INTRODUCTION — Celiac disease, also known as gluten-sensitive enteropathy or celiac sprue, is a common immune-mediated inflammatory disease of the small intestine caused by sensitivity to dietary gluten and related proteins in genetically predisposed individuals. It differs from food allergies, which are mediated by immunoglobulin E (IgE) or immunoglobulin G (IgG). A strict gluten-free diet is recommended for both diagnostic and therapeutic purposes in individuals with a provisional diagnosis of celiac disease. If celiac disease is confirmed, the diet should be continued for life. A strict gluten-free diet results in complete resolution of symptoms and mucosal healing for most individuals and also reduces the risks for long-term adverse health outcomes including intestinal lymphoma.

The diagnosis of celiac disease in children is discussed here. Other aspects of celiac disease in children are discussed in the following topic reviews:

●(See "Epidemiology, pathogenesis, and clinical manifestations of celiac disease in children".)

●(See "Management of celiac disease in children".)

The discussion below is consistent with guidelines developed by the European Society for Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) [1] and the recommendations in the clinical report from the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition (NASPGHAN) [2], which is available on the NASPGHAN website. Similar guidelines for the diagnosis and management of celiac disease have been issued by the American Gastroenterological Association (AGA) [3], the American College of Gastroenterology (ACG) [4], and the National Institutes of Health (NIH) [4,5].

WHOM TO TEST — Screening for celiac disease is recommended for patients with suggestive symptoms and also for children in groups at increased risk for having the disease, regardless of symptoms (table 1).

Screening of asymptomatic patients who do not have risk factors is not generally recommended. In theory, such a strategy might permit recognition and correction of subclinical nutritional deficiency states, resolution of mild symptoms, and potentially decrease the risk for malignancy. However, there are few data to support these beneficial effects [6], and the strategy would require many asymptomatic individuals to adhere to a difficult dietary regimen.

We recommend that serologic screening for celiac disease be performed in the following groups of children, provided that they are on a gluten-containing diet [1,4,7]:

Patients with symptoms suggesting celiac disease — We suggest screening patients with any symptoms and signs suggestive of celiac disease, if not otherwise explained. These may include classical gastrointestinal symptoms, nonspecific behavioral symptoms, and/or extraintestinal manifestations.

●Gastrointestinal symptoms are the most common classical manifestations of celiac disease (see "Epidemiology, pathogenesis, and clinical manifestations of celiac disease in children", section on '"Classical" gastrointestinal symptoms'). These include:

•Persistent diarrhea

•Abdominal bloating/distension

•Poor weight gain/weight loss

•Abdominal pain

•Constipation

•Vomiting

●Non-gastrointestinal manifestations may also be seen in patients with or without classical findings (table 2A-B) (see "Epidemiology, pathogenesis, and clinical manifestations of celiac disease in children", section on 'Non-gastrointestinal manifestations'). These include:

•Neurologic and behavioral symptoms

•Arthritis or arthralgias (even if there is a presumed underlying rheumatic disease)

•Abnormal liver biochemical tests

•Chronic fatigue

•Short stature or low height velocity

•Pubertal delay

•Iron deficiency anemia often poorly responsive to supplementation

•Dermatitis herpetiformis-like rash (picture 1A-B)

•Dental enamel hypoplasia of permanent teeth (symmetric distribution)

•Recurrent aphthous stomatitis

•Osteopenia, or fracture not explained by the level of trauma

Members of high-risk groups — We suggest screening all members of the following high-risk groups (table 1), regardless of symptoms (see "Epidemiology, pathogenesis, and clinical manifestations of celiac disease in children", section on 'High-risk groups'):

●First-degree relatives of patients with celiac disease

●Autoimmune thyroiditis

●Type 1 diabetes

●Autoimmune liver disease

●Down syndrome

●Selective immunoglobulin A (IgA) deficiency

●Turner syndrome

●Williams syndrome

●Juvenile chronic arthritis

If screening is undertaken for asymptomatic individuals in these high-risk groups, testing should first be performed at three years of age or older and on a gluten-containing diet for at least one year. If initial results are negative, screening tests should be repeated at intervals or at any time that symptoms develop. The optimal time interval for subsequent testing has not been studied, but in our practice, we screen asymptomatic members of these groups every three to five years during childhood. (See 'Asymptomatic members of a high-risk group' below.)

