Cross-Reactivity, Skin Microbiome, and Environmental Triggers in Dermatitis Herpetiformis#

Medical Disclaimer

This content is for informational purposes only. It is not medical advice. Read the full disclaimer.

Overview#

This document examines three under-explored dimensions of dermatitis herpetiformis (DH): non-gluten food triggers and cross-reactivity, the skin microbiome in DH, and environmental/early-life factors that may trigger or modulate the disease. While gluten is the established primary driver of DH, patients frequently report persistent symptoms on a strict gluten-free diet (GFD), raising questions about additional immunological triggers.

Part 1: Cross-Reactivity and Non-Gluten Food Triggers#

1.1 Oats and Avenin#

What is avenin? Avenin is the prolamin (alcohol-soluble storage protein) of oats, analogous to gliadin in wheat, hordein in barley, and secalin in rye. Avenin differs from gliadin in several key ways: - Avenin constitutes only ~10-15% of total oat protein (vs. ~80% prolamin content in wheat) - Avenin has lower proline and glutamine content than gliadin - The immunodominant epitopes differ, though some sequence homology exists

DH-specific oat studies:

Hardman et al. (1997) in the New England Journal of Medicine challenged 10 DH patients (on strict GFD for a mean of 15.8 years) with uncontaminated oats (62.5 g/day) for 12 weeks. No patients experienced adverse effects, and no rash developed. The study concluded that oat avenin showed no toxic effects in DH patients.
Reunala et al. (1998) challenged 11 DH patients in remission with 50 g oats daily for 6 months. Eight remained asymptomatic, two developed transient rash, and one withdrew due to mild but persistent rash (~9% symptomatic rate in this small cohort).
Reunala et al. (2020) conducted a large long-term safety study of 312 long-term treated DH patients. Of these, 256 (82%) were consuming oats as part of their GFD. Long-term follow-up showed no differences in illnesses, celiac disease complications, or medication usage between oat consumers and non-consumers.

What percentage of celiac/DH patients react to pure oats?

The estimates vary by study design and selection bias: - A systematic review concluded that the majority of adults and children with CD can tolerate moderate amounts of pure, uncontaminated oats - Lundin et al. (2003) in PLOS Medicine found that 3 of 9 celiac patients (~33%) had avenin-induced mucosal inflammation, but this cohort was selected for patients with oat exposure history, biasing the estimate upward - Beyond Celiac reports that while ~53% of celiac patients may experience some GI symptoms after eating uncontaminated oats, only ~8% showed increased immune response on blood testing - A 2019 study found acute IL-2 elevation (T-cell activation) in ~40% of 29 participants, but again noted significant selection bias toward oat-sensitive individuals

Is the immune cross-reactivity proven? Yes, in a subset. Lundin et al. identified avenin-reactive mucosal T-cells in some celiac patients that can cause mucosal inflammation. The T-cell response is HLA-DQ2-restricted and targets specific avenin peptides that structurally resemble gliadin epitopes after deamidation by tissue transglutaminase. However, the T-cell response to avenin is generally weaker and less consistent than the response to gliadin.

Current clinical guidance on oat introduction: - Most celiac disease guidelines (including those from Finland, UK, Canada, and Scandinavia) permit introduction of pure, uncontaminated oats after initial stabilization on a GFD - Health Canada's position states that the majority of CD patients can tolerate moderate amounts of pure oats - The recommendation is typically to introduce oats gradually, monitor for symptoms, and potentially test with follow-up serology or biopsy - For DH specifically, the evidence supports safety for the vast majority (~90%+), but individual monitoring is warranted

Sources: - NEJM - Absence of Toxicity of Oats in DH (Hardman 1997) - NEJM - Absence of Toxicity of Avenin in DH (Reunala 1999) - PLOS Medicine - Molecular Basis for Oat Intolerance (Lundin 2003) - PMC - Systematic Review: Coeliac Disease and Oats - Frontiers - To Be Oats or Not to Be (2019) - PMC - Long-Term Safety of Oats in DH (Reunala 2020) - Beyond Celiac - Subset Reacts to Oat Protein - Health Canada - Celiac Disease and Safety of Oats

1.2 Dairy and Casein#

Is there evidence of casein cross-reactivity with gluten peptides in celiac/DH?

