How should celiac disease be diagnosed?

Diagnosis of Celiac Disease

Celiac disease diagnosis requires a two-step approach: initial serologic testing with IgA tissue transglutaminase (TTG-IgA) plus total IgA level, followed by confirmatory upper endoscopy with multiple duodenal biopsies demonstrating characteristic histologic changes. 1, 2

Initial Serologic Testing

The preferred first-line test is IgA TTG antibody combined with total IgA measurement. 1, 2 This combination is critical because:

• TTG-IgA has 90.7% sensitivity and 87.4% specificity in adults (at 15 U/mL threshold), and 97.7% sensitivity with 70.2% specificity in children (at 20 U/mL threshold) 2, 3
• Total IgA must be measured simultaneously because 1-3% of celiac patients have selective IgA deficiency, which causes falsely negative IgA-based tests 1, 2
• In children under 2 years, combine TTG-IgA with deamidated gliadin peptide (IgG and IgA) to improve sensitivity 1

Critical pitfall: Patients must consume at least 10g of gluten daily for 6-8 weeks before testing to avoid false-negative results 1, 2, 4. Starting a gluten-free diet before completing diagnostic workup invalidates both serology and biopsy 1, 2, 4.

If IgA Deficiency is Detected

Switch to IgG-based testing immediately. 1 The preferred test is IgG deamidated gliadin peptide (DGP-IgG), which has superior accuracy with 93.6% sensitivity and 99.4% specificity in adults 2, 4. IgG TTG has limited utility (sensitivity only 40.6-84.6%) and should not be used as a primary screen when total IgA is normal 2, 3.

Confirmatory Serology

When TTG-IgA is positive, obtain endomysial antibody (EMA) testing for confirmation. 1, 2 EMA has excellent specificity of 99.6% in adults and 93.8% in children 1, 2. When TTG-IgA exceeds 10 times the upper limit of normal AND EMA is positive in a second blood sample, the positive predictive value approaches 100% 1, 2, 3.

Histologic Confirmation with Duodenal Biopsy

Upper endoscopy with duodenal biopsy is mandatory for diagnosis in adults and cannot be replaced by serology alone. 1, 2 The only exceptions are patients with coagulation disorders or pregnancy 2, 4.

Biopsy Technique Requirements

• Obtain at least 6 biopsy specimens: 1-2 from the duodenal bulb and a minimum of 4 from the second portion of the duodenum or beyond 1, 2
• Multiple biopsies are essential because mucosal changes can be patchy 1, 4
• Do not rely on duodenal bulb biopsies alone, as they may be compromised by Brunner's glands or peptic changes 1, 4
• Visual endoscopic appearance alone is insufficient to rule out celiac disease 4

Diagnostic Histologic Criteria

Definitive diagnosis requires villous atrophy (Marsh type 3) with: 2, 4

• Crypt hyperplasia with lengthening
• Increased intraepithelial lymphocytes (≥25 IELs per 100 enterocytes)
• Increased lamina propria cellularity (lymphocytes, plasma cells, eosinophils)

"Probable celiac disease" may be diagnosed when biopsies show ≥25 IELs without villous atrophy, combined with positive serology 2, 4. However, lymphocytic infiltration alone is not specific for celiac disease—consider alternative causes including H. pylori infection, small bowel bacterial overgrowth, NSAIDs, olmesartan, and autoimmune disorders 1, 4.

Managing Discordant or Equivocal Results

Positive Serology with Negative Biopsy

If serology is strongly positive but initial biopsies are negative: 2, 4

1. Confirm the patient consumed adequate gluten (≥10g daily for 6-8 weeks) when biopsies were performed 2, 4
2. Verify total IgA level was measured to exclude IgA deficiency 2, 4
3. Have an experienced GI pathologist review biopsy specimens to confirm proper orientation and exclude tangential sectioning artifact 4, 3
4. Consider repeat endoscopy with additional biopsies, as disease can be patchy 2, 4

Negative Serology with High Clinical Suspicion

If suspicion remains high despite negative serology, proceed directly to duodenal biopsy. 1 Seronegative celiac disease represents up to one-third of seronegative enteropathy cases in White populations 4.

Before proceeding, verify: 4, 3

• Patient was consuming adequate gluten during testing
• Total IgA level is normal (not deficient)
• Complete celiac serology panel was obtained (TTG-IgA, DGP, EMA)

HLA-DQ2/DQ8 Genetic Testing

HLA testing should NOT be performed routinely but is useful in select situations. 1 Approximately 95% of celiac patients have HLA-DQ2 and 5% have HLA-DQ8 4. The absence of both alleles has >99% negative predictive value and essentially excludes celiac disease. 1, 2

Use HLA testing in: 1

• Patients with equivocal small-bowel histologic findings
• Those following a gluten-free diet before testing was performed
• Discrepant celiac-specific serology and histology
• Patients with Down syndrome
• When the original diagnosis is in question

Critical limitation: HLA-DQ2 is present in 25-30% of the White population, so a positive result does not confirm disease 1.

Differential Diagnosis of Villous Atrophy

Villous atrophy is not specific for celiac disease. 1 When serology is negative but biopsy shows villous atrophy, systematically exclude: 4

Medication-Induced Enteropathy

• Olmesartan (frequently missed cause)
• NSAIDs
• Mycophenolate mofetil
• Chemotherapy agents

Infectious Causes

• Giardiasis (order stool antigen or PCR)
• Cryptosporidium (especially in AIDS)
• Whipple's disease
• Small intestinal bacterial overgrowth

Immune-Mediated Disorders

• Common variable immunodeficiency (look for absent/reduced plasma cells in lamina propria)
• Autoimmune enteropathy

Inflammatory Bowel Disease

• Crohn's disease (look for granulomas, perform colonoscopy with ileoscopy)

Tests NOT Recommended

Do not order: 1, 3

• Stool studies for celiac disease
• Small-bowel follow-through
• Intestinal permeability testing
• D-xylose testing
• Salivary antibody testing
• Non-deamidated (native) gliadin antibodies
• Capsule endoscopy for initial diagnosis (except in patients unable/unwilling to undergo upper endoscopy)

Differentiating Celiac Disease from Non-Celiac Gluten Sensitivity

Symptoms alone or symptom response to a gluten-free diet cannot reliably differentiate celiac disease from non-celiac gluten sensitivity. 1 This distinction is critical to:

• Identify patients at risk of nutritional deficiency and complications
• Determine risk in family members
• Influence the degree and duration of dietary adherence

Non-celiac gluten sensitivity should only be considered after celiac disease has been ruled out with appropriate serology, small bowel histology (in patients not on a gluten-free diet), and HLA-DQ typing if needed 1.

Common Diagnostic Pitfalls to Avoid

1. Never start a gluten-free diet before completing diagnostic workup 1, 2, 4, 3
2. Never diagnose based on serology alone without biopsy in adults 1, 2, 4
3. Always measure total IgA with IgA-based tests 1, 2
4. Do not accept poorly oriented biopsies—tangential sections can mimic villous atrophy 4, 3
5. Do not overlook medication history, especially olmesartan and NSAIDs 4
6. Do not rely on subjective response to gluten-free diet alone—this has very low positive predictive value 1