Secondary Logo

Journal Logo

Original Articles: Gastroenterology: Celiac Disease

The Association Between Celiac Disease and Eosinophilic Esophagitis

Mayo Experience and Meta-analysis of the Literature

Hommeida, Salim*; Alsawas, Mouaz†,‡; Murad, Mohammad H.†,‡; Katzka, David A.§; Grothe, Rayna M.*,§; Absah, Imad*,§

Author Information
Journal of Pediatric Gastroenterology and Nutrition: July 2017 - Volume 65 - Issue 1 - p 58-63
doi: 10.1097/MPG.0000000000001499

Abstract

What Is Known

  • Celiac disease and eosinophilic esophagitis are immune-mediated diseases.
  • Dietary triggers play a role in the pathogenesis.
  • Studies had variable results in terms of the association between celiac disease and eosinophilic esophagitis.

What Is New

  • A diagnosis of celiac disease in children is not associated with increased risk of eosinophilic esophagitis.
  • Higher incidence of eosinophilic esophagitis in patients with celiac disease in previous studies could be due to referral or selection biases.
  • Esophageal eosinophilia may respond to a gluten-free diet.

See “Eosinophilic Esophagitis and Celiac Disease: A True Association or Coincidence?” by Watkins and Blanchard on page 1.

Eosinophilic esophagitis (EoE) is a chronic immune/antigen-mediated esophageal disease characterized by eosinophil-predominant inflammation of the esophagus (1). The diagnosis of EoE based on the most recent guideline by the American College of Gastroenterology (ACG) requires mucosal eosinophilia that is isolated to the esophagus and persists after 2 months of proton pump inhibitor (PPI) trial with a peak value of 15 or more eosinophils per high-power field (Eos/HPF) and exclusion of other causes of esophageal eosinophilia (2). A study in Olmsted County, Minnesota, that included both adults and children estimated an EoE prevalence of 55 per 100,000 in 2006 (3). A pediatric-based study from a single institution's pathology database in Hamilton County, Ohio, estimated the incidence of EoE in children to be 1.28 per 10,000 in 2003 (4). The clinical presentation of EoE can differ by the age at presentation but usually includes feeding dysfunction, vomiting, abdominal pain, dysphagia, and food impaction (5,6).

EoE is strongly associated with other allergic conditions, such as asthma, environmental allergies, and atopic dermatitis (7–9). These atopic disorders differ not only by organ and presentation but further reflect a variety of allergy-based pathways.

CD is an immune-mediated disease of the small intestine that occurs in genetically predisposed individuals because of sensitivity to dietary gluten and its related proteins (10,11). This inflammation leads to histologic changes in the small intestine characterized by increased intraepithelial lymphocytes, crypt hyperplasia, and villous atrophy (12). Biopsy-confirmed epidemiologic studies reported a prevalence of 1:70 to 1:300 (3). The common gastrointestinal (GI) symptoms include abdominal pain, diarrhea, and failure to thrive (13). Gluten-free diet (GFD) is currently the mainstay for treatment of CD (14).

Both CD and EoE are immune-mediated diseases with dietary triggers playing a role in their pathogenesis. Avoidance of these food triggers can lead to resolution of symptoms and histologic changes in both CD and EoE (1,10,15). Both diseases result from an imbalance of the Th1/Th2 pathway (16,17). Despite the similarity in symptoms, immune-mediated pathogenesis, and management with avoidance of allergens, CD and EoE are 2 distinct diseases. The question of whether an association exists between CD and EoE has been the focus of multiple studies in both adults and children (18–23). These studies had variable results with wide differences in the strength of association or dissociation between these 2 conditions. Hence, the recommendations to obtain routine esophageal biopsies when evaluating for CD were also variable (22,24,25). In the present study, we present the experience of a tertiary care center in addition to a systematic review and meta-analysis of the literature studying the association between CD and EoE.