This suggestion of screening for asymptomatic individuals in these high-risk groups is consistent with several pediatric guidelines [1,2,7] but differs from guidelines used for adults in the United States, in which the recommendation is equivocal [4]. This difference in recommendations reflects a debate about the utility of screening for celiac disease among truly asymptomatic individuals belonging to a high-risk group because the benefit of treating such individuals has not been proven. Guidelines from the United Kingdom encourage testing for the first three of these high-risk groups and suggest consideration of testing for the other groups on this list, as well as for individuals with a variety of nonspecific symptoms [8]. (See "Diagnosis of celiac disease in adults", section on 'Who should be tested'.)

The debate continues about whether asymptomatic individuals in these high-risk groups should be screened, and recommendations may change as new information arises about the potential risks and benefits of screening. As an example, a study of a celiac screening program for children with type 1 diabetes mellitus compared clinical characteristics of 71 children with asymptomatic celiac disease with matched controls [9]. The children with celiac disease were slightly thinner (as indicated by a lower body mass index [BMI] Z-score), but height, bone mineral density, and diabetes control were similar. Thus, it is reasonable to question the need for celiac screening for patients with type 1 diabetes and no symptoms of celiac disease, and to make treatment decisions on a case-by-case basis, based on a discussion of estimated risks, symptoms, and treatment burden. (See "Associated autoimmune diseases in children and adolescents with type 1 diabetes mellitus".)

HOW TO TEST — The first step is to perform serologic testing. The serology results and level of clinical suspicion determine whether to move on to endoscopic biopsy (algorithm 1). Patients with positive results on a specific antibody test and characteristic histologic changes of the intestinal mucosa while on a gluten-containing diet are given a provisional diagnosis of celiac disease. The diagnosis is confirmed if there is normalization of antibodies and resolution of symptoms (if any) when the patient is on a gluten-free diet.

Antibody testing — Serologic tests for celiac disease are useful for screening and are an important first step in the diagnosis of the disease.

Tissue transglutaminase — For most patients, the most valuable test is for IgA antibodies against tissue transglutaminase (tTG-IgA) [4,10]. This test is highly sensitive, specific, and more cost-effective than other antibody tests (table 3). However, false-positive and false-negative results may still occur with some frequency in populations with a low risk for celiac disease. Testing should be performed while on a gluten-containing diet. Individuals previously started on a gluten-free diet without prior testing should resume a diet containing ideally at least 3 g gluten/day (equivalent to approximately one slice of bread daily) for at least six weeks before undergoing antibody testing, although the duration and amount of gluten required for diagnostic accuracy has not been fully clarified [11]. Serologies may become negative within weeks of beginning a gluten-free diet but may take up to a year or even longer in some cases [4,12].

The sensitivity and specificity of tTG-IgA antibodies for biopsy-proven celiac disease are generally above 96 percent, using second-generation enzyme-linked immunosorbent assays (ELISA) technology [2,13,14]. Sensitivities are reported to be somewhat lower in children younger than two years [15,16] (see 'Children younger than two years' below). Rapid office-based test kits have somewhat lower sensitivity and specificity, so a laboratory-based assay is recommended in patients with positive results of the rapid test, or those with negative results of the rapid test but strong clinical suspicion of celiac disease [17,18]. Case reports suggest that tTG-IgA antibodies may be falsely elevated during a febrile illness [19].

Other tests — Other serologic tests that are sometimes used to support the diagnosis of celiac disease include:

●Anti-endomysial antibodies – Testing for anti-endomysial antibodies (EMA) is as accurate as tTG-IgA, but this test is more expensive and somewhat dependent on operator interpretation. As a result, EMA is typically used as a second-line test to clarify the diagnosis in patients with equivocal results of tTG-IgA, including asymptomatic members of a high-risk group, as described below [10,20]. EMA is an immunofluorescence test for IgA antibodies to endomysium, a structure of the smooth muscle connective tissue [21]. (See 'Asymptomatic members of a high-risk group' below.)

●Deamidated gliadin peptide – Deamidated gliadin peptide (DGP) also has good diagnostic accuracy and may be particularly useful for young children; this is a second-generation anti-gliadin antibody test [22-25]. (See 'Children younger than two years' below.)

