Evidence-Based Management of Celiac Disease in Children 2016

Management for Celiac Disease in Children

Celiac disease (CD) is a systemic immune-mediated illness triggered by gluten in genetically susceptible persons and affects ∼1% of the world’s population. Although great progress has been made in the diagnosis and management of CD in recent years, important problems still exist – optimal follow-up care is not being provided.1

In an attempt to provide thoughtful and effective best practices for managing CD and associated disorders in children, a group of experts in the field, using evidence-based data were convened to critically review and discuss the available data to provide an evidenced-based approach to optimal care.  When the quality of evidence was not sufficient, expert opinion was used.2,3,4,5

The 6 categories examined included the following:

  1. Bone Health
  2. Hematologic Issues
  3. Endocrine Problems
  4. The Liver and CD
  5. Nutritional Problems and CD
  6. Testing and Monitoring

1. Bone Health

  • Background – CD can affect the bone health of children in several ways with a variety of signs and symptoms including bone pain, rickets, tetany, osteomalacia, osteopenia, osteoporosis, fractures associated with minimal trauma, or growth failure with or without symptoms of malabsorption. With the exception of osteopenia or osteoporosis identified by the evaluation of bone mineral density, these are now rarely the presenting signs and symptoms of CD in children because of the widespread use of CD serological testing to assist in diagnosis. Initiation of a GFD rapidly restores bone mass to normal levels in almost all children and some adolescents. Instruction on age-appropriate intake of calcium, vitamin D, and the need for exercise to promote bone health should be provided during nutritional counseling at the time of diagnosis. Detailed information on the mechanisms of bone injury, bone mineral density, fracture risk, effect of the GFD, and calcium and vitamin D intake are included in the Supplemental Information.

Bone Health – Best Practice 1  Should routine screening for bone health by using calcium, PO4, alkaline phosphatase, ± parathyroid hormone (excluding vitamin D) be done routinely for all children being evaluated for CD at the time of diagnosis?

  • With the advent of serological testing, children now present with a variety of milder symptoms and associated conditions. Abnormalities of the above tests most often occur with presentations that are associated with severe malabsorption, prolonged delay in diagnosis, or clinical presentations suggestive of bone disease including bone pain, rickets, osteomalacia, tetany, or fractures caused by minimal trauma. With these presentations, screening with the tests listed above should be conducted. If abnormalities are detected, patients should be treated with dietary calcium and vitamin D including supplements if necessary. Annual serial follow-up should be conducted until results normalize.

Bone Health – Best Practice 2 – Should screening for vitamin D status be done routinely for all children being evaluated for CD at the time of diagnosis?

  • Small case-control studies exist that differ in patient age, geographic location, time of year of measurement, clinical presentation, and results. Some demonstrate vitamin D deficiency. Furthermore, many children and adolescents do not receive adequate dietary calcium or vitamin D. Additional data on vitamin D status for children and adolescents with CD from multiple geographic areas across North America that control for sunlight exposure are needed to further define which children require vitamin D testing.

Bone Health – Best Practice 3 – Should screening using imaging studies (bone density) evaluating bone health be done routinely for all children and adolescents with CD when they are seen at 1-year follow-up?

  • When CD is diagnosed at a young age with a short duration of symptoms, bone density recovers rapidly and completely to normal values for age and size.  Routine follow-up bone density testing is not required or cost-effective. More data about bone density recovery in adolescents with various presentations of CD on a strict GFD are required. Abnormalities of bone density are likely to occur with presentations that are associated with severe malabsorption, prolonged delay in diagnosis, or clinical presentations suggestive of bone disease, including bone pain, rickets, osteomalacia, tetany, or fractures caused by minimal trauma. With these presentations, bone mineral density is the test of choice to obtain at diagnosis. If abnormalities are detected, serial follow-up every 1 to 2 years should be conducted until results normalize. This is especially true for adolescents where recovery may be slower and dietary compliance may be more problematic.