Multiple mechanisms have been proposed to explain why dairy worsens symptoms in celiac/DH patients:

Mechanism 1: Secondary lactose intolerance (most established) - Over 50% of celiac patients experience secondary lactose intolerance at diagnosis - Lactase enzyme is produced at the tips of intestinal villi; villous atrophy destroys lactase-producing cells first - This is typically reversible: most patients recover lactase production as villi heal on a GFD (may take months to years) - This mechanism is well-established and is the most common explanation for dairy intolerance in celiac patients

Mechanism 2: Cow's milk protein-induced enteropathy - A landmark study by Auricchio et al. (1985) in the BMJ found that ~50% of celiac patients showed mucosal inflammatory responses to cow's milk protein similar to gluten reactions - Bovine beta-casein can cause an enteropathy that mimics celiac disease, causing villous blunting and inflammation (Bovine Beta Casein Enteropathy) - Four children with celiac disease on a GFD showed persistent villous atrophy and elevated serologies that resolved only after eliminating cow's milk protein - Bovine milk intolerance in celiac disease has been specifically linked to IgA reactivity to alpha- and beta-caseins

Mechanism 3: Molecular mimicry/structural similarity - Casein is rich in glutamine and proline, the same amino acids that tissue transglutaminase (TG2) binds to in gliadin - There is a homologous amino acid sequence between beta-casein and gliadin that could enable antigenic mimicry - Casein's glutamine/proline content means TG2 may deamidate casein peptides, potentially creating neo-epitopes recognized by gliadin-reactive T cells - This could explain elevated celiac autoantibodies (anti-TG2) persisting despite strict GFD in some dairy-consuming patients

Mechanism 4: Increased intestinal permeability - The enteropathy of celiac disease increases mucosal permeability - This may allow immune responses to dietary antigens (including milk proteins) that would not normally cross the epithelial barrier - This is a non-specific mechanism and could apply to many food proteins

Is the cross-reactivity IgA-mediated? Evidence suggests it can be. Bovine milk intolerance in celiac disease has been shown to correlate with IgA reactivity to alpha- and beta-caseins specifically. However, IgE-mediated cow's milk protein allergy (a distinct mechanism) can also coexist.

Clinical bottom line: Dairy intolerance in celiac/DH is real and multifactorial. The most common cause (secondary lactose intolerance) is temporary and resolves with GFD. However, a subset of patients has genuine casein-driven enteropathy that may require dairy elimination independent of gluten status.

Sources: - PMC - Mucosal Reactivity to Cow's Milk Protein in Coeliac Disease - ScienceDirect - Bovine Milk Intolerance in CD Related to IgA Reactivity to Caseins - MedDocs - Milk Protein-Induced Villous Atrophy in Children with CD - Beyond Celiac - Cow's Milk Protein Allergy and Non-Recovery on GFD - Allergic Living - Dairy Intolerance Joins Celiac Disease

1.3 Corn (Zein)#

Background: Zein is the prolamin storage protein of corn. Some celiac/DH patients report symptoms from corn, leading to speculation about cross-reactivity.

The Vojdani study (2013): Vojdani and Tarash published a widely cited study in Food and Nutritional Sciences testing whether affinity-purified anti-alpha-gliadin 33-mer antibodies cross-react with various food antigens. They reported significant immune reactivity between gliadin antibodies and corn, among other foods.

However, this study has been heavily criticized:

Christina L. Graves, Ph.D. (UNC Chapel Hill, 2019) published a detailed critique identifying serious methodological problems: - The study used polyclonal and monoclonal antibodies against gliadin applied to food extracts in ELISA, but this does not demonstrate that the human immune system cross-reacts in the same way - Gluten contamination of the food samples was not adequately controlled for -- Vojdani & Tarash themselves acknowledged they "don't know" whether some reactions represented true cross-reactivity or contamination - The study tested IgG/IgA binding in vitro, which does not establish that T-cell-mediated responses (the driver of celiac pathology) would occur - Gluten cross-reactivity would only be relevant to patients with celiac disease (who have circulating anti-gliadin antibodies), not to people with non-celiac gluten sensitivity - There has been no independent replication of these findings, and no credible rebuttal to Graves' critique

Computational/sequence analysis: Some bioinformatic analyses predict sequence similarity between zein moieties and alpha-gliadin. However, celiac disease-specific antibodies do not appear to cross-react with corn zein in properly controlled experiments.

Clinical bottom line: There is currently no robust molecular evidence for cross-reactivity between zein and gliadin that would drive celiac or DH pathology. Reports of corn sensitivity in celiac/DH patients may reflect other mechanisms (corn contamination with gluten, independent food sensitivities, or increased intestinal permeability allowing immune responses to various food proteins).

