.
There were 34 patients who had normal biopsies, 26 children (76%) had tTG-IgA levels <5 U/mL, 7 children (21%) had tTG-IgA levels between 5 and 30 U/mL, and 1 child had a tTG-IgA level of 44 U/mL. There were 3 children who had Marsh 1 lesions but were not considered to be patients with CD. All of them had tTG-IgA levels between 5 and 30 U/mL.
ESPGHAN CD Diagnostic Criteria for Screening-Detected Patients
We assessed the 2012 ESPGHAN diagnostic criteria based on the serological levels of tTg-IgA exceeding 10 times the upper levels of normal. The ESPGHAN diagnostic criteria for the serology would have given 64 CD diagnoses of whom 63 (98%) were graded with Marsh 3 lesions for a cutoff of 5 U/mL and 101 CD diagnoses of whom 96 (95%) were graded with Marsh 3 lesions for a cutoff of 3 U/mL. One child would not have been considered to have CD in our study for the cutoff of 3 U/mL, whereas all of the children were considered to have CD for the higher cutoff of 5 U/mL.
To the best of our knowledge, the present study is the first one to investigate any potential association between the levels of tTG-IgA and enteropathy in a screening for CD in a general population of children. We found that levels of tTG-IgA >50 U/mL predicted gluten-induced enteropathy, but we also found an uncertainty to predict enteropathy in lower levels of tTG-IgA.
Several studies have found a strong correlation between tTG-IgA >100 U/mL and Marsh 3 lesions (21,24–26). These studies, however, were performed in children who had symptoms and signs suggestive of CD in a clinical setting. They also reported a poor correlation between the level of tTG-IgA and the severity of enteropathy.
In our study, all of the children who had levels exceeding 10 times the upper level of normal values at the recommended cutoff from the manufacturer (>5 U/mL), that is, 50 U/mL, were diagnosed as having CD. Recent studies using the same serological test in a clinical setting, however, recommended a cutoff of >3 U/mL. Among those with tTG >30 U/mL, 1 child did not have CD. Interestingly, Marsh 3 lesions were found more frequently in the group with tTG-IgA >50 U/mL (98%) than in the group with tTG-IgA between 30 and 50 U/mL (87%), which could be a reason for considering a higher cutoff level. The serology for CD has a high specificity and a high negative predictive value, which is useful when ruling out CD (34). Another reason for choosing a higher cutoff level is that increased values of tTG-IgA are not pathognomonic for CD. The level of tTG-IgA, however, is often lower in other conditions, whereas higher levels are more indicative of CD.
The blood test was obtained at school, and the families were later informed that their child had a suspicion of CD. The gluten load may have decreased compared with that in symptomatic patients in whom the motivation of continuing with gluten-containing food may be higher, to find the cause of the symptoms. Obtaining CD diagnosis without biopsy confirmation may affect the adherence, and being screening detected can also be a factor that affects the adherence (35,36). Our findings support the view that biopsy is necessary in patients with lower levels of tTG-IgA to obtain a correct diagnosis.
One of the strengths of our study is that a prospective study design with a similar protocol for the diagnostic process for all of the screened children was used and that all of the children were the same age. Furthermore, all of the serum samples were taken in connection with the biopsy and analyzed using the same serological test by the same laboratory (Phadia GmbH). All of the biopsies were evaluated by 1 experienced pathologist and then reevaluated blinded to the previous results by expert pathologists, thus minimizing interobserver variability. All of the patients with CD were DQ2 and/or DQ8 positive. The exceptional patients with CD, 1 patient with tTG-IgA >50 U/mL and Marsh 1 lesions and 4 patients with tTG-IgA >30 U/mL and Marsh 1 lesions, were EMA positive.
Although there are many pitfalls in the diagnosis of CD, there are some plausible explanations for the inconsistent association between the level of serological markers and intestinal mucosal damage. Every step in the diagnostic process has an impact on the diagnostic outcome. The histopathological evaluation is dependent on different variables, for example, biopsies from several different locations, biopsy method, and the evaluation by an experienced pathologist (29). The biopsy procedure in the present study has been thoroughly validated (29), and we found that multiple biopsies taken from both the proximal and the distal duodenum sharpened the diagnosis; all of the biopsies in the present study were reevaluated by the same experienced pathologist.
An interesting result was the subgroup with severe enteropathy in combination with low levels of tTG-IgA, which supports the results of other studies and underlines the necessity of performing a biopsy to obtain a correct diagnosis (16). It is somewhat easier to explain why some children with high levels of tTG-IgA had only mild enteropathy because it is known that the lesions can be patchily distributed in children that can lead to severe lesions being missed by the biopsy (37). Another interesting finding was that the correlation followed the same pattern regardless of sex, suggesting that boys and girls have a similar immunological response in the intestinal mucosa even though CD is more common in girls than in boys.
