Celiac disease is an autoimmune disorder in which dietary gluten induces an enteropathy that is characterized by chronic inflammation of the small intestinal mucosa with atrophy of intestinal villi (1,2). The symptoms vary from diarrhea with malabsorption and growth retardation to diffuse symptoms such as fatigue (3). In some individuals, the disease may be asymptomatic (4). Celiac disease is associated with a wide spectrum of complications such as anemia, osteoporosis, infertility, and autoimmune diseases related to celiac disease (5). The risk of complications is decreased if relying on gluten-free diet, which is the only available treatment. The development of serological markers indicative of celiac disease has facilitated active case finding and has made screening possible. However, the final diagnosis of celiac disease is still dependent on the identification of enteropathy in small-bowel biopsies (6).
For high accuracy in diagnosing celiac disease, several steps must be performed, from taking the biopsy to the histopathological report written by the pathologist. There are a number of studies in both children (7–10) and adults (11) indicating that the enteropathy may have a patchy distribution. Thus, the diagnosis will be missed if the biopsy happens to be taken only from an area with normal mucosa. The histological changes are graded using the Marsh classification (12).
According to the 1990 criteria by the European Society of Paediatric Gastroenterology, Hepatology, and Nutrition, villous atrophy of some degree (Marsh grade III a–c) is needed for celiac disease diagnosis (13). In recent years, clinical practice has changed, and the diagnosis is now often based on the following: clinical signs; elevated serological markers; and pathological changes in a small-bowel biopsy, including increased count of intraepithelial lymphocytes (IELs) without villous atrophy (Marsh grade I) (14). Several methods to facilitate detection of IELs have been developed, and in clinical practice, immunohistochemical staining for CD3+ lymphocytes is used widely (15). Previous studies have shown fairly good interobserver agreement between experienced pathologists, but how this correlates with clinical practice is not known (16).
Between 1984 and 1996, Sweden experienced an epidemic of celiac disease in children younger than 2 years of age. We have shown that half of the epidemic could be explained by changes in national recommendations that resulted in more infants being introduced abruptly to gluten without ongoing breast-feeding. When recommendations were again changed, the incidence returned to the preepidemic level (17,18). The epidemic has given us the opportunity to study the risk of developing celiac disease as a result of early infant-feeding practices. In the study ETICS—Exploring the Iceberg of Celiacs in Sweden—we are screening 2 birth cohorts that differ in infant feeding, that is, 12-year-olds born in 1993 during the epidemic (phase I) and in 1997 after the epidemic (phase II). Children of the 1993 birth cohort had a high prevalence of symptomatic celiac disease reported before the screening, 8.9 of 1000. However, additional screening-detected cases resulted in another 20 of 1000, giving the unexpectedly high total prevalence of 29 of 1000 (95% confidence interval [CI] 25–33) (19).
In the screening study, the confirmative diagnostic procedures with small-bowel biopsies were performed according to the clinical routine at each study site. To verify the accuracy of the celiac disease diagnosis, and thereby also the prevalence estimate, we conducted a follow-up study with a standardized evaluation of the biopsy specimens, performed by pathologists with expertise in gastrointestinal pathology together with a systematic review of the diagnostic procedure.
The aim of the present study was to evaluate the accuracy of the celiac disease diagnosis in the ETICS study phase I (12-year-olds born in 1993) and take advantage of lessons learned about each step of the small-bowel biopsy process for improving future diagnostic routines.
MATERIALS AND METHODS
Celiac Disease Screening Study
Five pediatric departments representing different geographical areas of Sweden collaborated in a screening study (ETICS) to evaluate the prevalence of celiac disease. The phase I screening was conducted from 2005 to 2006. It was school based with 10,041 12-year-old children invited; 7567 consented to participate. Of these, blood samples were collected from 7207 children.
The screening procedure was performed as described earlier. All of the serum samples were analyzed for anti-human tissue transglutaminase antibodies (tTG-IgA). If the level was intermediate, then endomysial antibodies-IgA was analyzed. All of the samples were also analyzed for serum-IgA, and samples <0.5 g/L were further analyzed for tTG-IgG, with intermediate values further analyzed for endomysial antibodies-IgG. Serological criteria for recommending a small-bowel biopsy are summarized in Table 1. This resulted in 192 children recommended for biopsy, of whom 184 accepted additional investigation.
Definition of Celiac Disease
Criteria for celiac disease diagnosis were elevated levels of serological markers (Table 1) and enteropathy of the small-bowel mucosa. Enteropathy was defined as villous atrophy, crypt hyperplasia, or increased count of IELs (>30 lymphocytes/100 enterocytes) (15,20,21). In children with elevated levels of serological markers who only had an increased count of IELs, additional requirements should be fulfilled for the celiac disease diagnosis. They should have symptoms and/or signs at presentation compatible with the disease and a clinical response to the gluten-free diet.
The small-bowel biopsy samples were taken either by endoscope or by suction capsule using a Storz or Watson capsule. Histopathological preparation and evaluation initially took place at the closest pathology laboratory according to local clinical practice. The mucosal specimens were stained with hematoxylin and eosin. At 1 study site CD3+ staining was performed as routine; otherwise, this was not routine. Diagnosis was made by the local pathologist according to recommendations from the Swedish Society of Pathology. This classification corresponds to the Marsh classification, with the difference that Marsh IIIb and IIIc are grouped together as 1 group named subtotal/total villous atrophy. Thereafter, an expert pathologist reevaluated all of the specimens blinded to the results of the local pathologists' assessment. Every fraction was classified according to Marsh and the quality was noted as good, suboptimal, or not interpretable. The overall histopathological diagnosis in each case was based on the highest grade of lesion. When there was diagnostic divergence between the local pathologist and the expert pathologist, specimens were further evaluated in the same blinded way, by a second expert pathologist, and diagnostic consensus was reached. Children with normal biopsy but persistent clinical suspicion of celiac disease were, if possible, rebiopsied using endoscopy. All of the children with persistent elevated serological markers and thereby considered to be potential cases with celiac disease were assessed with a clinical visit and reevaluation of serological markers 1 and 5 years, respectively, after the initial biopsy.
Microsoft Access was used for data handling. The material is primarily presented in a descriptive way. Fisher exact t test and Pearson χ2 test were performed using SPSS 16.0 for comparison between methods (SPSS Inc, Chicago, IL).
Effect of Biopsy Method
Small-bowel biopsies were performed in 184 children, of which 130 were by endoscopy and 54 by suction capsule. After the first biopsy and local routine diagnostics, 140 children were diagnosed as having celiac disease. Sixteen children with normal biopsies, 4 after endoscopy and 12 after suction capsule, were rebiopsied using endoscopy. Of these 146 endoscopic biopsies, 1 was diagnostically inconclusive (0.6%), as compared with 4 of 54 biopsies by suction capsule (7.4%; P < 0.0001, Fisher exact t test).
There were conclusive fractions from both proximal and distal duodenum available from 110 of the endoscopically performed investigations. Among these, 90 children had histopathological changes in both proximal and distal duodenum and 10 had normal mucosa in both locations. Ten children (9.1%) had enteropathy in 1 location but normal mucosa in the other location, so called patchy enteropathy. The results are shown in detail in Figure 1.
Effect of Immuohistochemical Staining
CD3+ staining was performed in 68 cases. Among these, 15% were diagnostically divergent between the local pathologist and expert pathologist, compared with 13% of non-CD3+–stained specimens. CD3+ staining did not increase the probability to diagnose Marsh I in biopsies without villous atrophy.
Effect of Reevaluation by an Expert Pathologist
In 177 of 184 children, the expert reevaluation verified the local diagnosis. However, in 6 children (3.3%) the diagnosis was changed after reevaluation and increased the total number of cases with celiac disease to 145. In 5 children, the diagnosis was changed from normal to celiac disease. One child was diagnosed as having celiac disease by the local pathologist and then cleared by the expert pathologist. In another child with celiac disease diagnosis after the local pathologist's examination (Marsh I), the material from the biopsy was considered not interpretable by the expert pathologist. Additional characteristics of this child are given in Table 2 (child no. 10). Rebiopsy failed and the family refused additional investigation; however, the child is still considered a case with celiac disease. The reason is that serological markers recovered on gluten-free diet and reoccurred on gluten challenge together with symptoms. This is 1 of the 4 cases with inconclusive material after capsule biopsy. A detailed description of the effect of the histopathological reevaluation is shown in Figure 2.
Effect of Rebiopsy
Rebiopsy was performed on 16 children. Four of them had already received the celiac disease diagnosis after reevaluation of the first biopsy by the expert pathologist. Endoscopic rebiopsy after primary suction capsule biopsy, the 4 above excluded, resulted in 8 new cases with celiac disease and a total of 153 cases with celiac disease. The 4 rebiopsies after a primary endoscopic examination did not result in any new case with celiac disease. The effect of rebiopsy is shown in Figure 2.
Effect of Diagnostic Follow-up on the Prevalence of CD
In a recent report from the ETICS study phase I we reported 145 children with newly diagnosed celiac disease, including 5 cases diagnosed after early rebiopsy (19). After further sharpening of the diagnostic procedure with the present follow-up, we found another 9 cases with celiac disease and cleared 1. Thus, the total number increased to 153. Initial histopathological diagnosis and diagnosis after reevaluation and rebiopsies are listed in Figure 2. Characteristics of those diagnosed as having celiac disease without having villous atrophy are given in Table 2. Serological markers were collected from 7207 children, resulting in a screening-detected prevalence of 21 of 1000 (153/7207). Adding the previously diagnosed cases (8.9/1000) results in a total prevalence of celiac disease of 30 of 1000 (CI 26–34).
There are many pitfalls in the diagnosis of celiac disease. We found that the enteropathy often is patchy, the diagnostic quality of endoscopic biopsies is better than that of suction capsule specimens, and the pathologist is a key player in obtaining an accurate diagnosis.
In the ETICS study we relied on local routines in the diagnostic process. This gives us the disadvantage of a heterogenic material, which is a weakness. To compensate for this, mucosal specimens from suction capsule biopsies that were evaluated as normal were followed by multiple endoscopic biopsies, and expert pathologists reevaluated the histopathological diagnosis. The present study design gave us the opportunity to study the effect of different diagnostic routines on the outcome.
Celiac disease is a lifelong disease associated with a wide range of complications. Even in cases with mild enteropathy the risk of early death is increased in comparison to the normal population (22). This is 1 of the reasons that we in the ETICS study followed the diagnostic trend to consider increased IELs in combination with positive serology, and symptoms or signs, to be enough to diagnose celiac disease. In a clinical situation, it is a challenge to find these children without overdiagnosing the condition. It should be noted that minimal lesions with negative serology seldom are celiac disease (23). By contrast, several studies have shown that high levels of tTG-IgA may indicate villous atrophy, and that a verifying small-bowel biopsy may not be needed in this subgroup of patients (24–28).
We found a higher percentage of patchy enteropathy, more often affecting the proximal duodenum, than previously described (7). An explanation could be that screening-detected cases may have an earlier or milder form of celiac disease than those identified due to symptoms. If we, in the present study, had relied only on local routine diagnostics, including suction capsule biopsies, we would have missed several cases. This urges us to recommend that multiple biopsy samples should be taken from multiple locations, including the proximal duodenum, to ensure that enteropathy is diagnosed also when present with a patchy distribution. In the current guidelines from the European Society of Paediatric Gastroenterology, Hepatology, and Nutrition, a jejunal biopsy by suction capsule is recommended (13). However, in most countries there is a trend toward endoscopic examinations, and there are many studies besides the present supporting this change in practice (29–33). At present no international gastroenterological association recommends duodenal bulb biopsies, but there is a growing awareness of the possibility of patchy enteropathy. In centers lacking the resources to perform a pediatric endoscopic examination, a suction capsule biopsy could be performed firsthand in children with suspected celiac disease. However, if the biopsy does not show enteropathy, then the child should be referred for an endoscopic examination.