Bone Health – Best Practice 4  Should instructions on age-appropriate intake of calcium and vitamin D, including information on the impact of geographic region and season of the year, be provided during the initial GFD counseling?

  • Data from multiple geographic regions demonstrate that children and adolescents consume diets deficient in vitamin D and calcium. Additional studies have revealed that recovery of bone mineral density will occur in children if a GFD with adequate nutrition is provided.  In some geographic areas, this can be accomplished by GFD alone. However, in other areas, vitamin D supplementation will likely be required.

Bone Health – Best Practice 5  Should screening using imaging studies (bone density) to evaluate bone health be done routinely for selected patients with CD who do not adhere to a GFD?

  • Maximum bone density is accrued in adolescence, late teens, and early 20s, the ages when adherence to a GFD is most difficult. At these ages, symptoms may not develop with poor dietary control and a reduction in bone density may occur that increases fracture risk and can result in early onset osteoporosis. If abnormalities in bone mineral density are identified, caregivers should provide an explanation of increased risk of bone disease, as well as dietary counseling, which includes instructions about calcium and vitamin D intake and supplementation.

2. Hematologic Problems

  • Background  CD has been associated with a variety of hematologic disorders, of which anemia is by far the most common. In fact, anemia may be the only clinical abnormality identified in many patients and can be the presenting feature of CD, especially in older children and adults. Anemia in children with CD can be the end result of several different, and sometimes interrelated causes; however, the single most common type of anemia is iron deficiency.

Hematologic Problems  Best Practice 6 – Should screening for anemia using a combination of tests including a complete blood cell (CBC) count (CBC plus evaluation of mean cell volume), ferritin, iron, and total iron-binding capacity be done routinely for all children being evaluated for CD at the time of diagnosis?

  • There is abundant evidence in the literature about the prevalence of anemia, especially iron-deficient, in children with CD at the time of diagnosis. This group unanimously supports the measurement of “iron studies” as reported above as an appropriate screening tool, in light of the possible need for replacement.

Hematologic Problems – Best Practice 7  Should a CBC count be obtained routinely for all children undergoing follow-up evaluation for CD?

  • Although a consensus was reached by our group on the question, the evidence to support this position comes from studies of small populations, comprising a variety of ages and clinical presentations.

Hematologic Problems – Best Practice 8  Should screening for folate deficiency be done routinely for all children being evaluated for CD at the time of diagnosis?

  • A consensus was reached by our group on the lack of need to routinely test for folate deficiency. However, again the available data come from small, case report studies addressing various ages and clinical presentations.

3. Endocrine-Associated Disorders in CD

  • Background Endocrine disorders frequently cooccur with CD, primarily due to their shared HLA predisposition, but the association is also affected by shared non-HLA variants.
  • Autoimmune thyroid disease and type 1 diabetes mellitus (T1DM) are the most common autoimmune diseases that occur with CD but Addison disease, parathyroid disorders, and growth hormone deficiency have also been reported; however, they are much less common.
  • The frequency at which CD is diagnosed in individuals with type 1 diabetes and autoimmune thyroid disorders ranges from 3% to 12% for type 1 diabetes and up to 7% for autoimmune thyroid diseases; this has led expert panels from the 2 largest pediatric gastroenterology societies, North American Society for Pediatric Gastroenterology, Hepatology and Nutrition (NASPGHAN) and European Society for Pediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN), to recommend that CD be routinely screened for in patients with these disorders.
  • However, the reverse recommendation has not been made because there are insufficient data to support screening for associated endocrine disorders in individuals with an existing diagnosis of CD.
  • Nine sets of recent reviews and guidelines, all published since 2005, mention screening for CD in individuals with endocrine disorders such as T1DM and thyroid disease. Three also mention screening in individuals with Addison disease. All of these guidelines also recommended additional screening for CD for children with short stature or delayed puberty.

Endocrine-associated Disorders in CD – Best Practice 9  Should individuals with CD be screened for T1DM or prediabetes with tests including islet autoantibodies?

