Celiac Disease

Celiac disease (CD), or gluten-sensitive enteropathy, is a nonallergic, autoimmune-mediated sensitivity to gluten in genetically susceptible individuals.  Initial testing for CD typically includes assessment of the serum immunoglobulin A (IgA) level, followed by the appropriate tissue transglutaminase (tTG) antibody test (IgA or IgG, depending on whether or not the patient is IgA deficient).    Duodenal biopsy remains the gold standard for diagnostic confirmation of CD, and HLA-DQ2 and HLA-DQ8 genotyping may be essential in risk estimation or disease exclusion. CD risk varies among ethnicities and geographic regions, suggesting that environmental and/or lifestyle risk factors may play a role in disease etiology. 

Quick Answers for Clinicians

Who should be tested for celiac disease?

Children and adolescents who have unexplained gastroenterologic symptoms, poor growth, delayed puberty, iron-deficiency anemia, and abnormal liver testing, or a specific risk factor for celiac disease (CD), even if asymptomatic, should be tested. 

Adults with unexplained gastrointestinal symptoms, unexplained iron deficiency, dermatitis herpetiformis, recurrent aphthous stomatitis, early-onset osteoporosis, delayed puberty/unexplained short stature, alopecia areata, or a risk factor for CD, even if asymptomatic, should be tested.    

See Indications for Testing for risk factors in children and adults.

Which are the primary tests used for celiac disease testing?

The tissue transglutaminase (tTG) immunoglobulin A (IgA) test is the recommended single screening test for celiac disease (CD) for most individuals.   However, individuals with CD have a higher chance of IgA deficiency, so serum evaluation of IgA is recommended as a first step.  IgA-deficient individuals must be tested for CD-specific IgG antibodies for an optimal diagnostic outcome. Endomysial antibody (EMA) IgA testing in those with a high probability of having CD increases the sensitivity of tTG testing. In children younger than 2 years with a high suspicion of CD, deamidated gliadin peptide (DGP) antibody testing should be used in conjunction with the tTG test.  Patients must be on a gluten-containing diet when undergoing serologic testing for CD. 

Duodenal biopsy is considered the gold standard test for CD diagnosis. When biopsy is not possible, or preferred, such as in children and adolescents, CD can be confirmed based on HLA-DQ2 or HLA-DQ8 positivity,  in the presence of strong serologic and clinical evidence.

When is it appropriate to use genetic (HLA) testing for suspected celiac disease?

HLA genotyping can be used to exclude a celiac disease (CD) diagnosis because the absence of HLA-DQ2 or HLA-DQ8 essentially rules out CD.   HLA testing is recommended in patients with strong serologic evidence and clinical suspicion for CD who prefer (or whose family members prefer, in the case of young children) that small bowel biopsy be avoided and in patients with negative CD-specific antibodies and an indeterminate proximal small intestinal biopsy. 

Which testing algorithms are related to this topic?
Celiac Disease Testing for Symptomatic Individuals Algorithm

Indications for Testing

Children and Adolescents

Unexplained gastroenterologic symptoms, poor growth, delayed puberty, iron-deficiency anemia, and abnormal liver testing in children or adolescents warrant testing for CD. Children and adolescents with a specific risk factor for CD should be tested as well, even if they are asymptomatic. Risk factors include autoimmune disorders (type 1 diabetes mellitus, autoimmune thyroiditis, autoimmune liver disease), syndromes associated with CD (Down, Turner, or Williams syndromes), selective IgA deficiency, and first-degree relatives with CD. 

Adults

Unexplained gastrointestinal symptoms, unexplained iron deficiency, dermatitis herpetiformis, recurrent aphthous stomatitis, early-onset osteoporosis, delayed puberty/unexplained short stature, or alopecia areata prompt CD testing in adults.     Adults with a risk factor for CD should also be tested, even if asymptomatic. Risk factors in adults include family history of CD, syndromes associated with CD (Down syndrome, Turner syndrome), autoimmune disease associated with CD (eg, type 1 diabetes mellitus, thyroid disease, inflammatory bowel disease [IBD]), or selective IgA deficiency. 

Criteria for Diagnosis

The European Society for Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN)  has suggested the following criteria for the diagnosis of CD:

  • Positive tTG IgA or endomysial antibody (EMA) IgA serologic test, plus biopsy consistent with CD, or
  • HLA-DQ2 or HLA-DQ8 positivity in the presence of CD symptoms and high levels of CD-specific antibodies (10 times the upper limit of normal)

The patient should be on a gluten-containing diet when undergoing biopsy or any other CD test.    For duodenal biopsy, the gold standard for diagnosis, five or more samples are preferred to increase probability, one or more from the bulb and four or more from the second and third parts of the duodenum.  Intestinal damage is assessed using the modified Marsh classification of histologic findings in CD  or the simplified system classification. 