METHODS

Mayo Clinic Retrospective Study

The goal of the first part of the present study was to study the association between CD and EoE in the children who were assessed and treated at our institution. The study was approved by the Mayo Clinic Institutional Review Board. Data were extracted from the electronic medical record using a combination of clinical notes, histopathologic reports, and diagnosis list from January 1, 1998 to December 31, 2015. The search yielded 1860 patients in the CD group and 1025 patients in the EoE group. A retrospective chart review was conducted by 2 physicians (S.H. and M.A.). Inclusion criteria included children younger than 18 years at the time of diagnosis of CD and/or EoE who underwent triple-site biopsies (esophagus, stomach, and duodenum) at the time of the endoscopy. The diagnosis of CD was made on the basis of the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition (NASPGHAN) guidelines (2,13), and the diagnosis of EoE was made on the basis of ACG guidelines (2). All children with a diagnosis of CD had positive tissue transglutaminase IgA antibodies. In patients with CD with only increased intraepithelial lymphocytes without villous atrophy, the diagnosis of CD was confirmed by obtaining a different serologic marker (endomysial immunoglobulin A antibody or deamidated gliadin antibodies) or by checking the status of CD-associated human leukocyte antigen DQ2/DQ8. Children with inflammatory bowel disease, eosinophilic gastroenteritis and those who did not undergo both duodenal and esophageal biopsies at the same procedure were excluded. A total of 13,109 esophagogastroduodenoscopies (EGDs) were performed on 10,201 children at our institution between January 1, 1998 and December 31, 2015; many children had >1 EGD. Children who were found to have both CD and EoE had extensive data collection, including clinical presentation, laboratory results, endoscopic/histologic reports, past medical history, and family history.

Systematic Review and Meta-analysis

The second part of the present study consisted of performing a systematic review and meta-analysis of the possible association between CD and EoE in children and reporting the characteristics and outcomes of children with both conditions.

The systematic review and meta-analysis were reported according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines (26).

Eligibility Criteria

Studies were included when the diagnosis of CD and/or EoE was based on histopathologic findings according to the ACG and NASPGHAN guidelines. Studies were excluded if there was no reference to the diagnostic criteria or the diagnosis was not supported by histopathologic findings. Case reports were excluded. Studies providing duplicated information or repeated abstracts presented at different meetings were excluded. Only studies that reported odds ratio (OR) or provided enough information to calculate an OR were included in the meta-analysis. Other studies with insufficient data for meta-analysis were reported narratively and used to provide information on the clinical characteristics and treatment approaches in patients with coexisting CD and EoE.

Data Sources and Search Strategies

A comprehensive search was conducted of several databases including Ovid Medline In-Process and Other Non-Indexed Citations, Ovid EMBASE, Ovid MEDLINE, Ovid Cochrane Central Register of Controlled Trials, Ovid Cochrane Database of Systematic Reviews, and Scopus from each database's inception to February 4, 2016. An experienced librarian designed and conducted the search in association with the study principal investigator. Studies of CD and EoE were searched for using a controlled vocabulary supplemented with keywords. The actual strategy is described in Supplemental Digital Content 1, Search Strategy, http://links.lww.com/MPG/A873.

Study Selection and Data Extraction

Citations were reviewed independently by 2 reviewers and relevant data were collected. Potentially eligible studies were selected for full-text assessment. Any conflicts about the eligibility of the full studies were resolved by consensus. Authors of selected studies were contacted when needed for clarification. In studies that included both pediatric and adult populations, only data on the pediatric population were used (Table 1).

TABLE 1
TABLE 1:
Characteristic demographics of studies in the meta-analysis and/or systematic review of the association between celiac disease and eosinophilic esophagitis

Risk of Bias

To assess the methodological quality of our included studies, we used a modified tool based on the Newcastle-Ottawa scale to evaluate selection, comparability, and outcome assessment.

Data Synthesis

We used random-effect models meta-analyses to estimate pooled ORs and 95% confidence intervals (CIs) for association between CD and EoE (27). We estimated inconsistency between trials not due to chance using the I2 statistic (28,29).