  • Insufficient data exist to establish the risk of diabetes in this population. Furthermore, because no preventative strategies exist, it is not recommended to screen for a prediabetic state outside of the research setting.

Endocrine-associated Disorders in CD – Best Practice 10 – Given the slightly increased risk of diabetes in individuals with CD, should counseling for signs and symptoms of diabetes be recommended?

  • The recommendation was based on panel’s judgment of the limited risk of counseling. However, evidence for the effectiveness of counseling on preventing morbidity or mortality is not available.

Endocrine-associated Disorders in CD – Best Practice 11 – Should thyroid disease be screened at the time of diagnosis in children with CD?

  • A significantly elevated overall risk of autoimmune thyroid disease, especially Hashimoto’s disease, exists in those with CD.84–88,90 Screening with thyrotropin is recommended in those with type 1 diabetes, and should be used also in CD.80 The serum thyrotropin assay is accurate and widely available to screen for all common forms of hypothyroidism and hyperthyroidism. In addition, effective therapies for thyroid disease are available.

Endocrine-associated Disorders in CD – Best Practice 12 – Should thyroid disease be screened for at the time of follow-up evaluation for children with CD?

  • Although thyroid disease has been determined to be a coexisting condition, the actual prevalence of thyroid disease in established CD has not been determined.

Endocrine-associated Disorders in CD – Best Practice 13  Should screening for thyroid disease be performed in children with CD by using antithyroid antibodies?

  • The natural history of thyroid autoimmunity and its relationship with the development of clinical thyroid disease has not been determined.

4. The Liver and CD

  • Background  The liver can be 1 of the major sites for extraintestinal manifestations of CD. A spectrum of liver abnormalities has been described, ranging from elevated aminotransferases (cryptogenic hypertransaminasemia) to celiac hepatitis to autoimmune liver disease.

The Liver and CD – Best Practice 14  Should screening for liver disease using tests including aspartate aminotransferase (AST) and alanine aminotransferase (ALT) be done routinely for all children being evaluated for CD at the time of diagnosis?

  • Many studies identify celiac hepatitis as a possible presenting sign of CD. The hepatitis usually resolves with a GFD.

The Liver and CD – Best Practice 15 – Should routine screening for hepatitis B immunization status be done for all children being evaluated for CD at the time of diagnosis?

  • The current literature estimates that 30% to 70% of patients with CD are nonresponsive to hepatitis B vaccine before treatment. If this is accurate, a serious public health concern exists. More large-scale studies are needed to validate this estimate.

5. Nutritional Problems and CD

  • Background – Nutritional problems in CD can occur as a result of intestinal inflammation from the disease process itself and as a consequence of medical nutritional therapy (MNT) with the GFD. It is important to consider this dynamic situation, as nutritional issues at the time of diagnosis may change after implementation of the GFD.

Nutritional Problems and CD – Best Practice 16 – Should assessment of height, weight, and BMI or weight for height ratio in children younger than 3 years old be done routinely for all children being evaluated for CD at the time of diagnosis and follow-up?

  • The assessment of anthropometric parameters is important in all children, but particularly important in children with CD who often have a suboptimal nutritional status at the time of diagnosis. Monitoring response in growth on the GFD is essential to assure normal growth and development.

Nutritional Problems and CD – Best Practice 17 – Should all children being evaluated and treated for CD have access to an experienced dietitian who is knowledgeable about CD?

  • The only treatment of CD is MNT with a strict GFD. Therefore, referral to an experienced registered dietitian, knowledgeable about CD, is the optimal way to provide thorough nutritional assessment and education related to the GFD.

Nutritional Problems and CD – Best Practice 18 – Should screening for zinc and other trace elements (besides iron) be routinely obtained for all patients being evaluated for CD at the time of diagnosis?

  • The evidence to support or refute the practice of routine screening for Zn and trace elements is weak, as it comes from studies of small populations, comprising a variety of ages and clinical presentations. More research is needed.