One child had a normal biopsy (Marsh 0) in combination with a relatively high tTG-IgA value of 44 U/mL at the time of biopsy. In the blood sample taken at the school, the tTG-IgA value was 33 U/mL. The mucosal specimen was evaluated to be of good quality and classified as Marsh 0, by both the local and the expert pathologist. The child did not obtain a CD diagnosis and therefore continued on gluten-containing food. The child, however, had a genetic susceptibility for CD and family members with CD. The serum samples were studied annually and showed a gradual decline. After 3 years, the tTG-IgA levels were normalized. A possible explanation to the initial high levels may be that the serological markers were transient or that the gluten load unintentionally decreased over time. The child may have potential CD.
Applying the 2012 ESPGHAN guidelines, when tTG-IgA levels exceed 10 times the upper limit of normal and biopsy can be omitted, seems reasonable even in screening-detected patients without symptoms suggestive of CD. The guidelines, which suggest that all of the screening-detected patients should be biopsied, may, therefore, be questioned. Notably, for the child with tTG-IgA 83 U/mL and Marsh 1 lesions, the guidelines would not have been applicable. The local pathologist, however, reported that the mucosal specimen was of poor quality and therefore the preparation and histopathological evaluation was suboptimal in this patient. This child also had positive EMA. The patient was subjected to a 3-month gluten challenge that resulted in increased levels of tTG-IgA (120 U/mL), followed by normalization after treatment with GFD. A possible explanation in this case may be a primary false-negative biopsy because of a patchy distribution and the specimen not representative of the true mucosal damage.
Approximately 90% of all of the children with type 1 diabetes mellitus have the same high-risk HLA haplotype DQ2 and/or DQ8, but only 10% of these children develop CD (38–40). Thus, this group of children is subjected to frequent CD screening, and omitting the small intestinal biopsy could be beneficial for the quality of life of these children. Popp et al showed that high tTG-IgA levels correlated with Marsh 3 lesions in nonsymptomatic children with type 1 diabetes mellitus (41). This observation of screening high-risk individuals is in line with our findings in the general population and suggests that the revised criteria for CD diagnosis not only are applicable in symptomatic patients but may also be of use in screening-detected patients. Also, CD has many different clinical appearances, and in another substudy of this particular screened population we found that symptoms associated with CD did not predict actual CD in a screening situation in the general population (42).
To conclude, our results indicate that high levels of tTG-IgA (>50 U/mL) are related to enteropathy in screening-detected patients and that biopsies could be omitted even in this group. Further studies are needed to investigate the need for biopsy confirmation and whether the ESPGHAN guidelines concerning CD diagnosis in screening-detected patients need to be revised.
We thank all of the participating children and their families and Phadia GmbH, Germany, for analyzing all of the blood samples.
1. Lebenthal E, Branski D. Celiac disease
: an emerging global problem. J Pediatr Gastroenterol Nutr
2. Green PH, Cellier C. Celiac disease
. N Engl J Med
3. Fasano A. Clinical presentation of celiac disease
in the pediatric population. Gastroenterology
4. Ludvigsson JF, Ansved P, Falth-Magnusson K, et al. Symptoms and signs have changed in Swedish children with coeliac disease. J Pediatr Gastroenterol Nutr
5. Collin P, Kaukinen K, Vogelsang H, et al. Antiendomysial and antihuman recombinant tissue transglutaminase antibodies in the diagnosis
of coeliac disease: a biopsy-proven European multicentre study. Eur J Gastroenterol Hepatol
6. Reilly NR, Fasano A, Green PH. Presentation of celiac disease
. Gastrointest Endosc Clin N Am
7. Uenishi RH, Gandolfi L, Almeida LM, et al. Screening
for celiac disease
in 1st degree relatives: a 10-year follow-up study. BMC Gastroenterol
8. Valitutti F, Barbato M, Cucchiara S. Screening celiac disease
in at-risk groups: effect of diagnosis
on perceived health of children and their families. J Pediatr Gastroenterol Nutr
9. Di Sabatino A, Corazza GR. Coeliac disease. Lancet