RESULTS

Mayo Clinic Retrospective Study

Of 10,201 children who had at least 1 endoscopy, 595 (5.83%) had EoE. Of those, 380 (63.86%) were boys. Mean (SD) age at diagnosis of EoE was 9.42 (5.0) years. A total of 546 children (5.35%) had CD. Of those, 337 (61.72%) were boys. Mean (SD) age at diagnosis of CD was 10.67 (4.67) years. Ten children were found to have both CD and EoE. The prevalence of EoE in children with CD was 1.83%, compared to 5.83% in all children who underwent EGD in our institution and were found to have EoE. The OR of having EoE in children with CD compared to those without CD was 0.29 (95% CI, 0.154–0.545), suggesting no increased risk of EoE in children with CD in this cohort.

Because EoE is an emerging disease that is increasingly being recognized, a separate subgroup analysis for the period between January 1, 2005 and December 31, 2015 was performed. We identified 505 children with EoE, 416 with CD and 9 with both CD and EoE. During that period 6175 children underwent diagnostic EGD due having GI symptoms and did not have CD or EoE. The odd ratio of having EoE in patients with CD in this group was 0.2645 (95% CI, 0.1358–0.5152) suggesting the same outcome of no increased risk of EoE in children with CD.

Characteristics of Children With Celiac Disease and Eosinophilic Esophagitis in Our Cohort

Seven of the 10 children with both CD and EoE were boys (Table 2). Mean (SD) age at diagnosis was 9 (4) years. Many children had >1 GI presenting symptom. The most common was abdominal pain in 8 patients, followed by diarrhea and vomiting. Two asymptomatic children had CD serology screening because of family history of CD, followed by serologic and endoscopic assessment that confirmed the CD diagnosis.

TABLE 2
TABLE 2:
Demographic, endoscopic, and histopathological characteristics of patients with celiac disease and eosinophilic esophagitis

The diagnosis of CD and EoE was made on the basis of the histopathologic findings at the initial endoscopic evaluation in all but 1 of the 10 children. One year after his diagnosis of CD, this patient underwent another EGD because of persistence of abdominal pain despite normalization of CD serologic markers. He was found to have 20 Eos/HPF on esophageal biopsies and resolution of the duodenal villous atrophy.

Nine of the 10 children (90%) had family history of autoimmune disease (CD, hypothyroidism, inflammatory bowel disease, type 1 diabetes mellitus, and systemic lupus erythematosus). Seven had other atopic disorders, such as asthma, allergic rhinitis, eczema, and food allergy. Some children had >1 atopic disorder (Table 2). There was no observed difference in the level of tissue transglutaminase IgA antibodies positivity between children with only CD and those with both CD and EoE. Peripheral eosinophilia was present in 4 patients, with a mean (SD) of 1.17 (0.335) × 109/L (reference range, 0.05–0.5 × 109/L).

Endoscopic and Histologic Findings in Children With Both Celiac Disease and Eosinophilic Esophagitis in Our Cohort

Of 10 children, only 3 had a duodenal endoscopic finding (scalloping) suggestive of CD and 4 had esophageal findings (linear furrows) suggestive of EoE. Of those, only 2 had both duodenal and esophageal findings (Table 2). Five children had normal-appearing esophagus and duodenum. On histologic examination of the duodenal biopsies, only 1 patient had complete villous atrophy; the rest had partial villous atrophy. The mean (SD) number of eosinophils on esophageal biopsies was 35 (15) Eos/HPF (Table 2).

Outcome and Response to Therapy in Children With Both Celiac Disease and Eosinophilic Esophagitis in Our Cohort

All 10 children were started on a GFD because of their CD diagnosis. Four were treated only with GFD; 3 others received adjunct therapy with topical glucocorticosteroids and targeted food elimination, 2 with topical glucocorticosteroid alone, and 1 with targeted food elimination alone. Symptoms resolved in 9 patients. One child who had persistent symptoms and positive celiac serology markers was found to be nonadherent to the GFD. He received topical glucocorticosteroids because of the persistence of his symptoms. He underwent another EGD, which showed persistent esophageal eosinophilia and villous atrophy.