We could not show that CD3+ staining sharpened the diagnostic accuracy. However, there are several other studies suggesting that CD3+ staining facilitates differentiation between lymphocytes and other cells. There are also indications that a top-heavy distribution pattern of CD3+ IELs is a sensitive feature to distinguish celiac disease (34). CD3+ staining is more expensive and time-consuming than hematoxylin and eosin staining. If villous atrophy is present, then no additional routine CD3+ staining is necessary but should be reserved for borderline cases in which it could be valuable (20).
The diagnosis of celiac disease is highly dependent on the investigating pathologist's experience and interest in small-bowel pathology. In many Swedish hospitals and also worldwide, small-bowel biopsy specimens are assessed by general pathologists. In our study, 3.3% were misdiagnosed by the local pathologists, and this illustrates that such a routine is inappropriate. Thus, the biopsy material needs to be evaluated by a pathologist with a special interest in gastrointestinal pathology and good knowledge of celiac disease.
This diagnostic follow-up supports our previously reported results from the ETICS study phase I that the birth cohort of 1993, born during the peak of the Swedish celiac disease epidemic, when reaching 12 years of age has a celiac disease prevalence of 29 of 1000 (95% CI 25–33) (19). The present study resulted in an increase with 8 cases shifting the prevalence to 30 of 1000 (95% CI 26–34), thus not significantly different from the previously reported estimate. This is the highest celiac disease prevalence ever reported in any general population, with the Saharawi children in Algeria being the exception (35). The ETICS study phase II is now underway with screening of 12-year-olds born in 1997, after infant feeding practices were changed, giving a prevalence to be compared with that now reported.
By sharpening the diagnostic procedure, we found not only 14 new cases of celiac disease that otherwise would have been missed but also 1 overdiagnosed case, an increase from 140 to 153 cases in the 184 children investigated. The experience from the present study is that each step in the diagnostic procedure is important and crucial to the diagnostic outcome. To summarize, we suggest multiple biopsies be taken from both proximal and distal duodenum, preferably by endoscopy. The mucosal material should be prepared in a standardized way, primarily stained with hematoxylin and eosin. Previous studies suggest usefulness of immunohistochemistry in borderline cases (20). Last but not least, it is important that the examining pathologist has expertise in gastrointestinal pathology and a good knowledge of celiac disease.
We thank all of the participating children and their families, research nurses, administrative staff, and collaborators within the school health services. We also thank our second expert pathologist, Ester Lörinc, for her important contribution to the reevaluation.
1. Jabri B, Sollid LM. Tissue-mediated control of immunopathology in coeliac disease. Nat Rev Immunol 2009; 9:858–870.
2. Di Sabatino A, Corazza GR. Coeliac disease. Lancet 2009; 373:1480–1493.
3. Fasano A. Clinical presentation of celiac disease
in the pediatric population. Gastroenterology 2005; 128:S68–S73.
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 2004; 38:181–186.
5. Haines ML, Anderson RP, Gibson PR. Systematic review: the evidence base for long-term management of coeliac disease. Aliment Pharmacol Ther 2008; 28:1042–1066.
6. 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.
7. Bonamico M, Thanasi E, Mariani P, et al
. Duodenal bulb biopsies in celiac disease
: a multicenter study. J Pediatr Gastroenterol Nutr 2008; 47:618–622.
8. Weir DC, Glickman JN, Roiff T, et al
. Variability of histopathological changes in childhood celiac disease
. Am J Gastroenterol 2010; 105:207–212.