Nutritional Problems and CD – Best Practice 19 – Should multivitamin supplementation be offered routinely to all children with CD at the time of diagnosis?

  • Although a consensus was reached by our group on the basis of expert opinion and practice, there are no well- designed studies evaluating the clinical benefit of providing a gluten-free multivitamin to children newly diagnosed with CD.

6. Testing and Monitoring

  • Background – Several tests are used to diagnose and monitor CD, including serologic tests, genetic testing, and histology.1–6 The current diagnostic algorithm for CD includes initial screening serological tests, followed by a confirmatory small intestinal biopsy revealing the autoimmune insult typical of CD in children and adults. The relative merits of these tests in various situations, including detailed information on initial diagnosis: serology, serology in IgA deficient patients, histology, use of HLA testing, monitoring successful compliance to the GFD, and current guidelines are included

Testing and Monitoring – Best Practice 20 – Should quantitative IgA and IgA anti-tissue transglutaminase (tTG) antibody be obtained routinely as the initial screening tests for all children being evaluated for CD?

  • Most of the studies published, including population studies and studies specifically focused at comparing commercially available assays, report extremely high specificity. However, well- designed studies to assess sensitivity have not been performed. These studies should be based on the use of endoscopy and histologic analysis as primary standard to validate the sensitivity of these tests.

Testing and Monitoring – Best Practice 21 – Should IgA anti-tTG antibody be obtained for all children diagnosed with CD at periodic intervals after diagnosis to help monitor compliance with the GFD?

  • There have been several reports with variable data revealing that tTG IgA antibody titers tend to decrease or completely return within normal limits after 6 to 12 months after the implementation of a GFD. However, while tTG IgA ELISA is a validated assay for the diagnosis of CD, this assay has not been validated for monitoring.

Testing and Monitoring – Best Practice 22 – Should the use of the antiendomysial antibody be limited to patients with comorbidities that increase the chance of false-positive tTG antibodies?

  • There are several reports in the literature revealing that low tTG IgA titers are often detected in patients affected by autoimmune diseases other than CD, including type 1 diabetes and autoimmune liver disease.

Testing and Monitoring – Best Practice 23 – A negative serologic evaluation cannot rule out CD.

  • There is strong evidence in the literature that ∼10% of celiac cases can test falsely negative to the tTG IgA test.

Testing and Monitoring – Best Practice 24 – Should HLA typing be considered in the evaluation of children at risk for CD who have negative serology?

  • Several studies suggested that screening for HLA DQ2/8 in at-risk children, who test negative for tTG IgA antibodies, can be cost-effective in deciding whether to continue (HLA compatible) or not continue (HLA not compatible) monitoring for CD in these children over time.

Testing and Monitoring – Best Practice 25 – Should the use of HLA typing be considered for use in patients regarded as diagnostic dilemmas, including children who have already been placed on a GFD?

  • In the ever growing situation in which children have been placed on a GFD before confirming the diagnosis of CD, the assessment of HLA status can be of great assistance in deciding whether to perform a gluten challenge.

Summary of Consensus Best Practices

1. Routine screening for bone health (biochemical studies and imaging)
2. Measure 25-OH vitamin D level
3. Measure bone density at 1 y
4. Provide counseling on age-appropriate intake of calcium and vitamin D supplementation by a dietitian
5. Measure bone density in patients not adhering to GFD despite dietary counseling

6. Routine screening for anemia (CBC, evaluation of mean cell volume, ferritin, iron, total iron-binding capacity)
7. Routinely obtain CBC at follow-up evaluation
8. Routine initial screening for folate deficiency (serum folate)

9. Routine screening for type 1 diabetes
10. Routine counseling about signs and symptoms of diabetes 11. Routine screening for thyroid disease at time of diagnosis (thyrotropin)
12. Routine screening for thyroid disease at follow-up (thyrotropin) 13. Screening for thyroid disease using antithyroid antibodies