In addition to this patient, 4 other children also underwent another endoscopy in an average of 11 months after the initial EGD to assess their response to the GFD and adjunct therapy (food elimination and/or topical glucocorticosteroids). They all had resolution of their duodenal villous atrophy and esophageal eosinophilia.

Outcome and Response to Therapy in Children Treated With Only Gluten-free Diet in Our Cohort

Of the 10 children who had both CD and EoE, 4 were treated with GFD only. Two of those children were asymptomatic and were screened for CD because of a positive family history. The presenting symptoms of the other 2 children included abdominal pain, vomiting, and diarrhea. Of the children who were treated with only GFD, 2 had a normal endoscopic appearance of the esophagus and duodenum. Compared to all patients with CD and EoE, they had the lowest number of Eos/HPF (15–20) on esophageal biopsies and partial villous atrophy on duodenal biopsies. None of these children underwent another EGD because their clinical symptoms resolved and CD serologic markers normalized within 6 months.

Systematic Review and Meta-analysis

The initial search identified 218 studies; after title and abstract screening, 39 studies were included in the full-text screening process. In this level, 17 studies were excluded for reasons reported in Figure 1, and inter reviewer agreement was 79.5%. Including the present Mayo Clinic retrospective study, a total of 22 studies were included in this systematic review (10 published articles and 12 abstracts presented at national or international meetings) (Table 1).

FIGURE 1
FIGURE 1:
Flowchart of the process in identifying and excluding studies in the meta-analysis and reason for exclusion.

Five studies provided data on the treatments and outcomes, 13 provided data on histologic manifestations, and 5 studies reported sufficient data to be included in meta-analysis. The effect of GFD alone in treating EoE was not reported in the studies that were included in the meta-analysis, except for the data from our cohort. One of the studies included in the systemic review reported no resolution in esophageal eosinophilia in 2 patients after being on GFD (30).

A diagnosis of CD did not increase the risk of EoE with or without inclusion of our cohort results in the analysis. The OR of having EoE in children with CD compared with those without CD, after including our data, was 0.525 (95% CI, 0.364–0.797) (Fig. 2). The OR after excluding our data was 0.672 (95% CI, 0.438–1.030). Inconsistency among these studies was moderate (I2 = 41%).

FIGURE 2
FIGURE 2:
Meta-analysis of the odds ratio of having eosinophilic esophagitis in patients with celiac disease.

The risk of bias in the included studies was moderate (Supplemental Digital Content 2, Table, http://links.lww.com/MPG/A874).

Characteristics of Children With Both Celiac Disease and Eosinophilic Esophagitis in the Systematic Review

Six studies with a total of 45 children reported characteristics of those in whom both CD and EoE were diagnosed. Twenty-nine children (64.4%) were boys (Supplemental Digital Content 3, Table, http://links.lww.com/MPG/A875), and mean (SD) age at diagnosis was 10.3 (2.6) years. Of 32 children whose family history was reported, 15 (46.87%) had a family history of CD. The most common presenting symptoms were abdominal pain, diarrhea, and vomiting. The diagnosis of EoE in the majority of these children was incidental, and only 13% had dysphagia.

Endoscopic Findings in Children With Both Celiac Disease and Eosinophilic Esophagitis in the Systematic Review

These studies also reported the endoscopic findings in patients with both CD and EoE (Supplemental Digital Content 4, Table, http://links.lww.com/MPG/A876). Some patients had >1 reported endoscopic finding of the esophagus (normal-appearing esophagus in 33%, linear furrows in 54%, white plaques in 40%, and ringed esophagus in 7%). Eighty-five percent of these patients had a normal-appearing duodenum, whereas 15% had scalloping of the duodenum.

DISCUSSION

To our knowledge, this is the largest study to date to report on the association between CD and EoE in children. It also includes the first meta-analysis of studies that based the diagnosis of CD and EoE exclusively on the histopathologic findings. Data from our center and meta-analysis of the literature suggested that children with CD did not have higher risk of developing EoE when compared to all children who underwent endoscopic assessment.