The American College of Gastroenterology,  the World Gastroenterology Association,  and the American Gastroesophageal Association  have issued similar criteria for CD diagnosis.

Laboratory Testing

Diagnosis

IgA Deficiency Testing

Serum IgA level by nephelometry is the recommended initial test to determine IgA levels before antibody testing. If the patient is IgA deficient, all serologic testing should be performed using IgG tests to prevent false-negative antibody results. 

Infants may present with transient, suboptimal levels of IgA and/or IgG that may not be related to immune deficiency.

Antibody Testing Based on IgA Level

Undetectable levels of IgA are typically reported as <7 mg/dL. 

  • If the patient is not IgA deficient (ie, level is within age-matched range), order tTG IgA, the preferred test for CD.
  • If the IgA level is ≥7.0 mg/dL but below the age-matched range, consider a CD dual-antigen screen with reflex, which tests for both IgA and IgG.
  • If the IgA level is <7.0 mg/dL, order tTG IgG and deamidated gliadin peptide (DGP) IgG tests.

IgA deficiency may be accompanied by other immunoglobulin deficiencies. Consider evaluating the patient for immune deficiency if IgA levels are below the limit of detection. 

See the Antibody Test Results Interpretation table below for details about interpretation of the following specific antibody tests.

Tissue Transglutaminase Antibodies

The tTG IgA test is the recommended single screening test for IgA-competent individuals with possible CD, particularly in patients older than 2 years.  The higher the titer, the greater the likelihood that the result is a true positive.  Combining several tests, rather than using tTG IgA alone for CD diagnosis, may slightly increase sensitivity but reduces specificity, so this combined testing approach is not recommended for those at low risk for CD.  In addition, correlation among CD serologic tests may be variable at low antibody titers and in early disease. To rule out transient seropositivity, patients with low antibody titers but normal small-intestine mucosa should have serologic tests repeated in 6 months while continuing to consume gluten.  In patients with IgA deficiency, tTG IgG testing is recommended (in conjunction with DGP IgG testing).

Endomysial and Deamidated Gliadin Peptide Antibodies

EMA testing, which is highly specific for CD, may help differentiate between false-positive and true-positive tTG results.  However, EMA testing is expensive and relies on observer expertise.  In addition, sera containing antismooth muscle antibodies may interfere with the detection of EMA IgG.

The tTG IgA test along with the DGP IgA and IgG tests should be considered for screening in children younger than 2 years.  

Antibody Test Results Interpretation

Test Result Interpretation and/or Next Step
tTG IgA High positive (≥41 U/mL) CD is likely; consider HLA genotypinga or biopsy
Weak-moderate (4-40 U/mL) Order EMA IgA by IFA, DGP IgA, and HLA genotypinga
Negative (≤3 U/mL)

CD is unlikely; exclude history of gluten-free diet or immunosuppressive drug, which can cause false-negative results

Consider HLA genotypinga (in light of age and associated diseases)
EMA and/or DGP Negative in presence of HLA positivitya Perform biopsy
Positive in presence of HLA negativitya Likely false-negative HLA genotypinga test; consider biopsy to confirm or rule out CD
Positive in presence of HLA positivity CD confirmed
Negative in presence of HLA negativitya CD ruled out; any positive tTG test preceding this result was likely false positive
CD dual-antigen screen with reflex Normal CD unlikely
Positive or equivocal Consider HLA genotypinga
tTG IgG and DGP IgG Normal CD unlikely
Positive or equivocal Consider HLA genotypinga

aSee HLA Genotyping section below.

IFA, indirect fluorescent antibody

HLA Genotyping​

Associations of HLA-DQ2 and HLA-DQ8 with CD are among the strongest HLA-disease associations discovered. HLA-DQ2 is present in approximately 90% of patients with CD and is encoded by the HLA-DQA1*05 allele in combination with the HLA-DQB1*02 allele (also referred to as HLA-DQ2.5) in cis or trans arrangement, whereas the remaining 5-10% of patients with CD carry HLA-DQ8 encoded by the HLA-DBQ1*03:02 allele. The presence of HLA-DQ2 or HLA-DQ8 is necessary but not sufficient for the development of CD, given that these alleles are widely present in the general population (20-30% of the general population carry HLA-DQ2,   and 10-20% carry HLA-DQ8). Although the presence of these genes cannot confirm CD, their absence essentially excludes CD.