14. Routine screening for ALT and AST
15. Screening for hepatitis B virus immunization status

16. Routine assessment of anthropometric measures
17. Access to an experienced dietitian
18. Routine screening for Zn and other trace elements at time of diagnosis
Exception: severe malabsorption, prolonged delay in diagnosis 19. Routine vitamin supplementation

20. Routine initial testing with quantitative IgA and IgA anti-tTG antibody
21. Routine testing with IgA anti-tTG Ab at periodic intervals to help monitor compliance with GFD
22. Use of IgA antiendomysial antibody limited to patients with comorbidities that increase the chance of false-positive tTG antibodies
23. Negative serologic evaluation cannot rule out CD
24. Consider use of HLA typing for children at risk for CD who have negative serology
25. Consider use of HLA typing for patients considered as diagnostic


  • John Snyder, MD, Chief of the Division of Gastroenterology, Hepatology and Nutrition at Children’s National Health System in Washington, DC, passed away suddenly in July, 2016 after sustaining a critical injury in a biking accident while traveling in France. Dr Snyder was board certified in pediatrics and pediatric gastroenterology and nationally recognized for his contributions to GI research in areas like H. pylori/peptic ulcer disease and celiac disease. He developed and led Children’s pre-eminent multidisciplinary pediatric Celiac Disease program and served as Professor of Pediatrics at the George Washington University School of Medicine and Health Sciences. Dr Snyder also worked extensively in the fields of infant nutrition and international child health. He was part of a team that set the international research priorities to reduce global mortality from childhood diarrhea and a respected international consultant on the use of culturally acceptable, simple, and inexpensive foods and fluids to treat diarrhea and malnutrition in community settings. He published extensively, and was a contributing author to several important textbooks in his field. His passionate advocacy for children was also realized in his work with the Centers for Disease Control and Prevention and the World Health Organization. Dr Snyder’s legacy of clinical and research expertise, and his commitment to his patients and to teaching the next generation will continue to inspire his colleagues for many years to come.
  • The authors thank Dr Sona Long for her help in researching and evaluating the information in the section on hematology.
  • Dr Guandalini helped with conceptualization and design of the analyses, performed the literature review and the primary data analysis on hematologic issues, was an active and voting participant in all of the deliberations related to each topic, and helped to draft the initial manuscript;
  • Dr Liu helped with conceptualization and design of the analyses, performed the literature review and the primary data analysis on associated endocrine problems, was an active and voting participant in all of the deliberations related to each topic, and helped to draft the initial manuscript;
  • Dr Newton helped with conceptualization and design of the analyses, performed the literature review and the primary data analysis on nutritional issues, was an active and voting participant in all of the deliberations related to each topic, and helped to draft the initial manuscript; and all authors approved the final manuscript as submitted.


  • ALT: alanine aminotransferase
  • AST: aspartate aminotransferase
  • CBC: complete blood cell
  • CD: celiac disease
  • ESPGHAN: European Society for Pediatric Gastroenter- ology, Hepatology and Nutrition
  • GFD: gluten-free diet Ig: immunoglobulin MNT: medical nutritional therapy NASPGHAN: North American Society for Pediatric Gastroenterology, Hepatology and Nutrition
  • T1DM: type 1 diabetes mellitus
  • tTG: tissue transglutaminase to draft the initial manuscript

PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275). Copyright © 2016 by the American Academy of Pediatrics
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: Supported by an unrestricted grant from the Children’s National Health System’s Celiac Disease Program.
POTENTIAL CONFLICT OF INTEREST: Dr Fasano receives payment for lectures including service on speakers bureaus for Mead Johnson Nutrition, has stock/stock options with Alba Therapeutics, and is a consultant for Pfizer; the other authors have indicated they have no potential conflicts of interest to disclose.
COMPANION PAPER: A companion to this article can be found online at www.pediatrics.org/cgi/doi/10.1542/peds.2016-1311.
The guidelines/recommendations in this article are not American Academy of Pediatrics policy, and publication herein does not imply endorsement.

1Evidence-Informed Expert Recommendations for the Management of Celiac Disease in Children

  • Evidence-Informed Expert Recommendations for the Management of Celiac Disease in Children
  • John Snyder, J. Decker Butzner, Amy R. DeFelice, Alessio Fasano, Stefano Guandalini, Edwin Liu and Kimberly P. Newton
  • Pediatrics 2016;138;; originally published online August 26, 2016; DOI: 10.1542/peds.2015-3147
  • The online version of this article, along with updated information and services, is located on the World Wide Web at: /content/138/3/e20153147.full.html
  • PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly publication, it has been published continuously since 1948. PEDIATRICS is owned, published, and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois, 60007. Copyright © 2016 by the American Academy of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275

Evidence was assessed by using standardized criteria for evaluating the quality of the data, grade of evidence, and strength of conclusions. Over 600 publications were reviewed, and 172 were chosen for inclusion. The thorough review of the results demonstrated that the quality of the data available was often insufficient to provide unequivocal best practices. However, using the available data and the clinical experience of the panel, a practical framework for the management of children with CD was created. These recommendations were developed by our expert panel and do not necessarily reflect the policy of the American Academy of Pediatrics. The potential usefulness of these best practices is underscored by the fact that consensus, measured by the outcome of anonymous voting, was reached by the panel for 24 of the 25 questions. We hope that these best practices may be useful to the pediatric gastroenterology and larger general pediatric communities.

3 METHODS: Six acknowledged experts in pediatric CD were chosen to provide a thorough assessment of the data and to develop best practices in 6 topic areas: bone disease, endocrine problems, hematologic issues, liver issues, nutritional problems, and testing to monitor CD activity. The assignment of 1 panel member to each topic area was made by the organizer and moderator (Dr Snyder) based on that person’s recognized knowledge of the topic and was made in consultation with North American senior experts in pediatric gastroenterology who were not part of the panel. Each panel member researched and summarized their topic area. The panel met and thoroughly reviewed each topic before evaluating and voting on each best practice. In addition to their expertise, the panel members were chosen to provide a geographic representation of the major pediatric CD programs in the United States and Canada. The number of experts chosen was also influenced by logistic and financial considerations because the project was funded by a nonrestricted grant from a nonprofit organization, the Celiac Disease Program of Children’s National Health System. The meeting was convened at Children’s National Health Center in Washington, DC, on January 25 and 26, 2013. The goal was to provide a critical review of the management of pediatric CD in North America and to develop a practical set of best practices for the Children’s National Health System Celiac Program by using evidence-based data and expert opinion.

4  Literature Search and Grading the Articles for Quality of Evidence: Each expert completed a thorough literature search combining the term “celiac disease” with multiple terms specific to their section using accessible databases including PubMed, Medline, Embase, Cochrane Library, BioSciences Information Services Previews, EBM Reviews, ISI Web of Science, and Scopus. The search included publications from 1973 to January 2013 and included publications of all types that presented or reviewed data on CD in patients younger than 20 years old. Publications were assessed by using criteria including study design, sample size, data analysis, synthesis of results, potential bias, and limitations. The search identified over 600 unique publications. Of these 600 articles, 172 were included after exclusion of publications that did not present relevant evidence, that did not present sufficient evidence for pediatric patients, or were commentaries, case reports, abstracts, or nonsystematic reviews. Reviews of the literature were used to find additional primary research references and to provide summaries of data, references for pathophysiology, and to point our team to more extensive background reading. Reviews were also used to support introductory statements. The review articles and guidelines are identified in the References. Only original clinical studies were used to develop the “Best Practice” management questions.

Voting on Best Practice Statements and Grading the Statements for Quality of Evidence:  Details about mechanism for anonymous voting, assessment of quality of data, use of the Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) system to evaluate the available evidence, and strength of recommendation are included in the Supplemental Information.