Because the co-occurrence of CD and EoE was described by Verzegnassi et al (31) in 2007, the association between CD and EoE has been investigated by multiple studies with a wide difference in the strength of association or dissociation (31). The co-occurrence of these diseases was reported to be as high as 35.2% (23). In studies in which patients underwent both esophageal and duodenal biopsies at the same time, the coexistence of CD and EoE was lower than previously reported; however, it was still significantly higher than the prevalence of CD and EoE in the general population (22,32). Incidence and prevalence of both CD and EoE are increasing (9,33). It is not completely clear whether this is a true increase in the incidence or whether it is due to increased awareness, availability of sensitive tests, and increase in the number of endoscopic tests in children (34). Children with either condition can present with similar GI symptoms, with abdominal pain and vomiting being the most commonly reported (13,35). Hence, having both conditions simultaneously should be expected especially in children presenting with suggestive GI symptoms seeking medical care. The strength of association and response to management should, however, be questioned. Is this coexistence coincidental or casual?

In our study cohort, children with CD did not have increased risk of EoE. Similarly, a meta-analysis of the literature, including 5 studies in which the diagnosis of CD and EoE was based on histopathologic findings and information was available to calculate the OR, showed no association between CD and EoE with variable ORs. The overall OR of having EoE in children with CD was 0.525 (95% CI, 0.364–0.797), suggesting no increased risk of EoE in children with CD. In fact, the meta-analysis suggests a possible protective role of CD in terms of developing EoE. All these patients will be on GFD, resulting in elimination of wheat, which is considered the second most common dietary trigger for EoE (36).

The discrepancy between our results and the previous reports of increased incidence (up to 10.7%) of EoE in children with CD could be due to referral and/or selection biases. All children included in these studies had GI complaints necessitating referral and endoscopic assessment, so the likelihood of finding GI disorders such as CD and/or EoE is expected to be higher than that in the general population. Because GFD is currently the recommended treatment for patients with CD, it is difficult to estimate the role of gluten in inducing EoE or esophageal eosinophilia (37). All the cases of EoE that were included in out cohort would have not responded to PPI therapy to meet the inclusion criteria for EoE, so we could not discuss the PPI responsive esophageal eosinophilia in this cohort. Abraham et al (38) reported that in 9 children with CD and EoE, GFD alone did not lead to endoscopic or histologic improvement in 6 of 7 patients (38). Although in our study, 4 of the 10 patients with CD and EoE had complete resolution of their clinical symptoms on GFD. None of those patients underwent another EGD to assess their histologic response.

In the studies that reported the endoscopic appearance of patients who had both CD and EoE, 30% of patients had a normal-appearing esophagus, 54% had linear furrows, 40% had white plaques, and 7% had ringed esophagus (19,22,38,39). The occurrence of these endoscopic findings is higher than what is reported in other studies, except for ringed esophagus, which was less frequent (40). Although endoscopic appearance of the esophagus has a low sensitivity for pathologic changes in children, the specificity was reported to be as high as 95% (40). Given that approximately one-third of these patients had a normal-appearing esophagus, obtaining biopsies even in the normal-appearing esophagus should be considered.

In the same studies in which the endoscopic appearance was reported, 85% of the patients had normal-appearing duodenum and 15% had scalloping of the duodenum, which supports the role of duodenal biopsies in confirming the diagnosis of CD (37).

Possible limitations of the present study include the retrospective nature of the cohort and the small number of children with both CD and EoE, which can decrease the confidence in conclusions and warrants low certainty (41). The small number of studies with sufficient data to be included in the meta-analysis limited our ability to assess the presence of true association between EoE and CD.

CONCLUSIONS

Based on our cohort and the observational data the diagnosis of CD in children is not associated with increased risk of EoE. Higher incidence of EoE in patients with CD in previous studies could be due to referral or selection biases. A normal-appearing esophagus and/or duodenum does not exclude histopathology. Esophageal eosinophilia may respond to a GFD. More controlled studies are needed to evaluate the role of GFD in esophageal eosinophilia.

