Celiac and Hashimoto's: The Thyroid Connection

People with celiac disease are three to four times more likely to develop Hashimoto's thyroiditis, and people with Hashimoto's carry a similar elevated risk of celiac. The link runs along three overlapping tracks: a shared HLA-DQ2 and HLA-DQ8 genetic haplotype, molecular mimicry between gliadin and thyroid peroxidase (TPO), and zonulin-mediated intestinal permeability that exposes the immune system to thyroid-cross-reactive antigens. If you have one, screen for the other. Every newly diagnosed Hashimoto's patient should get a celiac panel (tTG-IgA plus total IgA) before starting a gluten-free diet, and every newly diagnosed celiac patient should get TPO and thyroglobulin antibodies measured.

How Strong Is the Celiac-Hashimoto's Link?

The landmark review by Ch'ng, Jones, and Kingham (2007) in Clinical Medicine and Research consolidated prevalence data from dozens of smaller studies and established the three to four times bidirectional risk that every subsequent paper references. The precise numbers move with the cohort. Celiac prevalence in autoimmune thyroid disease runs 1.6% to 4.8% across populations, against a general-population baseline of roughly 1%. Autoimmune thyroid disease prevalence in celiac cohorts runs 10% to 30%.

The relationship is symmetric. Celiac diagnosis predicts future Hashimoto's. Hashimoto's diagnosis predicts future celiac. Neither is a random cluster. Subsequent meta-analyses through the 2010s and 2020s, including work from Roy and colleagues, have held the three-to-four-fold ratio steady. The mechanism is not incidental, and the epidemiology is not soft.

The clinical implication is direct. One autoimmune condition in the thyroid-gut axis dramatically shifts the pre-test probability for the other. Primary care clinicians who screen for celiac only when a patient presents with diarrhea or weight loss miss most of these cases.

Mechanism 1: Shared HLA-DQ2 and HLA-DQ8 Genetics

The first and most fundamental overlap is genetic. Roughly 95% of celiac patients carry HLA-DQ2 or HLA-DQ8, the two class II major histocompatibility alleles that present gluten-derived peptides to the immune system. A smaller but still substantial fraction of Hashimoto's patients carry the same alleles. Published estimates put HLA-DQ2/DQ8 prevalence in Hashimoto's cohorts at roughly 50% or higher, against a general-population carrier rate of 30% to 40%.

HLA molecules are the dinner plates on which the immune system serves peptide fragments to T cells. The DQ2 and DQ8 plates happen to fit deamidated gliadin peptides unusually well. The same structural quirk that makes gluten a problem in the gut appears to create collateral exposure to self-peptides in the thyroid. Several studies have mapped how HLA-DQ2 also efficiently presents TPO-derived and thyroglobulin-derived epitopes to T cells. One susceptibility locus, two autoimmune outcomes.

What HLA Testing Can and Cannot Tell You

HLA-DQ2 and DQ8 status is a susceptibility marker, not a diagnosis. A positive result means you carry the genetic architecture that permits celiac disease. It does not mean you have it. Roughly one in three people in the general population carry one of these alleles, and most of them will never develop either celiac or Hashimoto's.

Where the test earns its cost is as a rule-out. A negative HLA-DQ2/DQ8 result has a negative predictive value approaching 99% for celiac. If both alleles are absent, celiac is essentially ruled out for life, and no further serology or biopsy is needed when future symptoms arise. The test is unaffected by current diet, which makes it uniquely useful for patients who started a gluten-free trial before thinking to get tested.

Order HLA typing when serology is ambiguous, when the patient has already gone gluten-free, or when a first-degree relative has confirmed celiac and you want a one-time screen.

Mechanism 2: Gliadin Molecular Mimicry with TPO

Molecular mimicry is the phenomenon where a foreign protein shares enough structural similarity with a self-protein that antibodies raised against the invader also bind the host. Gliadin, the prolamine fraction of wheat gluten, contains amino acid sequences that overlap with epitopes on thyroid peroxidase and thyroglobulin.

Natter and colleagues (2001) published the first direct evidence that antibodies from celiac patients cross-react with thyroid tissue. The experimental setup was straightforward: take serum containing gliadin antibodies, incubate it with thyroid sections, and observe binding. The antibodies lit up the thyroid follicles. Naiyer and colleagues (2008), publishing in Thyroid, extended the finding with a more rigorous experimental design. They showed that anti-tissue-transglutaminase (anti-tTG) antibodies from celiac patients bind thyroid follicular cells and extracellular matrix, and proposed that this deposition contributes to thyroid dysfunction in celiac patients.

Tissue Transglutaminase Is Not Just a Gut Enzyme

The anti-tTG antibody is the workhorse of celiac serology. The enzyme it targets, tissue transglutaminase 2, was originally characterized as a gut-mucosal protein involved in wound healing. It turns out tTG2 is expressed in many tissues, including the thyroid. Antibodies that bind tTG2 in the small bowel will also bind tTG2 in thyroid follicles.

This has a clean clinical consequence. A patient with active celiac disease, even without villous atrophy on biopsy, can have anti-tTG antibodies depositing in thyroid tissue, driving localized inflammation, and accelerating the autoimmune process that produces Hashimoto's. Immunohistochemistry studies have confirmed antibody deposition in the thyroid follicles of patients with concurrent celiac and autoimmune thyroid disease. The mechanism is not speculative.

Mechanism 3: Zonulin, Leaky Gut, and Systemic Antigen Exposure

The third pathway was mapped by Alessio Fasano's lab at Massachusetts General Hospital over two decades of work. Fasano identified zonulin as the only known physiological regulator of intestinal tight junctions in humans. Gliadin triggers zonulin release by binding the CXCR3 receptor on intestinal epithelial cells. Zonulin then disassembles the ZO-1 and occludin proteins that seal the spaces between gut cells.

The result is what the functional medicine world calls leaky gut and what the peer-reviewed literature calls increased intestinal permeability. Partially digested gluten fragments, bacterial lipopolysaccharides, and other food antigens cross into the systemic circulation. The immune system encounters these molecules outside their normal compartment and mounts a response. Over time, the response can extend to self-tissues that share structural features with the escaped antigens.

The Three-Factor Model

Fasano has argued that autoimmunity requires three conditions operating together: a genetic predisposition, an environmental trigger, and a leaky gut that permits systemic antigen exposure. Remove any one of the three, and the autoimmune process does not initiate or does not progress. In celiac plus Hashimoto's, all three conditions are present. HLA-DQ2/DQ8 supplies the genetics. Gliadin supplies the trigger. Zonulin supplies the permeable barrier.

The model predicts something useful. Elevated gut permeability should appear in Hashimoto's patients even when celiac is absent. The prediction has held up in published cohorts. Zonulin levels and lipopolysaccharide-binding protein run higher in Hashimoto's patients than in matched controls, including in patients who test negative for celiac. The autoimmune thyroid process can proceed on a leaky-gut substrate that was opened by gluten but is maintained by a broader ecosystem of disruptors: dysbiosis, NSAIDs, alcohol, and chronic stress.

For a deeper look at how zonulin and tight junctions work and how to restore barrier integrity, see our leaky gut autoimmune protocol and the emerging evidence for larazotide as a zonulin blocker.

Non-Celiac Gluten Sensitivity and the Thyroid

Non-celiac gluten sensitivity (NCGS) is the diagnostic bucket for people who react symptomatically to gluten but test negative for celiac disease and wheat allergy. Prevalence estimates range from 0.6% to 6% of the general population, depending on definition and cohort. No validated biomarker exists. Diagnosis requires excluding celiac and wheat allergy, then demonstrating symptom improvement on a gluten-free diet and symptom return on a blinded rechallenge.

NCGS is where the celiac-thyroid conversation gets messy and where most patients actually live. A sizable minority of Hashimoto's patients have classic gluten-related symptoms, negative celiac serology, and no villous atrophy on biopsy. They do not have celiac, and they do not have a normal gut either. Krysiak and colleagues (2019) enrolled exactly this population in their randomized trial on gluten-free diet and thyroid autoimmunity, and the mechanism is plausible: sub-threshold zonulin release, chronic low-grade antigen exposure, and gliadin cross-reactive antibodies that do not reach the diagnostic cutoff for tTG-IgA.

The evidence for NCGS-driven thyroid autoimmunity is weaker than for celiac-driven thyroid autoimmunity. The rating is Grade C on mechanism alone and Grade B when anchored to the Krysiak trial. Absence of biomarker makes the diagnosis inherently noisy. Placebo and nocebo effects are not small. A structured elimination and rechallenge remains the most useful clinical tool.

For the broader evidence on gluten and systemic inflammation outside the gut, see our guide to gluten and joint inflammation.

Does a Gluten-Free Diet Lower TPO Antibodies?

This is the question most Hashimoto's patients actually want answered. The honest answer depends on which patient you are.

The Krysiak 2019 RCT

Krysiak, Szkróbka, and Okopieñ published the most-cited randomized trial on this question in Experimental and Clinical Endocrinology and Diabetes in 2019. The design was tight. Thirty-four Polish women with drug-naïve Hashimoto's and biopsy-supported non-celiac gluten sensitivity were randomized to a strict gluten-free diet or a control diet for six months. At the end of six months, the gluten-free group showed significant reductions in TPO antibodies and thyroglobulin antibodies, along with increases in 25-hydroxyvitamin D. The control group showed no significant change on any measure.

The trial is the best evidence we have for GFD lowering TPO in a Hashimoto's-specific population. It is also, honestly, a small trial. Thirty-four women. Single center. A specific phenotype (drug-naïve, NCGS-positive, early disease). Anyone who tells you the Krysiak result generalizes to every Hashimoto's patient is overselling.

The Mixed Picture

Osowiecka and Myszkowska-Ryciak published a 2023 systematic review of gluten-free diet effects on Hashimoto's that captured the heterogeneity honestly. Across the included studies, results ranged from significant TPO reduction to no detectable effect. The variance was not random. Trials enrolling patients with confirmed celiac or NCGS showed consistent benefit. Trials enrolling unselected Hashimoto's patients, regardless of gluten status, showed weak or null effects. The population defines the result.

This matters for how you set expectations. A patient with Hashimoto's plus confirmed celiac should expect TPO antibody reduction on a strict gluten-free diet. A patient with Hashimoto's plus gluten-triggered symptoms (even without a celiac diagnosis) has reasonable odds of TPO reduction, supported by Krysiak. A patient with Hashimoto's and no gluten reactivity, no celiac, and no NCGS markers has weaker odds. Gluten-free is not a universal thyroid cure.

Evidence Grades by Scenario

• Grade A. Gluten-free diet for confirmed celiac disease, whether or not Hashimoto's is also present. This is standard of care. The diet is lifelong and non-negotiable.
• Grade B. Gluten-free diet for Hashimoto's plus confirmed or clinically suspected non-celiac gluten sensitivity. Krysiak 2019 plus mechanistic plausibility. Worth a structured six-month trial with TPO antibody monitoring before and after.
• Grade C. Gluten-free diet for Hashimoto's without celiac or NCGS. Mixed evidence. A 60 to 90-day trial with labs and symptom tracking is reasonable but the base rate of meaningful TPO reduction is lower than most Hashimoto's blogs suggest.

Testing Strategy: What to Order and When

Three clinical scenarios cover almost every reader who finds this article. Each has a distinct testing sequence.

If You Were Just Diagnosed with Hashimoto's

Order the celiac panel before changing your diet. The antibodies that confirm celiac decline within weeks of gluten elimination, and a subsequent test can read falsely negative.

• tTG-IgA. The primary screening antibody. Sensitivity and specificity both exceed 95% in IgA-sufficient adults on a gluten-containing diet.
• Total serum IgA. Roughly 2% to 3% of celiac patients have selective IgA deficiency, which makes all IgA-based celiac tests falsely negative. If total IgA is low, order tTG-IgG or DGP-IgG instead.
• DGP-IgG (deamidated gliadin peptide IgG). The alternative serology for IgA-deficient patients.
• EMA-IgA (endomysial antibody). A highly specific confirmatory test when tTG-IgA is borderline or positive.
• HLA-DQ2/DQ8 typing. Optional. Most useful as a rule-out or when the patient has already gone gluten-free.

If You Were Just Diagnosed with Celiac

Order thyroid labs at diagnosis and repeat annually even if the first set reads normal. Hashimoto's can emerge years after celiac diagnosis, and the transition from antibody-positive to overt hypothyroidism is often silent.

• TPO antibody and thyroglobulin antibody. The two Hashimoto's autoantibodies.
• TSH, free T4, free T3. Functional status.
• 25-hydroxyvitamin D, ferritin, B12, zinc. Nutrient deficiencies are near-universal in untreated celiac and relevant to thyroid function.

If You Have TPO Antibodies and GI Symptoms but No Celiac Diagnosis

This is the most common real-world scenario and the one where sequencing matters most. Do not eliminate gluten before testing.

• Start with the full celiac panel (tTG-IgA, total IgA, DGP-IgG, EMA) while still eating gluten normally.
• If serology is positive, refer to a gastroenterologist for biopsy.
• If serology is negative but symptoms persist, HLA-DQ2/DQ8 typing can help stratify. Negative HLA makes celiac essentially impossible. Positive HLA does not confirm celiac but keeps the door open for NCGS.
• If celiac is excluded and HLA does not rule out susceptibility, consider a structured 60 to 90-day gluten elimination with TPO, vitamin D, and symptom tracking before and after. Reintroduce gluten at day 90 and watch for symptom return over the following two weeks.

For Hashimoto's-specific lab interpretation including functional versus conventional ranges for TSH, free T3, free T4, and antibodies, see Hashimoto's lab targets and TPO and thyroglobulin antibodies explained.

Villous Atrophy, Biopsy, and Silent Celiac

Small bowel biopsy remains the diagnostic gold standard for celiac in adults. The pathologist grades damage using the Marsh classification, from Marsh 1 (increased intraepithelial lymphocytes, normal villi) through Marsh 3c (total villous atrophy). Biopsy-free diagnosis is accepted in some pediatric scenarios with very high tTG-IgA plus positive EMA, per the 2020 ESPGHAN guidelines. Adult practice is more conservative, and most gastroenterologists still confirm serology with endoscopy before committing a patient to a lifelong gluten-free diet.

The less-discussed clinical reality is silent celiac. A patient has villous atrophy on biopsy, positive serology, and no digestive symptoms worth mentioning. Estimates vary, but roughly one in three celiac cases identified in Hashimoto's cohorts are silent at screening. Fatigue, brain fog, joint aches, and subclinical iron deficiency are often present when you ask specifically, but the patient came to the doctor about the thyroid, not the gut. They would never have been screened on GI symptoms alone.

This is the case for routine celiac screening in new Hashimoto's diagnoses. Silent celiac accelerates thyroid autoimmunity, produces nutrient deficiencies that mimic thyroid symptoms, and is only diagnosable while the patient is still eating gluten. The cost of the tTG-IgA plus total IgA panel is trivial. The cost of missing celiac for a decade is not.

Practical Protocol: What to Do Next

The right next step depends on which diagnosis you already have and which you are still investigating.

If You Have Hashimoto's Only

• Order tTG-IgA and total IgA before making any dietary change.
• If positive, get a gastroenterology referral and plan for biopsy confirmation and lifelong strict gluten-free diet.
• If negative, consider a structured 60 to 90-day gluten elimination with TPO antibody, vitamin D, and symptom tracking at baseline and at the end. Reintroduce gluten at the end and observe.
• Layer selenium 200 mcg per day of L-selenomethionine. The CATALYST trial and the 2024 Huwiler meta-analysis support Grade A evidence for TPO reduction. See selenium for Hashimoto's.
• Optimize vitamin D to 50 to 80 ng/mL. The VITAL trial (2022) showed 22% reduction in incident autoimmune disease at 2,000 IU daily over 5.3 years.

If You Have Celiac Only

• Order TPO antibody, thyroglobulin antibody, TSH, free T4, and free T3. Repeat annually.
• Commit to strict gluten-free, not casual gluten-free. Cross-contamination from shared fryers, shared toasters, and shared cutting boards produces ongoing antibody stimulation. The villi cannot heal on a "mostly" gluten-free diet.
• Address nutrient repletion. Iron, vitamin D, B12, zinc, and selenium are commonly depleted in untreated celiac and directly relevant to thyroid function.
• Every supplement must be certified gluten-free. Many excipients and capsule coatings contain wheat-derived ingredients.

If You Have Both

• Strict gluten-free is the foundation. Non-negotiable, lifelong, certified gluten-free products only.
• Layer selenium 200 mcg per day and vitamin D to 50 to 80 ng/mL.
• Support gut barrier healing. L-glutamine at therapeutic dosing, zinc carnosine, and in selected cases the research peptide BPC-157 all target epithelial repair. Discuss any peptide use with your physician.
• Consider the full AIP diet for Hashimoto's if gluten elimination alone does not meaningfully reduce symptoms at 90 days. Abbott and colleagues (2019) showed significant symptom reduction and inflammatory marker improvement on AIP in Hashimoto's, though TPO antibodies did not change.
• For the broader supplement framework graded by condition, see best supplements for autoimmune disease and the condition-specific supplements for Hashimoto's.
• For the pillar treatment overview, see Hashimoto's natural treatment.

Discuss all testing and dietary changes with both your endocrinologist and your gastroenterologist. The two specialists see different parts of the same problem, and coordinated management produces better outcomes than either working alone.

FAQ

Can gluten cause Hashimoto's?