HLA genotyping is not necessary for routine laboratory evaluation of CD due to its low positive predictive value, but testing may be indicated in patients at risk for CD, individuals who are repeatedly seropositive but biopsy negative, and patients avoiding biopsy. This testing can also be useful in patients who have equivocal small bowel histologic findings (Marsh I-II) and are seronegative for CD. Biopsy may not be needed if tTG IgA results are >10 times the manufacturer’s cutoff and are confirmed by EMA testing in a different blood sample, and HLA testing is positive. 

HLA typing is useful for ruling out CD in individuals who belong to groups at risk for the disease, such as first-degree relatives of patients with confirmed CD, or individuals with other conditions known to be associated with CD, such as autoimmune diseases, type I diabetes, liver and thyroid disease, and diseases associated with chromosomal abnormalities, such as Williams, Turner, and Down syndromes.

Interpretation of HLA Testing for Celiac Disease
Result Interpretation

Positive: Both alleles (HLA-DQA1*05 and HLA-DQB1*02) of HLA-DQ2 heterodimer identified in a symptomatic individual

Supports CD diagnosis

Further confirmatory CD testing is recommended

Positive: HLA-DQ8 identified in a symptomatic individual

Supports CD diagnosis

Further confirmatory CD testing is recommended

Positive: 1 portion of the HLA-DQ2 heterodimer identified

Rarely observed in CD

If strong suspicion of CD exists, further CD testing may be helpful

Negative: No copy of HLA-DQ2 heterodimer or HLA-DQ8 identified

Diagnosis of CD is likely excluded; no further CD testing is recommendeda

aRare exceptions to these HLA associations have been occasionally observed.

Prognosis

Antiactin (F-Actin) IgA​

Antiactin (F-actin) IgA testing by enzyme-linked immunosorbent assay (ELISA) should be performed in biopsy-confirmed cases of CD and can provide information about prognosis. Levels correlate with severity of mucosal damage: The presence of antiactin antibodies may indicate intestinal villus atrophy and a more severe form of disease.  Although the F-actin test result may indicate moderate to severe disease, the test lacks specificity. 

Screening

There is insufficient evidence to recommend screening for CD in asymptomatic people.  Serologic screening for CD is recommended for specific at-risk groups, including first-degree relatives of patients with CD and individuals who have the following :

For asymptomatic children with the above risk factors, begin screening after 2 years of age and when the child has been consuming wheat for at least 1 year, or sooner if CD signs and symptoms manifest. 

Monitoring

Monitoring should be performed to assess therapeutic response to change in diet and patient compliance with diet. Monitoring tests include the same assays recommended for diagnosis.  tTG, DGP, or EMA assays can be used, depending on previous results, to monitor adherence to a gluten-free diet. In cases of IgA deficiency, IgG testing should be used. 

Celiac serologic testing, particularly tTG IgA and DGP IgA, is recommended every 3-6 months after initial diagnosis until abnormal baseline results have normalized or until patient is clinically stabilized, then every 1-2 years.  If antibody levels remain elevated after more than 12 months on a gluten-free diet, consider repeating biopsy. A decline in antibody levels, including F-actin IgA, may correlate with normalization of the intestinal villi. 

ARUP Lab Tests

Diagnosis

Preferred Reflex Screening Test

Components: Initial test is IgA; additional tests may include tTG IgA or IgG, EMA IgA, DGP IgA or IgG, and the CD dual-antigen screen

Preferred Single Screening Tests
Acceptable Single Screening Tests
HLA Genotyping Tests

Polymerase chain reaction (PCR) followed by melting curve analysis

IgA Deficiency Test
Other Tests

Prognosis

Medical Experts

Contributor

Delgado

Julio Delgado, MD, MS
Julio Delgado, MD, MS
Professor of Clinical Pathology, University of Utah
Chief, Division of Clinical Pathology, University of Utah and ARUP Laboratories
Chief Medical Officer and Director of Laboratories at ARUP Laboratories
Contributor
Contributor
Contributor

References

Additional Resources

  • 22253278

    Jenkins HR
    Murch SH
    Beattie RM
    et al.
    Coeliac Disease Working Group of British Society of Paediatric Gastroenterology, Hepatology and Nutrition

    Diagnosing coeliac disease.

    Arch Dis Child
    2012
    97
    5
    393-4
    PubMed
    • Resources from the ARUP Institute for Clinical and Experimental Pathology®

    Related Information