REFERENCES

1. Furuta GT. Eosinophilic esophagitis: update on clinicopathological manifestations and pathophysiology. Curr Opin Gastroenterol 2011; 27:383–388.
2. Dellon ES, Gonsalves N, Katzka DA, et al. ACG clinical guideline: evidenced based approach to the diagnosis and management of esophageal eosinophilia and eosinophilic esophagitis (EoE). Am J Gastroenterol 2013; 108:679–692.
3. Gujral N, Freeman HJ, Thomson AB. Celiac disease: prevalence, diagnosis, pathogenesis and treatment. World J Gastroenterol 2012; 18:6036–6059.
4. Noel RJ, Putnam P, Rothenberg ME. Eosinophilic esophagitis. N Engl J Med 2004; 351:9.
5. Baxi S, Gupta SK, Fitzgerald JF, et al. Clinical presentation of patients with eosinophilic inflammation of the esophagus. Gastrointest Endosc 2006; 64:473–478.
6. Rodrigues M, D’Amico MFM, Sipahy AM, et al. Clinical manifestations, treatment, and outcomes of children and adolescents with eosinophilic esophagitis. Erratum appears in J Pediatr (Rio J) 2015; 91:507Sipahy, Aytan M [corrected to Sipahi, Aytan M]]. Jornal de Pediatria. 89(2):197-203.
7. Roy-Ghanta S, Larosa DF, Katzka DA. Atopic characteristics of adult patients with eosinophilic esophagitis. Clin Gastroenterol Hepatol 2008; 6:531–535.
8. Orenstein SR, Shalaby TM, Kocoshis SA, et al. The spectrum of pediatric eosinophilic esophagitis beyond infancy: a clinical series of 30 children [Erratum appears in Am J Gastroenterol 2001;96(7):2290]. Am J Gastroenterol 2000; 95:1422–1430.
9. Prasad GA, Alexander JA, Talley NJ, et al. Epidemiology of eosinophilic esophagitis over three decades in Olmsted County, Minnesota. Clin Gastroenterol Hepatol 2009; 7:1055–1061.
10. Shan L, Molberg O, Khosla C, et al. Structural basis for gluten intolerance in celiac sprue. Science 2002; 297:2275–2279.
11. Arentz-Hansen H, McAdam SN, Sollid LM, et al. Celiac lesion T cells recognize epitopes that cluster in regions of gliadins rich in proline residues. Gastroenterology 2002; 123:803–809.
12. Marsh MN, Crowe PT. Morphology of the mucosal lesion in gluten sensitivity. Baillieres Clin Gastroenterol 1995; 9:273–293.
13. Hill ID, Dirks MH, Seidman EG, 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.
14. Kelly CP, Bai JC, Leffler DA, et al. Advances in diagnosis and management of celiac disease. Gastroenterology 2015; 148:1175–1186.
15. Mishra A, Marsh MN, Crowe PT, et al. Fundamental signals that regulate eosinophil homing to the gastrointestinal tract. J Clin Invest 1999; 103:1719–1727.
16. Maiuri L, Ciacci C, Londei M, et al. Association between innate response to gliadin and activation of pathogenic T cells in coeliac disease. Lancet 2003; 362:30–37.
17. Zhu X, Wang M, Mishra A, et al. Interleukin-15 expression is increased in human eosinophilic esophagitis and mediates pathogenesis in mice. Gastroenterology 2010; 139:182.e7–193.e7.
18. Ahmed OI, Qasem SA, Hill ID, et al. Esophageal eosinophilia in pediatric patients with celiac disease: is it a causal or an incidental association? J Pediatr Gastroenterol Nutr 2015; 60:493–497.
19. Dharmaraj R, Hagglund K, Lyons H. Eosinophilic esophagitis associated with celiac disease in children. BMC Res Notes 2015; 8:263.
20. Jensen ET, Eluri S, Dellon ES, et al. Increased risk of esophageal eosinophilia and eosinophilic esophagitis in patients with active celiac disease on biopsy. Clin Gastroenterol Hepatol 2015; 13:1426–1431.
21. Johnson JB, Boynton KK, Peterson KA. Co-occurrence of eosinophilic esophagitis and potential/probable celiac disease in an adult cohort: a possible association with implications for clinical practice. Dis Esophagus 2016; 29:977–982.