Gluten does not cause Hashimoto's in isolation. In genetically susceptible people, specifically HLA-DQ2 or HLA-DQ8 carriers, gluten is a documented trigger through molecular mimicry with TPO and zonulin-mediated intestinal permeability. It is one factor among several, alongside viral infection (Epstein-Barr is a notable trigger), psychological stress, iodine excess, and selenium or vitamin D deficiency. For a susceptible patient, removing gluten can slow or partially reverse the autoimmune process. For a non-susceptible patient, it will not.

Will going gluten-free lower my TPO antibodies?

If you have confirmed celiac disease, yes. Strict gluten-free diet reduces thyroid antibodies consistently in this population. If you have non-celiac gluten sensitivity plus Hashimoto's, the Krysiak 2019 randomized trial showed meaningful TPO reduction over six months. If you have Hashimoto's without celiac or NCGS, the evidence is mixed. A 60 to 90-day trial with TPO labs before and after is reasonable and low-risk, but do not expect the Krysiak numbers to apply to every phenotype.

Should everyone with Hashimoto's be tested for celiac?

Yes. The three-to-four-fold prevalence overlap, the frequency of silent celiac in Hashimoto's cohorts, and the low cost of the serology panel make universal screening the right default. Order tTG-IgA plus total IgA at the time of Hashimoto's diagnosis, before any dietary change. Repeat screening every few years is reasonable if the initial test was negative and symptoms evolve.

What is the difference between celiac and non-celiac gluten sensitivity for my thyroid?

Celiac produces villous atrophy and specific antibodies (tTG-IgA, EMA) that are directly implicated in thyroid antibody deposition. The thyroid link is mechanistically strongest here. NCGS has no biomarker and is diagnosed by exclusion. Both can drive thyroid inflammation through overlapping pathways including molecular mimicry and permeability, but celiac does so more reliably and with clearer evidence.

Do I need a biopsy to confirm celiac?

In adults, almost always yes. Small bowel biopsy during upper endoscopy remains the diagnostic gold standard. Some pediatric guidelines allow biopsy-free diagnosis when tTG-IgA is greater than ten times the upper limit of normal plus positive EMA, but adult gastroenterology is more conservative. Biopsy also rules out differential diagnoses that mimic celiac serology (giardia, bacterial overgrowth, autoimmune enteropathy).

I already started a gluten-free diet. Can I still get tested?

Accurate celiac serology requires active gluten exposure. Antibodies fall within weeks of elimination, and a test done after three months of gluten-free eating will usually read negative even in true celiac patients. Options: restart gluten at roughly 3 to 10 grams per day (about two slices of bread) for two to six weeks, known as a gluten challenge, then retest. Alternatively, HLA-DQ2/DQ8 typing can be done anytime and is unaffected by diet. A negative HLA result rules out celiac for life. Discuss the challenge with your doctor, especially if you felt significantly better off gluten.

What supplements help Hashimoto's patients with celiac?

Selenium 200 mcg per day of L-selenomethionine has Grade A evidence for TPO antibody reduction. Vitamin D to a serum 25(OH)D of 50 to 80 ng/mL has Grade A evidence for reduced autoimmune incidence (VITAL trial, 2022). Iron is frequently depleted in untreated celiac and directly affects thyroid function; target ferritin above 70 ng/mL. B12 and zinc are commonly low. All supplements must be certified gluten-free. Discuss dosing with your physician, particularly if you are on levothyroxine, because iron and calcium can interfere with absorption and must be taken at least four hours apart from the thyroid hormone.

The Bottom Line

Celiac and Hashimoto's cluster because they share the same genetic substrate, the same trigger, and the same barrier pathology. HLA-DQ2 and HLA-DQ8 supply the genetics. Gliadin supplies the trigger and the molecular mimic. Zonulin supplies the leaky barrier. The three-to-four-fold bidirectional prevalence is the clinical signature of a unified mechanism, not a coincidence.

Test in both directions at diagnosis. Sequence matters. Order the celiac panel before the pantry purge. A gluten-free diet is Grade A for confirmed celiac, Grade B for Hashimoto's plus gluten sensitivity, and Grade C for Hashimoto's alone. Run the right trial for your phenotype, not the trial that sounds most dramatic on social media.

Take the free three-minute AutoimmuneFinder quiz to build a personalized, evidence-graded protocol matched to your specific conditions, severity, and current medications.

This article is for educational purposes only and does not constitute medical advice. Celiac disease and Hashimoto's thyroiditis require proper medical diagnosis, and any testing strategy or dietary change should be coordinated with your physician, endocrinologist, and gastroenterologist. Do not start, stop, or change any supplement, diet, or medication without consulting your healthcare provider. All dosage recommendations should be discussed with your doctor before implementation.