10. Sollid LM. Coeliac disease: dissecting a complex inflammatory disorder. Nat Rev Immunol
11. Dickson BC, Streutker CJ, Chetty R. Coeliac disease: an update for pathologists. J Clin Pathol
12. Marsh MN. Grains of truth: evolutionary changes in small intestinal mucosa in response to environmental antigen challenge. Gut
13. Bonamico M, Thanasi E, Mariani P, et al. Duodenal bulb biopsies in celiac disease
: a multicenter study. J Pediatr Gastroenterol Nutr
14. Rashid M, MacDonald A. Importance of duodenal bulb biopsies in children for diagnosis
of celiac disease
in clinical practice. BMC Gastroenterol
15. Husby S, Koletzko S, Korponay-Szabo IR, et al. European Society for Pediatric Gastroenterology, Hepatology, and Nutrition guidelines for the diagnosis
of coeliac disease. J Pediatr Gastroenterol Nutr
16. 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
17. Villalta D, Bizzaro N, Tonutti E, et al. IgG anti-transglutaminase autoantibodies in systemic lupus erythematosus and Sjogren syndrome. Clin Chem
18. Ferrara F, Quaglia S, Caputo I, et al. Anti-transglutaminase antibodies in non-coeliac children suffering from infectious diseases. Clin Exp Immunol
19. Tursi A, Brandimarte G, Giorgetti GM. Prevalence of antitissue transglutaminase antibodies in different degrees of intestinal damage in celiac disease
. J Clin Gastroenterol
20. Rostami K, Kerckhaert J, Tiemessen R, et al. Sensitivity of antiendomysium and antigliadin antibodies in untreated celiac disease
: disappointing in clinical practice. Am J Gastroenterol
21. 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
22. Liu E, Bao F, Barriga K, et al. Fluctuating transglutaminase autoantibodies are related to histologic features of celiac disease
. Clin Gastroenterol Hepatol
23. Fabiani E, Catassi C. International Working GroupThe 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-tTG. Eur J Gastroenterol Hepatol
24. Mubarak A, Wolters VM, Gmelig-Meyling FH, et al. Tissue transglutaminase levels above 100 U/mL and celiac disease
: a prospective study. World J Gastroenterol
25. Kotze LM, Utiyama SR, Nisihara RM, et al. IgA class anti-endomysial and anti-tissue transglutaminase antibodies in relation to duodenal mucosa changes in coeliac disease. Pathology
26. Alessio MG, Tonutti E, Brusca I, et al. Correlation between IgA tissue transglutaminase antibody ratio and histological finding in celiac disease
. J Pediatr Gastroenterol Nutr
27. Myleus A, Ivarsson A, Webb C, et al. Celiac disease
revealed in 3% of Swedish 12-year-olds born during an epidemic. J Pediatr Gastroenterol Nutr
28. Ivarsson A, Myléus A, Norström F, et al. Prevalence of childhood celiac disease
and changes in infant feeding. Pediatrics
29. Webb C, Halvarsson B, Norstrom F, et al. Accuracy in celiac disease
diagnostics by controlling the small-bowel biopsy process. J Pediatr Gastroenterol Nutr
30. Burgin-Wolff A, Dahlbom I, Hadziselimovic F, et al. Antibodies against human tissue transglutaminase and endomysium in diagnosing and monitoring coeliac disease. Scand J Gastroenterol
31. Hill PG, Holmes GK. Coeliac disease: a biopsy is not always necessary for diagnosis
. Aliment Pharmacol Ther
32. Dahlbom I, Korponay-Szabo IR, Kovacs 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
33. 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
34. 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
35. Bardella MT, Molteni N, Prampolini L, et al. Need for follow up in coeliac disease. Arch Dis Child
36. Mayer M, Greco L, Troncone R, et al. Compliance of adolescents with coeliac disease with a gluten free diet. Gut
37. Bonamico M, Mariani P, Thanasi E, et al. Patchy villous atrophy of the duodenum in childhood celiac disease
. J Pediatr Gastroenterol Nutr
38. Gutierrez-Achury J, Coutinho de Almeida R, Wijmenga C. Shared genetics in coeliac disease and other immune-mediated diseases. J Intern Med
39. Bybrant MC, Ortqvist E, Lantz S, et al. High prevalence of celiac disease
in Swedish children and adolescents with type 1 diabetes and the relation to the Swedish epidemic of celiac disease
: a cohort study. Scand J Gastroenterol
40. Larsson K, Carlsson A, Cederwall E, et al. Annual screening
detects celiac disease
in children with type 1 diabetes. Pediatr Diabetes
41. Popp A, Mihu M, Munteanu M, et al. Prospective antibody case finding of coeliac disease in type-1 diabetes children: need of biopsy revisited. Acta Paediatr
42. Rosen A, Sandstrom O, Carlsson A, et al. Usefulness of symptoms to screen for celiac disease