Sources: - SCIRP - Cross-Reaction between Gliadin and Food Antigens (Vojdani 2013) - Celiac.com - Is Cross-Reaction a Celiac Disease Myth? (Graves critique) - Paleo Foundation - 19 Gluten Cross-Reactive Foods Myth Busted

1.4 Rice, Soy, and Other Grains#

Rice: - Vojdani's 2013 study reported antibody binding between gliadin antibodies and rice, but this suffers from the same methodological issues noted above - Anecdotal reports of rice sensitivity exist in celiac patient communities, but no clinical studies have established rice as a trigger for celiac or DH through cross-reactivity - The cited rice/gluten cross-reactive study relied on IgE-mediated responses with different epitopes than those recognized by anti-alpha-gliadin antibodies, making extrapolation problematic

Soy: - Soy contains prolamins, but at much lower concentrations than wheat - Soy is rich in globulin storage proteins, which have been studied for potential immunogenicity in celiac disease - There is limited evidence that soy prolamins can trigger celiac-type immune responses; the literature is much more robust for wheat, barley, rye, and (in a subset) oats - Anecdotal reports exist but lack clinical validation

Millet: - Vojdani's study included millet among foods showing antibody reactivity, subject to the same caveats - No independent clinical studies confirm millet as a DH or celiac trigger

Other "safe" grains (quinoa, buckwheat, amaranth, teff): - These are generally considered safe for celiac patients - Quinoa prolamins have been theoretically flagged by some researchers, but clinical studies have not shown meaningful reactivity - The primary concern with any "gluten-free" grain remains cross-contamination during growing, harvesting, and processing rather than intrinsic cross-reactivity

Key principle: Increased intestinal permeability in active celiac/DH may allow immune responses to various food proteins that would not occur in a healthy gut. This does not represent true molecular mimicry or cross-reactivity but rather a consequence of barrier dysfunction. As the gut heals on GFD, many of these secondary food sensitivities may resolve.

Sources: - SCIRP - Vojdani 2013 Cross-Reaction Study - Nature - Characterization of Globulin Storage Proteins in Relation to Celiac Disease - Frontiers - Database of Celiac Disease-Associated Peptides

1.5 Coffee#

The Vojdani finding: The 2013 Vojdani study reported that alpha-gliadin antibodies reacted with instant coffee at 82% of the level of gliadin-gliadin binding. Some subsequent reports suggested that highly processed coffees (instant, commercial ground) showed more cross-reactivity than organic whole-bean coffees.

Critical assessment: - This finding comes from the same methodologically flawed study critiqued by Graves (2019) - The result was obtained using purified anti-gliadin antibodies in an ELISA plate, not from human immune cells in vivo - The possibility of gluten contamination in processing was not excluded - No independent replication exists - The proposed mechanism (that coffee processing creates proteins structurally similar to gliadin) has no supporting structural biology evidence

Other potential coffee-symptom connections: - Coffee increases gastric acid secretion and gut motility, which could exacerbate GI symptoms in patients with active enteropathy - Caffeine may independently affect intestinal permeability - Some instant coffees contain additives (maltodextrin, etc.) that may have independent effects - These pharmacological effects are distinct from immune cross-reactivity

Clinical bottom line: There is no credible evidence that coffee proteins cross-react with gliadin to drive celiac or DH pathology. If coffee worsens symptoms, the mechanism is more likely pharmacological (motility, acid secretion) or related to additives in processed coffee products rather than immune cross-reactivity.

Sources: - SCIRP - Vojdani 2013 Study (includes coffee finding) - Celiac.com - Cross-Reaction Myth Analysis - Gluten Free Society - Is Coffee Safe?

1.6 Processed Food Additives#

Microbial Transglutaminase (mTG) -- the most concerning additive:

Microbial transglutaminase is a heavily used food processing enzyme (labeled as "enzyme" or sometimes not labeled at all) that functionally imitates tissue transglutaminase (TG2), the autoantigen of celiac disease. This is the most biologically plausible food additive concern:

mTG catalyzes protein cross-linking in processed meats, dairy products, baked goods, and other foods
mTG can cross-link gliadin peptides in vitro, creating complexes that are immunogenic in celiac patients
mTG enhances intestinal permeability, suppresses mucus barriers, and has anti-phagocytic activity
mTG mimics its family member TG2 structurally and functionally
The industrial food additive mTG has been shown to be immunogenic in celiac disease patients (antibodies against mTG are elevated)
mTG is classified as "generally recognized as safe" (GRAS) by regulatory authorities, but accumulating evidence challenges this designation for celiac patients
Since TG3 (epidermal transglutaminase) is the DH autoantigen, and mTG mimics TG2, there is a theoretical pathway for mTG to exacerbate DH by amplifying transglutaminase-related autoimmunity

Carrageenan: - Commonly used as a thickener in dairy alternatives, processed foods, and supplements - Induces intestinal inflammation in animal models by disrupting the intestinal barrier and reducing mucin content - Decreases populations of beneficial gut bacteria, particularly Akkermansia muciniphila - Increases expression of pro-inflammatory cytokines (TNF-alpha, IL-6) - Long-term consumption is associated with inflammation and gut microbiota changes - Inflammatory effects are reversible upon removal from the diet - Particularly problematic for individuals with existing intestinal permeability issues or autoimmune conditions - No DH-specific studies exist, but the mechanism (barrier disruption + inflammation) is relevant to DH pathogenesis

Maltodextrin: - Ubiquitous in processed foods, often used as a filler or thickener - Promotes endoplasmic reticulum stress-driven mucus depletion and exacerbates intestinal inflammation - Impairs the intestinal mucus barrier and accelerates colitis in animal models - Increases pathogenic phenotypes of Crohn's disease-associated bacteria - Disrupts tight junction proteins, potentially increasing intestinal permeability - May predispose to low-grade inflammation and could be a risk factor for inflammatory bowel disease - Frequently used (ironically) as a placebo in gut health clinical trials, calling into question the validity of some study designs

Modified Food Starch: - Can be derived from wheat, corn, potato, or tapioca - When wheat-derived, it may contain residual gluten (though typically below 20 ppm threshold) - In the US, if derived from wheat it must be labeled; in other jurisdictions, labeling requirements vary - The primary concern for celiac/DH patients is hidden gluten rather than cross-reactivity

General principle for additives and DH: Celiac/DH patients already have compromised intestinal barrier function. Additives that further increase permeability (carrageenan, maltodextrin) or mimic disease-relevant enzymes (microbial transglutaminase) represent biologically plausible mechanisms for symptom persistence or exacerbation, even on a strict GFD. However, direct clinical evidence in DH patients is lacking for all of these additives.

Sources: - PMC - Food Additives as Triggering Factors in Celiac Disease (2019) - PMC - Microbial Transglutaminase as Environmental Inducer of CD - PMC - Processed Food Additive mTG and Cross-Linked Gliadin Complexes - PMC - mTG Is Immunogenic in Pediatric Celiac Disease - Frontiers - Maltodextrin Impairs Intestinal Mucus Barrier - ScienceDirect - Maltodextrin Promotes ER Stress-Driven Mucus Depletion - PMC - Carrageenan in the Diet: Friend or Foe for IBD - Frontiers - Role of Carrageenan and CMC in Intestinal Inflammation

1.7 Iodine as a Non-Gluten Trigger (Unique to DH)#

Iodine deserves special mention as the only non-gluten dietary factor with a well-established, mechanistically understood role in DH flares.

Mechanism: - IgA antibodies to transglutaminase 3 (TG3) complexes deposited in DH skin are enzymatically active - This enzymatic activity is dramatically increased in the presence of iodide - Iodine does not cause DH de novo; it exacerbates existing disease by amplifying the TG3-IgA inflammatory cascade - The observation that iodine triggers DH dates to 1891, though the molecular mechanism was only recently characterized

Sources of problematic iodine: - Iodized salt - Seaweed/kelp (very high iodine content) - Seafood (especially shellfish) - Dairy products (from iodine-containing sanitizers used in dairy processing) - Dietary supplements containing potassium iodide or sodium iodide - Radiologic contrast media - Medications (amiodarone, povidone-iodine)

Diagnostic use: - The potassium iodide patch test has been used historically in DH diagnosis - All 5 patients with active, untreated DH in one study had positive patch tests - Only 2/6 patients on GFD and 1/8 on dapsone tested positive, reflecting disease control

Clinical relevance: - DH patients resistant to dapsone should be evaluated for excessive dietary iodine intake - Some DH patients may need to moderate (not eliminate) iodine-rich foods, particularly during flares - This effect is independent of gluten and represents a direct amplifier of the DH-specific skin inflammation

Sources: - PMC - DH Resistant to Dapsone Due to Dietary Iodide - Celiac Disease Foundation - DH and Iodine Exposure - PubMed - Potassium Iodide Patch Test in DH - JAMA Dermatology - Iodide-Induced Immunofluorescence in DH - JAAD - DH Flare from Triiodomethane Packing Strips

Part 2: Skin Microbiome in DH#

2.1 What Lives on DH Skin?#

Current state of research: Very limited

Despite extensive skin microbiome research in atopic dermatitis, psoriasis, and acne, no published studies have used 16S rRNA gene sequencing or shotgun metagenomics to characterize the skin microbiome specifically in DH patients (lesional vs. non-lesional skin vs. healthy controls). This represents a significant research gap.

What we know from related conditions:

From atopic dermatitis research (the most studied inflammatory skin condition for microbiome): - Lesional skin shows dramatically reduced microbial diversity compared to non-lesional skin - Staphylococcus aureus dominates lesional sites during flares, often exceeding 90% relative abundance - Recovery from flares is associated with return of microbial diversity - Common healthy skin colonizers include Corynebacterium, Cutibacterium (formerly Propionibacterium), Micrococcus, Staphylococcus epidermidis, Streptococcus, and various Proteobacteria

Why the gap matters for DH: DH lesions have a distinct immunological microenvironment (IgA/TG3 deposits, neutrophilic infiltration, complement activation) that likely creates different selective pressures on microbial communities compared to atopic dermatitis (Th2-driven, eosinophilic). Characterizing the DH skin microbiome could reveal: - Whether specific bacteria amplify or modulate the neutrophilic response - Whether microbial metabolites interact with TG3 enzymatic activity - Whether dysbiosis at lesional sites contributes to the itch-scratch-infection cycle

Sources: - PMC - Skin and Gut Microbiome in Common Dermatologic Conditions - MDPI - Skin Microbiome Shifts in Various Dermatological Conditions - PMC - Bacterial 16S rRNA Sequencing in Cutaneous Research

2.2 Staphylococcus and Secondary Infection#

No DH-specific S. aureus colonization studies exist, but extensive data from atopic dermatitis provides a relevant framework:

S. aureus in inflammatory skin disease: - S. aureus colonization rates of 57-100% are reported in atopic dermatitis, compared to ~5-30% in healthy controls - S. aureus produces V8 protease, which directly activates pruriceptor sensory neurons through PAR1, driving itch independent of the immune system - S. aureus produces superantigens (SEA, SEB) that non-specifically activate T cells, amplifying inflammation - S. aureus biofilm formation impairs targeted treatment and allows persistent colonization - Skin barrier disruption (from any cause) promotes S. aureus colonization; colonization further damages the barrier (vicious cycle)

The itch-scratch-infection cycle in DH: DH is one of the most intensely pruritic skin conditions. The scratch reflex: 1. Mechanically disrupts the epithelial barrier 2. Creates micro-wounds that serve as entry points for S. aureus and other pathogens 3. Introduces bacteria from subungual (under-nail) reservoirs 4. S. aureus colonization amplifies itch through V8 protease-PAR1 signaling 5. Superantigen production by S. aureus drives additional T-cell activation 6. The cycle perpetuates, potentially masking whether the primary driver is ongoing immune reactivity or secondary infection

Clinical implications: - Impetiginization (secondary bacterial infection of existing skin lesions) can complicate DH and may be underrecognized - Treatment of secondary S. aureus infection may be necessary alongside dapsone and GFD - The contribution of microbial superantigens to DH flare severity is unstudied but biologically plausible

Sources: - PMC - Interactions Between Atopic Dermatitis and S. aureus Infection - Frontiers - S. aureus Colonization in Atopic Dermatitis - ScienceDirect - S. aureus V8 Protease-PAR1 Axis Drives Itch - PMC - Role of Bacterial Skin Infections in Atopic Dermatitis (International Eczema Council)

2.3 Skin-Gut Microbiome Axis in DH#

Gut dysbiosis in celiac disease/DH:

Celiac disease patients show consistent patterns of gut dysbiosis: - Decreased beneficial microbes: Bifidobacterium, Lactobacillus - Increased potentially pathogenic bacteria: Bacteroides, E. coli, Proteobacteria - Increased gram-negative bacteria overall - These imbalances persist even after GFD adoption, though they partially normalize

DH-specific microbiome findings (limited): - DH patients show Bacillota (Firmicutes) phylum dominance in gut microbiota - Gut microbiota composition differs between celiac patients with GI symptoms and those presenting with DH, suggesting the microbiome may influence disease phenotype (gut vs. skin manifestation) - All reviews examining DH identify gut microbiota dysbiosis as a common feature - Altered gut microbiota increases production of pro-inflammatory cytokines, impairs the mucosal barrier, and produces microbial transglutaminase -- all relevant to DH pathogenesis

Gut-skin axis mechanisms:

The gut-skin axis operates bidirectionally: 1. Gut-to-skin signaling: Intestinal dysbiosis increases systemic inflammation via bacterial translocation, LPS-driven immune activation, and altered short-chain fatty acid production. This systemic inflammatory tone can modulate skin immune responses. 2. Metabolite-mediated effects: Gut bacteria produce metabolites (SCFAs, tryptophan derivatives, bile acid metabolites) that reach the skin via circulation and influence keratinocyte differentiation, antimicrobial peptide production, and local immune responses. 3. Immune cell trafficking: Gut-primed immune cells (particularly IgA-secreting plasma cells and T cells) can migrate to skin. In DH, this is the direct pathogenic mechanism -- IgA-producing B cells activated in the gut deposit IgA-TG3 complexes in the papillary dermis. 4. Skin-to-gut signaling: UVB exposure on skin increases serum vitamin D levels, which in turn increases gut microbiome beta-diversity. This bidirectional relationship may partly explain geographic/seasonal variation.

Probiotics in DH: - A 2025 umbrella review found severely limited evidence on probiotics specifically for DH - Proposed mechanisms include immune modulation, reduced intestinal permeability, and anti-inflammatory effects - Most evidence is indirect, extrapolated from celiac disease or atopic dermatitis studies - Bifidobacterium and Lactobacillus species can produce endopeptidases that digest gliadin epitopes, reducing their immunogenicity - No published clinical trial results exist specifically for probiotics in DH treatment

Sources: - PMC - Gut-Skin Axis: Microbial Dysbiosis and Skin Conditions - MDPI - Gut-Skin Axis Interrelationship - Frontiers - Gut Microbiota in Celiac Disease: Role for Probiotics? - PMC - Gut Microbiota Alteration and Probiotics in Celiac Disease - Frontiers - Efficacy of Probiotics in DH Management (2025 Umbrella Review) - Frontiers - Skin Gluten-Related Disorders: New and Old Cutaneous Manifestations - PMC - DH: A Common Extraintestinal Manifestation of Coeliac Disease

Part 3: Environmental and Early Life Triggers#

3.1 Infections as Triggers#

Viral infections are the most established environmental triggers for celiac disease onset.

Reovirus (the landmark finding): - Jabri et al. (2017) published in Science that infection with reovirus -- a common, ordinarily harmless virus -- can trigger the immune system to break oral tolerance to gluten - Reovirus suppresses peripheral regulatory T cell (pTreg) conversion and promotes Th1 immunity to dietary antigens - The mechanism depends on interferon regulatory factor 1 (IRF1) and type-1 interferon signaling - Celiac patients had significantly higher anti-reovirus antibody titers compared to healthy controls, suggesting prior reovirus infection - This study was groundbreaking because it showed a seemingly innocuous virus (not causing overt illness) could permanently alter immune tolerance - The finding raises the possibility that vaccines against reovirus could potentially prevent celiac disease in genetically predisposed individuals

Enterovirus: - Prospective data from the Norwegian MIDIA study found enterovirus infections were significantly more frequent before celiac disease autoantibody appearance (OR 6.3, 95% CI 1.8-22.3; p = 0.005) - The TEDDY study (multi-center, prospective) found cumulative stool enteroviral exposures between ages 1-2 years increased celiac disease autoimmunity risk - Critically, there was a significant interaction between enterovirus exposure and gluten consumption -- the risk from enteroviruses was amplified in children with higher gluten intake - Coinfection with both parechovirus and enterovirus was associated with markedly increased risk

Rotavirus: - Evidence is mixed - One prospective study found children infected with rotavirus had higher celiac prevalence, with a dose-response relationship (OR 1.94 for one infection, 3.76 for two or more) - However, subsequent cohort studies did not replicate this association - Some evidence suggests molecular mimicry between rotavirus VP-7 protein and tissue transglutaminase, but a dedicated study (PMC 2016) found lack of evidence of rotavirus-dependent molecular mimicry

Swedish epidemic data: - During the Swedish celiac disease epidemic (1987-1997), repeated neonatal infections were linked to celiac disease onset (OR 1.52) - This epidemiological observation is consistent with the viral trigger hypothesis

SARS-CoV-2 / COVID-19: - Theoretical mechanisms proposed (cytokine storm increasing intestinal permeability, allowing gliadin passage) - However, a large study found previous SARS-CoV-2 infection was not associated with increased celiac disease autoimmunity in children - A broader autoimmunity study found modestly increased celiac disease risk post-COVID (adjusted risk ratio 1.80), but this may reflect increased testing or detection bias - Long-term follow-up is needed

DH-specific data on viral triggers: None. All viral trigger research has focused on celiac disease onset. Whether viral infections specifically predispose to the DH phenotype (skin vs. gut manifestation) is unknown.

Sources: - Science - Reovirus Triggers Inflammatory Responses to Dietary Antigens (Jabri 2017) - UChicago Medicine - Seemingly Innocuous Virus Can Trigger Celiac Disease - Frontiers - Enterovirus Infections Associated with Celiac Disease Development - PMC - TEDDY Study: Cumulative Effect of Enterovirus and Gluten - PLOS Pathogens - A Viral Trigger for Celiac Disease - PMC - Lack of Evidence of Rotavirus-Dependent Molecular Mimicry - PMC - COVID-19 and Celiac Disease: A Pathogenetic Hypothesis - Frontiers - New-Onset Autoimmune Disease After COVID-19

3.2 Antibiotic Exposure#

The evidence is conflicting, with large studies on both sides.

Studies finding an association: - A Swedish population-based case-control study of 2,933 celiac patients found antibiotic use was associated with celiac disease (OR 1.40, 95% CI 1.27-1.53) - A meta-analysis of 6 studies on antibiotic exposure found any infection was associated with increased celiac disease risk (OR 1.37, 95% CI 1.2-1.56, p < 0.001) - A dose-response pattern was observed: earlier exposure and repeated prescriptions showed higher risk

Studies finding no association: - The TEDDY study (prospective, HLA-genotyped newborns from Finland, Germany, Sweden, and the US) found early antibiotic use did not increase celiac disease risk - Multiple analyses within the TEDDY framework showed the same pattern: no association

Proposed mechanism (for those finding an association): - Antibiotics reduce beneficial bacteria (Bifidobacterium, Lactobacillus) and alter the microbial balance - In infancy, when the immune system is still developing tolerance, microbiome disruption may interfere with normal immune training - This window of vulnerability may explain why some studies find timing-dependent effects

Confounding issues: - The association between antibiotics and celiac disease may be confounded by the infections that prompted antibiotic use (reverse causation / indication bias) - Infections themselves are independently associated with celiac disease risk - Separating the effect of the antibiotic from the effect of the infection is methodologically challenging

DH-specific data: The gut microbiota of DH patients shows Bacillota dominance and differs from both celiac patients with GI symptoms and healthy controls. However, no studies have examined whether antibiotic exposure specifically predisposes to the DH phenotype.

Sources: - PubMed - Infection, Antibiotic Exposure, and Risk of Celiac Disease Meta-Analysis - Gastroenterology - Association Between Antibiotics in First Year of Life and CD - PMC - Antibiotic Exposure and Development of Coeliac Disease (Swedish case-control) - Beyond Celiac - Antibiotics Don't Increase Celiac Risk (TEDDY)

3.3 Early Life Factors#

Timing of gluten introduction:

The consensus has shifted significantly over the past decade: - Previous recommendation (pre-2014): Introduce gluten at 4-6 months, ideally while breastfeeding, to create a "window of tolerance" - NASSCD statement (2015): Neither the timing of gluten introduction nor the duration/maintenance of breastfeeding influences the risk of celiac disease -- "back to the drawing board" - Two large multicenter randomized trials (PreventCD, CELIPREV) found no effect of gluten introduction timing on celiac disease risk - Recent emerging evidence (2025): Some data suggest introducing gluten as early as 4 months may reduce risk at age 3, and gluten intake between 7-15 months was linked to increased risk -- but these findings are not yet reflected in consensus guidelines - Current guidance: introduce gluten between 4-6 months of age, with no specific window shown to prevent celiac disease

Breastfeeding: - Large prospective studies show breastfeeding does not prevent celiac disease development - However, observational studies suggest breastfeeding may delay symptom onset -- breastfed celiac patients present at older ages on average - Breastfeeding during gluten introduction and longer breastfeeding duration correlated with delayed disease onset - Recent data (2025) suggests breastfeeding in the first 6 months may be protective, though this conflicts with earlier randomized trial data

C-section vs. vaginal delivery: - C-section delivery produces measurable gut microbiome differences in infants: - Decreased Bacteroides vulgatus and Bacteroides dorei - Increased hospital-associated organisms (Enterococcus, Klebsiella) - Reduced microbial diversity overall - Decreased folate biosynthesis pathway - The CDGEMM study found that C-section, formula feeding, and early antibiotics together produced distinct gut microbiome patterns linked to immune dysfunction in at-risk infants - However, a meta-analysis found cesarean section is NOT associated with increased celiac disease risk in offspring - The microbiome effects of C-section are real, but their translation to celiac disease risk appears insufficient in isolation

DH-specific early life data: None exists. All early life factor research pertains to celiac disease. Whether early life factors influence the probability of developing DH (vs. celiac without skin manifestations) is unknown.