9. Rashid M, MacDonald A. Importance of duodenal bulb biopsies in children for diagnosis
of celiac disease
in clinical practice. BMC Gastroenterol 2009; 9:78.
10. Manuel PD, Walker-Smith JA, France NE. Patchy enteropathy in childhood. Gut 1979; 20:211–215.
11. Scott BB, Losowsky MS. Patchiness and duodenal-jejunal variation of the mucosal abnormality in coeliac disease and dermatitis herpetiformis. Gut 1976; 17:984–992.
12. Marsh MN. Grains of truth: evolutionary changes in small intestinal mucosa in response to environmental antigen challenge. Gut 1990; 31:111–114.
13. Walker-Smith JA, 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.
14. Kaukinen K, Collin P, Maki M. Latent coeliac disease or coeliac disease beyond villous atrophy? Gut 2007; 56:1339–1340.
15. Dickson BC, Streutker CJ, Chetty R. Coeliac disease: an update for pathologists. J Clin Pathol 2006; 59:1008–1016.
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; 39:26–29, discussion 30–32.
17. Ivarsson A, Persson LA, Nystrom L, et al
. Epidemic of coeliac disease in Swedish children. Acta Paediatr 2000; 89:165–171.
18. Ivarsson A, Hernell O, Stenlund H, et al
. Breast-feeding protects against celiac disease
. Am J Clin Nutr 2002; 75:914–921.
19. 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 2009; 49:170–176.
20. Veress B, Franzen L, Bodin L, et al
. Duodenal intraepithelial lymphocyte-count revisited. Scand J Gastroenterol 2004; 39:138–144.
21. Grant C, Hogberg L, Falth-Magnusson K, et al
. The clinical relevance of duodenal intraepithelial lymphocyte counts in children treated for coeliac disease. Acta Paediatr 2008; 97:1133–1135.
22. Ludvigsson JF, Montgomery SM, Ekbom A, et al
. Small-intestinal histopathology and mortality risk in celiac disease
. JAMA 2009; 302:1171–1178.
23. Biagi F, Bianchi PI, Campanella J, et al
. The prevalence
and the causes of minimal intestinal lesions in patients complaining of symptoms suggestive of enteropathy: a follow-up study. J Clin Pathol 2008; 61:1116–1118.
24. 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.
25. Hill PG, Holmes GK. Coeliac disease: a biopsy is not always necessary for diagnosis
. Aliment Pharmacol Ther 2008; 27:572–577.
26. 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.
27. 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.
28. 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 2010; 50:140–146.
29. Pais WP, Duerksen DR, Pettigrew NM, et al
. How many duodenal biopsy specimens are required to make a diagnosis
of celiac disease
? Gastrointest Endosc 2008; 67:1082–1087.
30. Branski D, Faber J, Freier S, et al
. Histologic evaluation of endoscopic versus suction biopsies of small intestinal mucosae in children with and without celiac disease
. J Pediatr Gastroenterol Nutr 1998; 27:6–11.
31. Meijer JW, Wahab PJ, Mulder CJ. Small intestinal biopsies in celiac disease
: duodenal or jejunal? Virchows Arch 2003; 442:124–128.
32. Thomson M, Kitching P, Jones A, et al
. Are endoscopic biopsies of small bowel as good as suction biopsies for diagnosis
of enteropathy? J Pediatr Gastroenterol Nutr 1999; 29:438–441.
33. Eltumi MA, Ong PS, Francis ND, et al
. A comparison of endoscopic and capsule small intestinal biopsy techniques in children with upper gastrointestinal disorders. J Paediatr Child Health 1996; 32:255–256.
34. Jarvinen TT, Collin P, Rasmussen M, et al
. Villous tip intraepithelial lymphocytes as markers of early-stage coeliac disease. Scand J Gastroenterol 2004; 39:428–433.
35. Catassi C, Ratsch IM, Gandolfi L, et al
. Why is coeliac disease endemic in the people of the Sahara? Lancet 1999; 354:647–648.