Which blood tests should be ordered to evaluate a patient with suspected celiac disease while they are still consuming a normal gluten‑containing diet?

Blood Tests for Celiac Disease

Order IgA tissue transglutaminase antibody (tTG-IgA) combined with total IgA level measurement as your first-line serologic screen for celiac disease. This combination represents the most efficient initial diagnostic approach and must be performed while the patient consumes at least 10g of gluten daily (approximately three slices of wheat bread) for 6–8 weeks prior to testing. 1, 2

Initial Screening Strategy

• The tTG-IgA test demonstrates excellent diagnostic performance: sensitivity of 90.7% and specificity of 87.4% in adults at a 15 U/mL threshold, and sensitivity of 97.7% with specificity of 70.2% in children at a 20 U/mL threshold. 1, 2

• Measuring total IgA level simultaneously is mandatory because IgA deficiency occurs in 1–3% of celiac patients and causes falsely negative results on all IgA-based antibody tests. 1, 2, 3

• The combination of tTG-IgA plus total IgA is identified as the single most efficient screening approach by major gastroenterology societies. 1, 2

Confirmatory Testing After Positive tTG-IgA

• When tTG-IgA is positive, order IgA endomysial antibody (EMA) as confirmatory testing because EMA provides superior specificity of 99.6% in adults and 93.8% in children. 1, 2

• If the tTG-IgA level exceeds 10 times the upper limit of normal and a repeat sample shows positive EMA, the positive predictive value for celiac disease approaches 100%. 1, 4

• This two-step approach (tTG-IgA followed by EMA) achieves sensitivity of 85.7%, specificity of 98.6%, positive predictive value of 71.7%, and negative predictive value of 99.7%. 5

Testing in Patients with IgA Deficiency

• If total IgA is low or absent, switch to IgG-based serology. Order IgG deamidated gliadin peptide (DGP-IgG) as the preferred test, which demonstrates sensitivity of 93.6% and specificity of 99.4% in adults. 1, 2

• IgG EMA or IgG tTG can also be used in IgA-deficient patients, but IgG DGP is preferred due to superior diagnostic accuracy. 1

• Do not order IgG tTG as a primary screening test when total IgA is normal—it lacks specificity in that setting and should only be used when IgA deficiency is confirmed. 1

Tests That Should NOT Be Ordered

The following tests are not recommended for celiac disease diagnosis and should be avoided: 6

• Stool studies for celiac antibodies
• Small-bowel follow-through imaging
• Intestinal permeability testing
• D-xylose testing
• Salivary antibody testing
• Non-deamidated IgA/IgG gliadin antibodies (older generation tests that confer no additional diagnostic benefit) 5

Critical Patient Preparation Requirements

• Patients must consume adequate gluten before testing: at least 10g of gluten daily for 6–8 weeks prior to blood draw to avoid false-negative results. 1, 2, 3

• If a patient has already started a gluten-free diet, they must resume regular gluten intake (e.g., three slices of wheat bread daily) for 1–3 months before repeat serologic testing. 1, 2

• Reducing or eliminating gluten before testing markedly lowers the sensitivity of both serologic assays and subsequent duodenal biopsy. 1, 2

Common Pitfalls to Avoid

• Never rely on serology alone for diagnosis in adults—duodenal biopsy remains mandatory for confirmation except in special circumstances such as coagulopathy or pregnancy. 2, 4, 3

• Do not start a gluten-free diet before completing the full diagnostic workup (serology plus biopsy), as this leads to false-negative results and inconclusive biopsies. 6, 2

• Always measure total IgA level alongside tTG-IgA—missing IgA deficiency is a frequent diagnostic error that results in false reassurance from negative IgA-based tests. 1, 2, 3

• Be aware that a normal tTG level does not predict recovery of villous atrophy in patients already on a gluten-free diet—44% of patients with persistent villous atrophy had normal tTG levels in one study. 5

Role of HLA-DQ2/DQ8 Genetic Testing

• HLA-DQ2/DQ8 testing should not be performed routinely but is useful in select scenarios, particularly when diagnosis remains uncertain after serology and biopsy. 2, 4

• The test has a negative predictive value exceeding 99%—absence of both HLA-DQ2 and HLA-DQ8 essentially rules out celiac disease. 1, 2, 4

• Approximately 95% of celiac patients have HLA-DQ2 and 5% have HLA-DQ8, but these alleles are present in 25–30% of the general white population, making them unsuitable as primary diagnostic tests. 2, 7