22. Leslie C, Mews C, Ravikumara M, et al. Celiac disease and eosinophilic esophagitis: a true association. J Pediatr Gastroenterol Nutr 2010; 50:397–399.
23. Quaglietta L, Coccorullo P, Staiano A, et al. Eosinophilic oesophagitis and coeliac disease: is there an association? Aliment Pharmacol Ther 2007; 26:487–493.
24. Ooi CY, Day AS, Lemberg DA, et al. Eosinophilic esophagitis in children with celiac disease. J Gastroenterol Hepatol 2008; 23 (7 pt 1):1144–1148.
25. Ludvigsson JF, Aro P, Ronkainen J, et al. Celiac disease, eosinophilic esophagitis and gastroesophageal reflux disease, an adult population-based study. Scand J Gastroenterol 2013; 48:808–814.
26. Moher D, Liberati A, Altam DG, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ 2009; 339:b2535.
27. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986; 7:177–188.
28. Higgins JP, Thompson SG, Altam DG, et al. Measuring inconsistency in meta-analyses. BMJ 2003; 327:557–560.
29. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med 2002; 21:1539–1558.
30. Croaker D, Swann J, Beardsley C, et al. Heterogeneity in the diagnosis of coeliac disease in paediatric patients. Arch Dis Child 2012; 97:A208Conference: 4th Congress of the European Academy of Paediatric Societies Istanbul Turkey. Conference Start: 20121005 Conference End: 20121009. Conference Publication: (var.pagings).
31. Verzegnassi F, Bua J, Ventura A, et al. Eosinophilic oesophagitis and coeliac disease: is it just a casual association? Gut 2007; 56:1029–1030.
32. Thompson JS, Lebwohl B, Green PH, et al. Increased incidence of eosinophilic esophagitis in children and adults with celiac disease. J Clin Gastroenterol 2012; 46:e6–e11.
33. Ludvigsson JF, Aro P, Ronkainen J, et al. Increasing incidence of celiac disease in a North American population. Am J Gastroenterol 2013; 108:818–824.
34. Straumann A, Simon H-U. Eosinophilic esophagitis: escalating epidemiology? J Allergy Clin Immunol 2005; 115:418–419.
35. Straumann A, Spichtin HP, Simon HU, et al. Natural history of primary eosinophilic esophagitis: a follow-up of 30 adult patients for up to 11.5 years. Gastroenterology 2003; 125:1660–1669.
36. Kagalwalla AF, Shah A, Nelson SP, et al. Identification of specific foods responsible for inflammation in children with eosinophilic esophagitis successfully treated with empiric elimination diet. J Pediatr Gastroenterol Nutr 2011; 53:145–149.
37. Rubio-Tapia A, Hill ID, Murray JA, et al. ACG clinical guidelines: diagnosis and management of celiac disease. Am J Gastroenterol 2013; 108:656–676. quiz 677.
38. Abraham JR, Persad R, Huynh HQ, et al. Gluten-free diet does not appear to induce endoscopic remission of eosinophilic esophagitis in children with coexistent celiac disease. Can J Gastroenterol 2012; 26:521–524.
39. Stewart MJ, Shaffer E, Storr MA, et al. The association between celiac disease and eosinophilic esophagitis in children and adults. BMC Gastroenterol 2013; 13:96.
40. Kim HP, Vance RB, Dellon ES, et al. The prevalence and diagnostic utility of endoscopic features of eosinophilic esophagitis: a meta-analysis. Clin Gastroenterol Hepatol 2012; 10:988.e5–996.e5.
41. Murad MH, Montori VM, Guyatt G, et al. How to read a systematic review and meta-analysis and apply the results to patient care: users’ guides to the medical literature. JAMA 2014; 312:171–179.
Keywords:

celiac disease; celiac sprue; children; eosinophilic esophagitis; meta-analysis; systematic review

Supplemental Digital Content

© 2017 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology,