Coeliac disease (CD) is a multifactorial disease of the small intestine. Gluten-induced enteropathy may manifest at all ages, with different severity of the enteral symptoms and diverse extraintestinal manifestations. Some patients present as young children with classical CD and severe malabsorption leading to diarrhoea, abdominal distension, growth failure, and general malaise (1). Others develop only mild enteral symptoms or remain asymptomatic for a long time despite having a severe small intestinal lesion (2). Furthermore, it is not known why adults can become symptomatic after decades of silent disease, but infections, changes in gluten intake, or higher nutrient demands such as pregnancy have been implicated (3,4).
At present, the CD diagnosis relies on the histological examination of 1 to 3 biopsies from the upper part of the small intestine, showing at least a Marsh III lesion (5–7). However, recent reports indicate that the mucosal deterioration of patients with gluten sensitivity can be highly variable, and minor changes such as increased numbers of intraepithelial lymphocytes are frequently observed (8). Most patients with CD display serum antibodies against endomysium (EMA) or the enzyme tissue transglutaminase (9,10), and these antibodies can appear in the circulation before villous alterations develop (11). Immunoglobulin (Ig) A antibodies against tissue transglutaminase (IgA-TGA) measured with a human recombinant antigen are a good indicator of active CD, whereas the significance of IgG-TGA antibodies is less clear, except in subjects with IgA deficiency (12,13).
Dermatitis herpetiformis (DH) can be classified as one of the extraintestinal manifestations of CD because these patients only infrequently have abdominal complaints (14). The structural changes of the small bowel villous structure in DH are similar to those seen in CD, but the spectrum is wider and villi can be preserved in a portion of the patients (15).
The cause of the various forms of clinical manifestation in CD is poorly understood. Neither the degree of villous atrophy nor the autoantibody production has been found to correlate with the severity of the clinical symptoms (16), although patients with subtotal or total villous atrophy more often display EMA than those with milder lesions. However, it has been reported that patients with a flat lesion may lack CD-related antibodies in serum (17).
Assessments of the biopsy findings are subjective, dependent on the quality and orientation of the tissue specimens used for evaluation and are sometimes difficult to interpret (18). Therefore, quantitative measurements of autoantibodies may, for at least some patients, be a more accurate indicator of the mucosal destruction.
The aim of this study was to evaluate the potential of quantitative detection of IgA-TGA and IgG-TGA antibodies in serum for the prediction of the mucosal condition and disease severity in IgA-competent patients with gluten-induced enteropathy.
PATIENTS AND METHODS
Included in this study were serum samples from 134 patients with untreated DH (median age 9.5 years: range 2.4–68.8, 17 adults), consecutively diagnosed in the years 1989 to 2001 at the Heim Pál Children's Hospital by skin immunofluorescent study showing granular IgA deposition in the dermal papillae. They also underwent a small-intestinal biopsy, although only 9.2% of them had enteral symptoms.
During the same time period as the patients with DH, 921 untreated patients with CD without skin manifestations were diagnosed according to the revised European Society for Paediatric Gastroenterology, Hepatology, and Nutrition criteria (5). Patients with clinical symptoms suggesting CD underwent the biopsy procedure and 1 to 3 biopsy samples were taken from the duodenum by upper endoscopy or Watson capsule. Among these patients, 98.3% were EMA positive. For comparison, the patients were randomly selected among patients with 3 typical clinical manifestations of the disease: early childhood onset and/or severe malabsorption (group I, n = 52, median age 1.6 years, range 0.6–13.4), late childhood onset and/or moderate symptoms (group II, n = 59, median age 8.1 years, range 2.6–16), and adult onset with gastrointestinal symptoms (group III, n = 59, median age 39.5 years, range 21.8–77.8). In children, priority was given to the severity of symptoms over age for the classification and all of the children with severe malabsorption were included in group I. Severe malabsorption in children was defined by the presence of at least 3 of the following criteria: diarrhoea (>3 loose stools/day), weight loss (any in children), anaemia (haemoglobin <10 g/L or below 7 mmol/L), or low serum total proteins (<60 g/L). During a gluten-free diet, follow-up samples from 10 children in group I (43 samples) and 10 children in group II (32 samples) were collected for 1 to 27 months.
Also included were 131 noncoeliac subjects (median age 10.8 years, range 0.8–58.2) having various gastrointestinal disorders without villous atrophy (Table 1). These patients had a serum sample submitted to the coeliac laboratory of Heim Pál Children's Hospital, Budapest, Hungary, for antibody testing, regardless of the decision for a small-intestinal biopsy.
All of the patients had total serum IgA values within the normal range for their age, as determined with routine methods. Biopsies and EMA measurements were performed as part of the clinical evaluation of the subjects and written consent was obtained for the invasive procedures. Investigation of stored serum samples for TGA antibodies was authorised by the ethical committee of the Heim Pál Children's Hospital, Budapest.
Measurement of TGA Autoantibodies
All patient sera were stored at −40°C until analysed, and TGA antibodies were determined with human recombinant tissue transglutaminase. IgA-TGA were measured with the Celikey assay (Phadia GmbH, Freiburg, Germany) according to the manufacturer's instructions, and IgG-TGA levels were measured with a research ELISA (Phadia), as previously described (19). Antibody levels were calculated in units per millilitre using a standard curve containing 0, 3, 7, 16, 40, and 100 U/mL of TGA antibodies. Patient samples yielding a value above the highest calibrator were further tested in appropriate dilution to obtain a value within the linear range of the calibrator curve. The within- and between-assay variability was 3.8% and 6.4%, respectively, for IgA-TGA and 4.4% and 5.9%, respectively, for IgG-TGA. The cutoff value for IgA-TGA as determined by receiver operating curves (ROC) was 3 U/mL, with an area under the curve of 0.994 (95% CI, 0.988–1, P < 0.0001). The cutoff value for IgG-TGA, as determined by ROC, was 2 U/mL with an area under the curve of 0.96 (95% CI, 0.947–0.979, P < 0.0001). However, 3 U/mL was chosen as a cutoff for both IgA-TGA and IgG-TGA.
Measurements of EMA
IgA EMA and IgG EMA were assessed by indirect immunofluorescent method on unfixed cryostat sections of monkey oesophagus (from Cercopithecus aethiops), human jejunum, and human appendix, as described elsewhere (20). Patient sera were diluted 1:2.5 and positive results were quantified further by titration. Optimal serum sample and conjugate dilutions were previously established by chessboard titration to ensure high sensitivity.
Evaluation of Histopathology Changes
Small intestine histology changes were graded 0 to 5 (Table 2) according to the description of Fontaine and Navarro (21), which was officially used as consensus by Hungarian pathologists at the time of the study. This classification allows a detailed grading of partial villous atrophy; grade 2 or more severe lesions were considered to be compatible with CD. Correspondent stages according to the Marsh classification (22) are shown in Table 2. A subset of 100 randomly selected biopsy samples was re-evaluated also by the Marsh classification, which showed that all of the samples with grade 3 to 5 lesions were Marsh IIIb or IIIc and samples with grade 2 lesions were at least Marsh IIIa.
Antibody levels are reported as median (5th and 95th percentiles) units per millilitre unless otherwise stated. ROC analyses for estimation of cutoff levels for the TGA antibodies were carried out with both CD and DH patients regarded as 1 disease group, and the disease control subjects as a nondisease group. The Kruskal-Wallis test with Dunn's multiple posttest was performed to compare antibody levels between the groups of untreated patients subgrouped according to symptom presentation and according to biopsy findings. Multiple regression analysis, adjusted for age, sex, and biopsy findings, was used to compare autoantibody levels according to symptoms. Correlations were calculated using the Spearman rank correlation coefficient (95% confidence interval in brackets). Kaplan-Meyer survival curves were used to analyse the time for normalisation of antibodies in patients during a gluten-free diet, using the log-rank method to compare survival curves. A P value < 0.05 was considered significant.
TGA Autoantibodies Correlate With the Clinical Presentation and Age
All 170 patients with CD (100%), 127 of 134 (95%) of the patients with DH, and 1 of 131 (0.7%) of the control subjects had IgA-TGA levels higher than the cutoff of 3 U/mL. The median IgA-TGA level was 481 (65–5890) U/mL for children with an early onset and severe malabsorption (group I) and 148 (27–1130) U/mL for children with a late onset or milder symptoms (group II). The median IgA-tTG level for adult patients (group III) was 185 (15–2790) U/mL and 46 (3–590) U/mL for the patients with DH. Differences in the median IgA-TGA levels between the patient groups are shown in Figure 1A. The control subjects had a lower median IgA-TGA level than the other patient groups (P < 0.0001) and group I had a higher IgA-TGA level than all other groups, as analysed with multiple regression adjusted for age, sex, and biopsy findings.
IgG-TGA antibodies above the cutoff were present in 143 of 170 (84%) patients with CD, in whom positive results were observed in 52 of 52 (100%) cases in group I, 48 of 59 (81%) cases in group II, and 43 of 59 (73%) cases in group III. Among the patients with DH, 90 of 134 (67%) had positive levels of IgG-TGA antibodies and 2 of 131 control subjects (1.5%) were positive. The children in group I had a higher median IgG-TGA level, 30 (13–1550) U/mL, than all other groups (P < 0.01), and the children in group II had a higher median antibody level, 18 (1–310) U/mL, than patients in group III, 6 (1–54) U/mL and patients with DH, 4 (2–130) U/mL (P < 0.01) (Fig. 1B). The control patients had a lower median IgG-TGA level than all of the other patient groups (P < 0.0001). Furthermore, group I had a higher IgG-TGA level than all of the other groups as analysed with multiple regression adjusted for age, sex, and biopsy findings.
There was a positive correlation between the IgA-TGA levels and IgA EMA titres (r = 0.915, 0.892–0.929), with a 99.5% agreement between positive and negative results. Moreover, IgG EMA was investigated in 65 patients with CD and 100 patients with DH, revealing a positive correlation between the IgG-TGA levels and IgG EMA titres (r = 0.827, 0.770–0.871), with 85% agreement between positive and negative results. The correlation between the levels of IgA-TGA and IgG-TGA for the entire study population was 0.778 (0.737–0.784).
In 14 of 170 CD patients, the IgG-TGA levels were more than 1.5 times higher than IgA-TGA; 8 of them belonged to group I and the remaining 6 belonged to group II. Among the IgA-TGA–positive patients with DH, 4 of 134 had more than 1.5 times higher levels of IgG-TGA than IgA-TGA. A weak inverse correlation between age and IgA-TGA was observed (r = −0.2413, −0.3819 to −0.08968, P < 0.01) and the IgG-TGA levels correlated negatively with age (r = −0.4958, −0.6043 to −0.3692, P < 0.0001).
TGA Antibodies Correlate With the Condition of the Small Intestinal Mucosa
Both the IgA-TGA and IgG-TGA levels correlated with the grade of villous atrophy within the CD and DH groups (P < 0.0001) (Fig. 2A and B). All of the patients with at least a grade 2 villous atrophy (Marsh III, villous height/crypt ratio of 1 or less) had IgA-TGA levels above 3 U/mL and 79% of them also had elevated levels of IgG-TGA.
There was no difference between the median IgA-TGA levels for patients with CD and patients with DH with a grade 3 or 5 villous atrophy. However, for patients with a grade 4 villous atrophy, the IgA-TGA levels were lower in the group of patients with DH (P < 0.001) (data not shown).
IgA-TGA and IgG-TGA in Combination Related to Biopsy Findings and Clinical Symptoms
Because the levels of IgA-TGA and IgG-TGA each correlated with the grade of mucosal villous atrophy, we investigated whether high levels of these antibodies in combination could predict the extent of mucosal lesions. The study population was, therefore, divided according to the presence of at least a grade 3 villous atrophy (257 patients) or milder mucosal alterations (175 patients). The levels for a combined cutoff were estimated to be 30 U/mL for IgA-TGA and 15 U/mL for IgG-TGA, because no higher levels were found in patients with a normal or slightly altered mucosa. Using the combined cutoff, 100 of 257 (39%) of the patients with at least a grade 3 lesion and 1 (0.6%) patient with milder lesions displayed higher antibody levels (Fig. 3), indicating that the presence of at least a grade 3 lesion could be predicted in untreated gluten-sensitive patients with 99% specificity. The same cutoff values predicted severe villous atrophy (Marsh IIIb or IIIc) also in the subset of 100 samples where antibody levels were directly compared with histology results according to the Marsh classification (Table 3).
Among the subgroups of patients with CD, 38 of 52 (73%) children in group I, 31 of 59 (53%) children in group II, and 19 of 59 (32%) adult patients in group III had antibody levels exceeding the proposed combined cutoff for predicting a grade 3 lesion. Additionally, 13 of 131 (10%) patients with DH were positive. The proportion of positive children with CD in group I was higher as compared with the other groups (P < 0.05), and the proportion of positive patients with DH was lower (P < 0.05). Additionally, the proportion of positive subjects in group II was higher than that observed in group III (P < 0.05).
Isotype-dependent Decrease of TGA Levels in Children During a Gluten-free Diet
The dynamics of IgA-TGA and IgA EMA were not significantly different, whereas IgG-TGA declined slower than IgA-TGA (P < 0.05) for the majority of children in group I (Fig. 4A). All of these children reverted to normal IgA-TGA levels and all but 1 child attained normal IgG-TGA levels on a gluten-free diet. The median time for normalisation of the IgA-TGA levels was 6 months (range 2.3–16.1 months) and IgA EMA decreased at a similar rate (median 6.2, range 2.2–16.1 months), whereas the normalisation time for IgG-TGA was 13.6 months (range 2.3–17.1 months). Among the children in group I, the initial levels of IgA-TGA correlated with the normalisation time (r = 0.766, P < 0.01), whereas no such correlation was observed for IgG-TGA.
A reduction in the IgA-TGA and IgG-TGA levels was also observed in the children with less severe symptoms during the gluten-free period (Fig. 4B). However, IgA-TGA levels were normalised in 7 of 10 children. The remaining 3 had elevated IgA-TGA, and IgG-TGA remained elevated in 2 of them at the end of the studied gluten-free period. The median duration for normalisation of IgA-TGA and IgG-TGA for the entire group was 13.6 months (range 7.2–18 months) and 14.4 months (range 7.2–27.1 months), respectively. There was no difference in the normalisation period for IgA-TGA compared with IgG-TGA in children with moderate forms of CD, and the initial levels of IgA-TGA or IgG-TGA did not correlate with the time required for normalisation.
Our study population comprised patients who were stratified into subgroups according to their clinical presentation of CD, spanning from the early classic manifestations of CD to patients with extraintestinal symptoms, represented by a large number of patients with DH. We showed that serum levels of IgA-TGA and IgG-TGA correlate with the severity of the gluten-induced enteropathy and that the antibody isotypes evinced separate dynamics in clinically different patient groups.
Furthermore, we found that the serum concentrations of IgA-TGA as well as IgG-TGA were greatly increased in patients with CD with the most severe enteropathy, whereas the levels were considerably lower and less frequent in the patients with DH. Even if the severe malabsorptive mode of CD was associated with the presence of IgG-TGA, the highest levels were observed among the youngest children, irrespective of clinical presentation. An inverse correlation between IgG1-TGA and age has previously been observed (23). However, age alone did not account for high or low antibody levels in our study because no such relation was observed for the adult patients with CD and the entire DH group.
We also found that IgA-TGA and IgG-TGA strongly corresponded to the condition of the small intestinal mucosa in subjects with CD and subjects with DH. This finding is in agreement with earlier studies reporting that the prevalence and levels of IgA-TGA differ between patients with DH with severe and minor mucosal lesions (24). Our large group of patients with DH was particularly suitable for such analysis because their diagnoses of gluten sensitivity were confirmed by a skin biopsy, independent of the histological findings in the small bowel. As 32% of the patients with DH had minor mucosal abnormalities, milder forms of small-intestinal lesions could also be included with accuracy, which is more difficult with CD case series. Nonetheless, a difference in the prevalence of IgA-TGA antibodies has already been noted when comparing patients with CD with severe lesions (Marsh IIIb and IIIc) and mild alterations (Marsh I and II) (25).
The presence of IgG-TGA was comparable in the adult patients with CD and the DH group, although several of them lacked elevated levels of IgG-TGA. Because the vast majority of these patients had higher IgA-TGA levels than IgG-TGA, the determination of IgA-TGA seems to be more suitable in the clinical evaluation and screening of these patient groups. Conversely, some of the untreated children with CD had even higher levels of IgG-TGA than IgA-TGA. Furthermore, in children with CD the IgA-TGA antibodies disappeared rapidly after the introduction of a gluten-free diet, whereas the clinical recovery and recession of IgG-TGA antibodies were much slower. Most of the untreated children with CD with the severe mode of CD displayed a particularly rapid decrease in IgA-tTG as compared with the IgG-tTG antibodies, and the difference between the normalisation rates could partly be dependent on the longer half-life of IgG in serum. Taken together, our results indicate that measuring both IgG-TGA and IgA-TGA, at least in children, may increase the possibility of detecting untreated CD cases.
Traditionally, CD diagnosis depends on the finding of characteristic histological abnormalities in the small intestinal mucosa. However, recent reports have suggested that serum IgA-TGA above a certain level could replace biopsy for diagnosing CD, because a highly elevated IgA-TGA level was associated with a Marsh III histopathology (26–28). The results of the present study support the notion that patients with high IgA-TGA levels (above 30 U/mL) had at least a Marsh III lesion. Although IgA-TGA antibodies are highly specific and predictive for gluten-induced diseases (29), a single positive serology result alone is currently unacceptable as proof of a CD diagnosis (7). False seropositivity may occur for various technical reasons, but even if the serological analysis is correct, IgA-TGA levels may fluctuate over time between undetectable and occasionally extremely high levels (30). The natural history of CD development in such instances is still enigmatic, and there is still a lack of reliable evidence that all seropositive cases need treatment. Thus, the detection of IgA-TGA alone cannot replace biopsy and the final diagnosis of CD can currently only be established in patients evincing villous atrophy of the small intestine.
Here we provide evidence that the presence of both IgA-TGA and IgG-TGA above certain combined cutoff levels reliably predicts the presence of severe villous atrophy and that this noninvasive procedure may alleviate the need for a routine biopsy in some cases. Moreover, a recent study measuring only IgA-TGA with the same commercial test reported a correlation between antibody levels exceeding 10 times the cutoff and histological alterations compatible with CD (28). However, these authors accepted mild lesions such as crypt hyperplasia without villous atrophy in the diagnosis, even if this finding is not always specific for CD (5,11,18).
The chosen cutoff values predicting severe villous atrophy in our study may differ from other commercial tests because the measured antibody levels in all of the tests are arbitrarily calculated. Of note is that the IgG-TGA results reported here were obtained with a research ELISA. However, reanalysing some of the included samples with the commercially available assay did not indicate a need for further adjustment of the combined cutoff level regarding IgG-TGA (data not shown). With the chosen cutoff, our present data clearly show that IgG-TGA is also important for an accurate prediction of mucosal atrophy, but because the numbers of patients with slight mucosal alterations included in the study were in the minority, further studies are needed. Interestingly, despite the low cutoff levels only 2 control children had elevated IgG-TGA levels, supporting the findings that IgG-TGA only occasionally are elevated among subjects with other diseases (31).
Furthermore, it is not clear whether the IgG-TGA antibodies contribute to the CD progression or whether they originate from the intestine. It has previously been reported that the length of the inflamed portion of the small intestine may influence the clinical presentation of CD symptoms (32), and it is possible that the levels of IgG-TGA and IgA-TGA depend on how much of the small intestine is affected. Although further studies are needed to explore the clinical implication of IgG-TGA production, the detection of both IgA-TGA and IgG-TGA provides an objective tool to classify patients with CD according to their antibody levels.
In summary, even if IgA-TGA and IgG-TGA are highly specific for CD, a small-intestinal biopsy is still the gold standard in diagnosing CD. The presence of increased IgA-TGA or IgG-TGA in serum alone should be interpreted with caution, and neither of these markers alone can as yet replace the biopsy in diagnosing CD. However, our results demonstrate that a combination of high serum levels of both IgA-TGA and IgG-TGA correlates with a more severe mucosal villous atrophy. Hence, at least for children with high levels of both IgA and IgG antibodies against TGA, the biopsy procedure may not be needed to establish the diagnosis of CD. Further studies are needed to elucidate the possibility to reduce the need for a biopsy in adult patients with suspected CD.
The authors thank Éva Török, Margit Lörincz, and Anikó Nagy for their work in the clinical care of the patients. The kits for tissue transglutaminase antibody testing were kindly provided by Phadia AB, Uppsala, Sweden.
1. Green PH, Cellier C. Celiac disease. N Engl J Med 2007; 357:1731–1743.
2. Maki M, Mustalahti K, Kokkonen J, et al. Prevalence of Celiac disease among children in Finland. N Engl J Med 2003; 348:2517–2524.
3. Carroccio A, Cavataio F, Montalto G, et al. Treatment of giardiasis reverses “active” coeliac disease to “latent” coeliac disease. Eur J Gastroenterol Hepatol 2001; 13:1101–1105.
4. Corrado F, Magazzu G, Sferlazzas C. Diagnosis of celiac disease in pregnancy and puerperium: think about it. Acta Obstet Gynecol Scand 2002; 81:180–181.
5. Walker-Smith J, Guandalini S, Schmitz J, et al. Revised criteria for diagnosis of coeliac disease. Report of Working Group of European Society of Paediatric Gastroenterology and Nutrition. Arch Dis Child 1990; 65:909–911.
6. United European Gastroenterology. When is a coeliac a coeliac? Report of a working group of the United European Gastroenterology Week in Amsterdam, 2001. Eur J Gastroenterol Hepatol 2001; 13:1123–1128.
7. 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.
8. Rostami K, Mulder CJ, Stapel S, et al. Autoantibodies and histogenesis of celiac disease. Rom J Gastroenterol 2003; 12:101–106.
9. 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 2002; 37:685–691.
10. Dieterich W, Laag E, Bruckner-Tuderman L, et al. Antibodies to tissue transglutaminase as serologic markers in patients with dermatitis herpetiformis. J Invest Dermatol 1999; 113:133–136.
11. Salmi TT, Collin P, Jarvinen O, et al. Immunoglobulin A autoantibodies against transglutaminase 2 in the small intestinal mucosa predict forthcoming coeliac disease. Aliment Pharmacol Ther 2006; 24:541–552.
12. Korponay-Szabo IR, Dahlbom I, Laurila K, et al. Elevation of IgG antibodies against tissue transglutaminase as a diagnostic tool for coeliac disease in selective IgA deficiency. Gut 2003; 52:1567–1571.
13. Bilbao JR, Vitoria JC, Ortiz L, et al. Immunoglobulin G autoantibodies against tissue-transglutaminase. A sensitive, cost-effective assay for the screening of celiac disease. Autoimmunity 2002; 35:255–259.
14. Reunala T, Kosnai I, Karpati S, et al. Dermatitis herpetiformis: jejunal findings and skin response to gluten free diet. Arch Dis Child 1984; 59:517–522.
15. O'Mahony S, Vestey JP, Ferguson A. Similarities in intestinal humoral immunity in dermatitis herpetiformis without enteropathy and in coeliac disease. Lancet 1990; 335:1487–1490.
16. Brar P, Kwon GY, Egbuna II, et al. Lack of correlation of degree of villous atrophy with severity of clinical presentation of coeliac disease. Dig Liver Dis 2007; 39:26–32.
17. Abrams JA, Diamond B, Rotterdam H, et al. Seronegative celiac disease: increased prevalence with lesser degrees of villous atrophy. Dig Dis Sci 2004; 49:546–550.
18. 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 2005; 17:85–91.
19. Dahlbom I, Olsson M, Forooz NK, et al. Immunoglobulin G (IgG) anti-tissue transglutaminase antibodies used as markers for IgA-deficient celiac disease patients. Clin Diagn Lab Immunol 2005; 12:254–258.
20. Sulkanen S, Halttunen T, Laurila K, et al. Tissue transglutaminase autoantibody enzyme-linked immunosorbent assay in detecting celiac disease. Gastroenterology 1998; 115:1322–1328.
21. Fontaine JL, Navarro J. Small intestinal biopsy in cows milk protein allergy in infancy. Arch Dis Child 1975; 50:357–362.
22. 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.
23. Agardh D, Borulf S, Lernmark A, et al. Tissue transglutaminase immunoglobulin isotypes in children with untreated and treated celiac disease. J Pediatr Gastroenterol Nutr 2003; 36:77–82.
24. Sugai E, Smecuol E, Niveloni S, et al. Celiac disease serology in dermatitis herpetiformis. Which is the best option for detecting gluten sensitivity? Acta Gastroenterol Latinoam 2006; 36:197–201.
25. Ravelli A, Bolognini S, Gambarotti M, et al. Variability of histologic lesions in relation to biopsy site in gluten-sensitive enteropathy. Am J Gastroenterol 2005; 100:177–185.
26. 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.
27. 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.
28. Hill PG, Holmes GK. Coeliac disease: a biopsy is not always necessary for diagnosis. Aliment Pharmacol Ther 2008; 27:572–577.
29. Rostom A, Dube C, Cranney A, et al. The diagnostic accuracy of serologic tests for celiac disease: a systematic review. Gastroenterology 2005; 128:S38–46.
30. 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.
31. Villalta D, Alessio MG, Tampoia M, et al. Testing for IgG class antibodies in celiac disease patients with selective IgA deficiency. A comparison of the diagnostic accuracy of 9 IgG anti-tissue transglutaminase, 1 IgG anti-gliadin and 1 IgG anti-deaminated gliadin peptide antibody assays. Clin Chim Acta 2007; 382:95–99.
32. Macdonald WC, Brandborg LL, Flick AL, et al. Studies of celiac sprue. IV. The response of the whole length of the small bowel to a gluten-free diet. Gastroenterology 1964; 47:573–589.
© 2010 Lippincott Williams & Wilkins, Inc.