●Anti-reticulin antibodies – Tests of anti-reticulin antibodies have reasonably high specificity, but lower sensitivity, and are no longer commonly used [26].

●Anti-gliadin antibodies – Standard (first-generation) IgA or IgG anti-gliadin antibodies are considerably less reliable and are not recommended [4,27,28].

(See "Diagnosis of celiac disease in adults", section on 'Serologic evaluation'.)

Special populations — Variations in the diagnostic approach outlined above are appropriate for the following populations:

Asymptomatic members of a high-risk group — For children at increased risk for celiac disease (eg, those with Down syndrome, type 1 diabetes, or with a first-degree relative with celiac disease) who are asymptomatic, we use a slightly different approach (algorithm 1). We initiate screening around three years of age and repeat the screening every three to five years during childhood. (See 'Members of high-risk groups' above.)

●If the results of tTG-IgA antibodies are strongly positive (>3 times the upper limit of normal [ULN]), we proceed directly to endoscopic biopsy.

●If tTG-IgA antibodies are only mildly elevated (levels that are positive but <3 times the ULN), we suggest sequential testing with EMA antibodies. This practice is based on expert opinion, with some support from a study showing that this approach helps to distinguish the patients who are likely to have celiac disease [29].

•If the result of the EMA test is positive, the diagnosis of celiac disease is likely, and we suggest proceeding to endoscopy to confirm the diagnosis. (See 'Intestinal biopsy' below.)

•If the result of the EMA test is negative, it is reasonable to observe the patient and defer doing a biopsy because the chances of finding the characteristic histologic changes of celiac disease in these cases are low. The decision to observe versus proceed to intestinal biopsy is made on a case-by-case basis, depending on the reasons that the patient underwent screening and the preference of the patient and family. Those who choose to not have a biopsy should be followed closely and undergo repeat serologic testing if they ever develop symptoms that are suggestive of celiac disease.

●If tTG-IgA antibodies are negative, it is reasonable to observe the patient and repeat the test if symptoms develop.

An alternative strategy for asymptomatic individuals in these high-risk groups, and especially for those with Down syndrome, is to assess for the human leukocyte antigen (HLA) types DQ2 and DQ8 as the initial screening test [1,30,31]. Those who test positive for HLA-DQ2 or DQ8 should then undergo serial screening for celiac-associated antibodies every three to five years as described above. Those who test negative for HLA-DQ2 and DQ8 do not require additional screening, because they have a very low risk for developing celiac disease. Although HLA testing is substantially more costly than serologic screening for celiac disease, it does not require serial testing. The cost-effectiveness of this strategy has not been established and depends on the pre-test risk for celiac disease in the population tested, including the likelihood of celiac disease susceptibility alleles [32]. This strategy is most appropriate for individuals with Down syndrome, in whom the likelihood of a celiac risk allele is similar to that of the general population [4]. It is not likely to be useful for members of other high-risk groups (eg, first-degree relatives of patients with celiac disease), because a high proportion of individuals in these groups will carry HLA susceptibility alleles, so HLA testing is unlikely to be useful to exclude celiac disease [33]. (See "Epidemiology, pathogenesis, and clinical manifestations of celiac disease in children", section on 'Genetic factors'.)

Children younger than two years — For children younger than two years, tTG-IgA is the best initial screening test for IgA-sufficient children, as it is for older individuals, with >97 percent sensitivity for detecting celiac disease [34,35]. Total IgA should be measured concurrently to exclude IgA deficiency. If the tTG-IgA is negative and there is a high clinical suspicion for celiac disease (or if the patient is IgA-deficient), we suggest sequential testing with DGP-IgG to detect the few patients with positive DGP-IgG but normal tTG-IgA [1,4]. Although early studies raised the possibility that DGP-IgG may be more sensitive than tTG-IgA in this age group [36,37], subsequent studies indicate that tTG-IgA is at least as sensitive as DGP-IgG (except for patients with IgA deficiency) [34,35].

Immunoglobulin A deficiency — For individuals with known selective IgA deficiency, testing should be performed with IgG antibodies to tTG (tTG-IgG) instead of the usual IgA-based antibody test [4]. Because tTG-IgG has limited specificity, concurrent measurement of a second IgG-based antibody, such as DGP, is suggested [1]. If one or both of these IgG-based tests is elevated, the patient should be further evaluated with an intestinal biopsy; a non-biopsy diagnosis is not appropriate for this group of patients. (See 'Is biopsy necessary for all patients?' below.)

Approximately 2 percent of children with celiac disease will have previously unrecognized IgA deficiency. Although European guidelines recommend routine measurement of serum total IgA in all patients, we think a more cost-effective population approach is to measure total IgA in those children with negative results of tTG-IgA but a high clinical suspicion of celiac disease. For all other children, the yield is very low. In one report of population screening in Sweden, measurement of total IgA contributed to the diagnosis of celiac disease in only 2 of 7208 children screened [38]. (See "Selective IgA deficiency: Clinical manifestations, pathophysiology, and diagnosis", section on 'Evaluation and diagnosis'.)

Patients already on a gluten-free diet — For patients who are already on a gluten-free diet, we suggest serologic testing for tTG-IgA and/or EMA. If these are elevated, an intestinal biopsy should be obtained to confirm the diagnosis of celiac disease. If these antibodies are not elevated, celiac disease is not excluded. In this case, we suggest HLA-DQ2/DQ8 testing to determine whether the patient is genetically susceptible to celiac disease [4,10]. If these results are negative, celiac disease is essentially excluded (see "Epidemiology, pathogenesis, and clinical manifestations of celiac disease in children", section on 'Genetic factors'). If the patient tests positive for HLA-DQ2/DQ8, then a gluten challenge should be performed by reintroducing moderate amounts of wheat into the diet. The patient should be retested for tTG-IgA when symptoms recur or after six months if the patient remains asymptomatic [30]. (See "Diagnosis of celiac disease in adults", section on 'Patients on a gluten-free diet'.)

Dermatitis herpetiformis — Patients with dermatitis herpetiformis established by skin biopsy are presumed to have celiac disease and can be treated with a gluten-free diet without other diagnostic studies. However, a baseline measurement of tTG-IgA is valuable to monitor improvement after institution of a gluten-free diet.

Intestinal biopsy — Individuals with positive tTG-IgA or EMA antibodies should undergo an intestinal biopsy to establish the diagnosis of celiac disease [4]. A possible exception is selected patients with very high results of serologic tests and classic symptoms of celiac disease, as discussed in the next section. The biopsy should be performed with the patient on a gluten-containing diet. Multiple biopsies should be taken (four from the distal duodenum and at least one from the duodenal bulb) because the disease may have a patchy distribution or initially be confined to the duodenal bulb [4,39-43]. Endoscopy and biopsy are also appropriate for patients with negative serologic testing if there is a strong clinical suspicion of celiac disease [4]. Capsule endoscopy should not be used as a substitute for endoscopy for the initial diagnosis of celiac disease, because its diagnostic yield for this disorder is much lower [4].

Is biopsy necessary for all patients? — Although intestinal biopsy is valuable for any patient to achieve maximal diagnostic certainty, it is considered optional for selected patients.

●Criteria for non-biopsy diagnosis – Diagnosis without an intestinal biopsy is an option for patients with all of the following characteristics, as outlined in the European pediatric guidelines [1,31,44]:

•Very high results of tTG testing (tTG-IgA >10 times the ULN)

•Positive results of an EMA-IgA from a second sample of blood

•Symptoms compatible with celiac disease

If patients with these characteristics choose to forgo intestinal biopsy, the diagnosis is only confirmed if there is a clear clinical response to a gluten-free diet, with normalization of the tTG titers and improvement of symptoms.

●Caveats

•Asymptomatic patients – For asymptomatic children who meet the above laboratory criteria (tTG-IgA >10 times the ULN and positive EMA-IgA), the European guidelines suggest that a non-biopsy diagnosis may also be considered [1]. However, the positive predictive value of a tTG-IgA >10 times the ULN may be lower in asymptomatic than symptomatic children; therefore, the decision to forgo a biopsy in these cases requires careful consideration and should involve the parents in a shared decision-making process.

•IgA deficiency – Patients with IgA deficiency should always have a biopsy-based diagnosis, even if their tTG-IgG antibody test is very high, because IgG-based antibody tests have not been sufficiently validated as biomarkers of disease [45]. Moreover, tTG-IgG is not a reliable marker of response to a gluten-free diet.

●Evidence – Support for a non-biopsy diagnosis in selected patients comes from three large case series, which reported that more than 98 percent of symptomatic patients with very high tTG-IgA and positive EMA had intestinal biopsy results consistent with celiac disease [29,46,47]. In one of these studies, the combination of very high tTG-IgA and positive EMA had sensitivity, specificity, and positive predictive value of 69.3, 96.8, and 99.6 percent, respectively [46]. If malabsorption symptoms (diarrhea, weight loss, poor weight gain, or iron deficiency anemia) were included, specificity and positive predictive value rose to 100 percent. The need to include the second antibody test in this diagnostic approach has been questioned since adding EMA testing does not improve the positive predictive value [48]. HLA typing has also been suggested to add strength to the diagnosis (because celiac disease would be effectively excluded in any patient that does not have HLA types DQ2 or DQ8) but is no longer required, because the additional diagnostic value of HLA typing in this setting appears to be small [1,46].

●Advantages of a biopsy-based diagnosis – It is important to recognize the potential value of including intestinal biopsy as part of the diagnostic process for all patients, including those with very high tTG antibodies. Histologic confirmation provides for maximal diagnostic certainty, which is important because of the lifelong burden of a gluten-free diet. In particular, the endoscopy permits evaluation for concurrent mucosal inflammation other than celiac disease, including eosinophilic esophagitis, Helicobacter pylori, and Crohn disease of the proximal small intestine [49-51]. Moreover, despite the high sensitivity estimates reported above, occasional false-positive results occur [29,46,47]. Finally, there are concerns about the quantitative variability and lack of standardization between commercially-available serologic tests for celiac disease. These considerations should be discussed with the patient and family before deciding whether to proceed with intestinal biopsy to confirm the diagnosis before embarking on a gluten-free diet.

Abnormal findings — The intestinal biopsies should be interpreted by an expert pathologist. The histologic features of celiac disease are usually described using the Marsh classification (table 4 and figure 1); they range from a mild alteration characterized only by increased intraepithelial lymphocytes (Marsh type 1 lesion), to a flat mucosa with total mucosal atrophy, complete loss of villi, enhanced epithelial apoptosis, and crypt hyperplasia (Marsh type 3 lesion) (picture 2) [52-57]. The Marsh type 4 lesion has the same histologic features seen in the type 3 lesion except that the crypts are hypoplastic. A gradient of decreasing severity from the proximal to the distal small intestine is often observed, correlating with the higher proximal concentration of dietary gluten. The degree of the villous atrophy does not necessarily correlate with the severity of clinical symptoms, and sampling errors can occur due to some inhomogeneity of mucosal inflammation.

Patients with these abnormal findings (Marsh type 2 or higher) are given a provisional diagnosis of celiac disease and should have a trial of treatment with a gluten-free diet (see 'Gluten-free diet trial' below). Rarely, other disorders can cause similar histologic findings (table 5). For patients with equivocal findings (eg, Marsh type 1 lesions), management decisions depend on the level of suspicion for celiac disease, including type and severity of symptoms and results of celiac-specific antibody tests.

Normal findings — Patients with positive tests for tTG-IgA or EMA but normal results of small bowel biopsies are considered to have potential celiac disease (table 6) (see "Epidemiology, pathogenesis, and clinical manifestations of celiac disease in children", section on 'Potential celiac disease'). It is important that the evaluation of such patients include expert review of multiple intestinal biopsies since the histologic abnormalities can be patchy. In some cases in which the diagnosis of celiac disease is unclear, testing for HLA-DQ2/DQ8 may be helpful because if the result is negative, this essentially excludes the possibility of celiac disease [4].

In general, we suggest not treating such patients with a gluten-free diet if they do not have symptoms. However, these patients should be carefully monitored for growth failure and other symptoms that might suggest active celiac disease and should be rebiopsied if symptoms develop. Decisions about whether to begin a gluten-free diet for patients with positive serologic tests but normal biopsy results should be made on a case-by-case basis with the family, after consideration of the patient's level of symptoms, appropriate exclusion of other causes of the symptoms, the burden of maintaining a gluten-free diet, and the adequacy of the biopsied tissue samples.

These decisions are informed by the following considerations, which take into account a patient's level of symptoms and possibly age or other risk factors that are predictors of the risk of developing overt celiac disease:

●Symptomatic children with positive serologic tests for celiac disease but apparently normal biopsies are likely to have celiac disease, based on limited evidence. This was suggested by a small series of eight children with some gastrointestinal symptoms and positive serologic tests (EMA) but normal intestinal histology who continued to consume a gluten-containing diet. Within two years, seven of these eight children had developed marked mucosal atrophy and were diagnosed with celiac disease [58]. These observations suggest that a substantial number of symptomatic children with positive serologic tests but normal biopsies will develop celiac disease within a few years, while in a smaller number, the antibodies disappear completely.

●By contrast, the majority of asymptomatic children with positive serologic tests and normal biopsy results will not develop celiac disease during childhood [59,60]. In one study, two-thirds of children with potential celiac disease at baseline still had normal biopsies at nine years of follow-up [60]. Such patients usually are identified during routine testing performed because they are members of a group at increased risk for celiac disease (eg, first-degree relatives of individuals with celiac disease). The natural history of these patients is described separately. (See "Epidemiology, pathogenesis, and clinical manifestations of celiac disease in children", section on 'Potential celiac disease'.)

Gluten-free diet trial — A trial of a gluten-free diet is recommended for both diagnostic and therapeutic purposes for all children in one of the following groups:

●Children with characteristic findings on intestinal biopsy and symptoms consistent with celiac disease (including nonspecific symptoms such as constipation or abdominal pain).

●Children with characteristic findings on intestinal biopsy and belonging to one of the above high-risk groups (eg, relatives of patients with established celiac disease or patients with type 1 diabetes (see 'Members of high-risk groups' above)), whether or not there are associated symptoms.

Treatment of individuals with confirmed celiac disease consists of a lifelong gluten-free diet, which improves gastrointestinal symptoms as well as most of the non-gastrointestinal symptoms of celiac disease and may reduce the risk of gastrointestinal malignancies and other long-term adverse health consequences. Patients with dermatitis herpetiformis confirmed by skin biopsy also should be treated with a gluten-free diet. Details of the gluten-free diet and monitoring are discussed separately. (See "Management of celiac disease in children".)

We do not recommend beginning a gluten-free diet before confirming the diagnosis, because the symptoms of celiac disease are nonspecific and a confirmed diagnosis of celiac disease mandates following a strict gluten-free diet for life. The gluten-free diet is burdensome, is not easy to follow, is associated with increased costs, and has far-reaching implications for the child's quality of life. For all of these reasons, a gluten-free diet should not be instituted unless there is good evidence for celiac disease. A diagnostic approach for patients who are already on a gluten-free diet is outlined above. (See 'Special populations' above.)

DIAGNOSIS — A provisional diagnosis of celiac disease is made if the results of celiac-specific antibody testing and the intestinal biopsy are both abnormal. The diagnosis is confirmed when symptoms resolve subsequently on a gluten-free diet (algorithm 1). Serologic tests that revert from positive to negative on a gluten-free diet may be used as supportive evidence of the diagnosis and are particularly valuable in individuals with minimal symptoms [61]. Patients who do not demonstrate a robust serologic response to a gluten-free diet, or whose initial endoscopic biopsies are not typical for celiac disease, may require a second endoscopy with biopsy while on a gluten-free diet to clarify the diagnosis.

Note that at least one endoscopy with intestinal biopsies is required to establish the diagnosis for most patients; a serologic response alone to a gluten-free diet is not sufficient (see 'Intestinal biopsy' above). Similarly, symptom response to a gluten-free diet alone is not sufficient to diagnose celiac disease, because these features do not distinguish between celiac disease and "non-celiac gluten sensitivity," which is defined as dose-dependent intolerance to gluten-containing grains after exclusion of celiac disease or wheat allergy [2,11]. (See "Epidemiology, pathogenesis, and clinical manifestations of celiac disease in children", section on 'Non-celiac gluten sensitivity' and "Diagnosis of celiac disease in adults", section on 'Discordant serology and small bowel biopsy'.)

European guidelines offer the option of an alternate approach to diagnosis for selected patients with very high results of serologic tests (tissue transglutaminase antibody [tTG-IgA] >10 times the upper limit of normal [ULN]) and a positive anti-endomysial antibody (EMA) obtained on a separate serum sample [1]. This diagnostic approach is more reliable in patients with typical symptoms of celiac disease. The diagnosis is confirmed if the serologic tests revert to normal and symptoms resolve on a gluten-free diet. Considerations relevant to the diagnostic approach for this group of patients are discussed above. (See 'Is biopsy necessary for all patients?' above.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Celiac disease" and "Society guideline links: Dermatitis herpetiformis".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: Celiac disease (The Basics)")

●Beyond the Basics topic (see "Patient education: Celiac disease in children (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Serologic testing for celiac disease should be performed for children who present with findings consistent with classic celiac disease, which include persistent diarrhea, chronic constipation, recurrent abdominal pain, and failure to thrive, and for those who present with dermatitis herpetiformis-type rash (picture 1A-B). Testing is also appropriate for children who present with other associated non-gastrointestinal symptoms (table 2A), in whom no other underlying cause can be identified. (See 'Patients with symptoms suggesting celiac disease' above.)

●We also suggest screening asymptomatic individuals with the following disorders, which are associated with an increased risk for celiac disease: first-degree relatives of patients with celiac disease, autoimmune thyroiditis, type 1 diabetes, Down syndrome, Turner syndrome, Williams syndrome, and selective IgA deficiency (Grade 2C). (See 'Members of high-risk groups' above.)

●A gluten-free diet should not be started prior to a full evaluation for celiac disease. This is because the symptoms of celiac disease are nonspecific, and both antibody testing and intestinal biopsy results are only valid if they are performed while the patient is exposed to gluten. (See 'Gluten-free diet trial' above and "Diagnosis of celiac disease in adults", section on 'Patients on a gluten-free diet'.)

●Serologic screening is performed by measuring one of several antibodies that are specific for celiac disease, with the patient on a gluten-containing diet (table 3). The most clinically useful test is for IgA antibodies against tissue transglutaminase (tTG-IgA), which is highly sensitive, specific, and more cost-effective than other antibody tests. (See 'Antibody testing' above.)

●Most patients with positive results of the serologic screen should undergo endoscopy, with biopsies from several areas of the duodenum including the duodenal bulb (algorithm 1). The histologic features of celiac disease range from a mild alteration characterized only by increased intraepithelial lymphocytes (Marsh type 1 lesion) to a flat mucosa with total mucosal atrophy, complete loss of villi, enhanced epithelial apoptosis, and crypt hyperplasia (Marsh type 3 lesion) (figure 1 and picture 2). (See 'Intestinal biopsy' above.)

●Intestinal biopsy is valuable for all patients to achieve maximal diagnostic certainty and exclude other gastrointestinal diseases (eg, eosinophilic esophagitis). However, the biopsy is considered optional for selected patients with very high levels of tTG-IgA, positive anti-endomysial antibodies (EMA), and typical symptoms. For patients with these characteristics, the decision about whether to perform endoscopy and biopsy should be made collaboratively with the patient and family. (See 'Is biopsy necessary for all patients?' above.)

●A provisional diagnosis of celiac disease is made if the results of celiac-specific antibody testing and the intestinal biopsy are both abnormal. Such patients should be treated with a gluten-free diet and monitored for changes in symptoms (including growth parameters). The diagnosis is confirmed when symptoms resolve subsequently on a gluten-free diet. Serologic tests that revert from positive to negative on a gluten-free diet may be used as supportive evidence of the diagnosis and are particularly valuable in individuals with minimal symptoms. (See 'Gluten-free diet trial' above and 'Diagnosis' above.)

●Patients with positive results of the serologic screen but normal intestinal biopsies (sometimes classified as "potential celiac disease") may not require treatment with a gluten-free diet if they are asymptomatic. However, because some of these patients will go on to develop intestinal damage and symptoms, they should be monitored closely and treatment should be considered if they develop symptoms. In patients who have symptoms but a normal intestinal biopsy, it is also reasonable to consider treatment with a gluten-free diet, after exclusion of other conditions and careful consideration of the potential benefits and treatment burden with the family. (See 'Normal findings' above and "Epidemiology, pathogenesis, and clinical manifestations of celiac disease in children", section on 'Potential celiac disease'.)

●Implementation of a gluten-free diet and other details of treatment and monitoring of children with established celiac disease are discussed in a separate topic review. (See "Management of celiac disease in children".)