Sources: - PMC - NASSCD Statement: Gluten Introduction, Breastfeeding, and Celiac Disease - Celiac Disease Foundation - New Clues: Breastfeeding, Gluten Intake, Birth Order (2025) - PMC - Influence of Early Feeding Practices on Celiac Disease in Infants - PMC - Celiac Disease Prevention - PubMed - Cesarean Section and Celiac Disease Risk: TEDDY Study - Celiac Disease Foundation - CDGEMM: Microbes in At-Risk Infants - BMC Microbiome - Multi-omics Analysis of Gut Microbiota in At-Risk Infants

3.4 Seasonal Patterns#

Direct evidence for seasonal DH flares is sparse. The literature identifies the following relevant considerations:

What is established: - The primary DH flare triggers are gluten exposure, iodine intake, and NSAID use -- none of these have inherent seasonal patterns - The symptoms of DH "can come and go throughout your life" with periods of remission and flares

What is plausible but unstudied: - Heat and sweat: DH lesions commonly occur on extensor surfaces (elbows, knees, buttocks) -- areas prone to friction and sweating. Heat-induced sweating could theoretically exacerbate itch and mechanical irritation of existing lesions, but no studies have examined this - UV exposure: UVB increases vitamin D synthesis and modulates skin immune responses. UV has been shown to increase gut microbiome beta-diversity. Whether UV exposure has net positive or negative effects on DH is unknown - Seasonal dietary changes: Iodine intake may fluctuate seasonally (seafood consumption patterns), which could affect DH through the iodine-TG3 mechanism - Seasonal infections: Winter viral infections could theoretically modulate immune responses, though no DH-seasonal data exists

Birth season and celiac disease: - Children born in summer who receive gluten for the first time during winter (when vitamin D synthesis is low) show higher celiac disease rates than those born in winter - This suggests vitamin D status at the time of first gluten exposure may influence disease development - Whether this applies to DH phenotype specifically is unknown

The tobacco clue: Interestingly, tobacco smoking has a protective effect on DH. Patients with DH are less likely to smoke or have hypercholesterolemia. The mechanism is unstudied but may relate to nicotine's immunomodulatory effects (anti-inflammatory in some contexts). This observation has not been explored seasonally.

Sources: - StatPearls - Dermatitis Herpetiformis - PMC - Possible Role of Vitamin D in Celiac Disease Onset - PubMed - DH and Cigarette Smoking - Medscape - DH Etiology and Pathophysiology

3.5 Geographic Patterns#

Established prevalence patterns:

Region	Prevalence (per 100,000)	Incidence (per 100,000/year)
Finland	Up to 75.3	~3.5
UK	~30	~1.8 (declining to 0.8)
North America (European descent)	Similar to Northern Europe	0.4-2.6
Southern Europe	Lower than Northern Europe	Lower
Africa	Extremely rare	Rare case reports only
Asia (incl. Japan, China)	Extremely rare	Rare case reports only

The DH-to-celiac ratio is shifting: Currently ~1:8 (DH:CD). DH incidence is declining (from 1.8 to 0.8 per 100,000 between 1990-2011) while celiac disease incidence is increasing. This paradox is likely explained by improved celiac disease screening catching patients before they develop DH.

Genetic factors (the dominant explanation): - HLA-DQ2 is found in ~85% of DH patients; HLA-DQ8 in ~15% - HLA-B8/DR3/DQ2 haplotype frequency is very low in East Asian populations (<1% in Japan) - Japanese DH patients typically lack HLA-DQ2/DQ8 entirely, and most cannot be linked to celiac disease -- suggesting Japanese DH may have a distinct pathogenesis - The absence of predisposing HLA haplotypes is the primary explanation for rarity in African and Asian populations - North American populations of European descent show similar rates to Northern Europe, strongly implicating genetics over geography

Latitude and vitamin D: - Celiac disease prevalence is greater at higher latitudes (away from equator), mirroring other autoimmune diseases - Lower UVB exposure at higher latitudes reduces skin vitamin D synthesis - Vitamin D regulates both innate and adaptive immune system activity, including oral tolerance mechanisms - However: A meta-analysis found that even after adjusting for latitude, significant geographic variation persists, suggesting additional factors

Vitamin D paradox in celiac disease: - Children with celiac disease show high rates of vitamin D deficiency (~80% in one study) even in high-sunlight regions - This likely reflects malabsorption from villous atrophy rather than a primary deficiency - Vitamin D deficiency may therefore be both a cause (contributing to immune dysregulation) and a consequence (from celiac-related malabsorption) of the disease

Dietary patterns: - Low wheat consumption in parts of Asia and Africa likely contributes to low DH prevalence independently of genetics - The interaction between genetic susceptibility (HLA haplotype) and environmental exposure (wheat-containing diet) is necessary for disease expression - Populations transitioning to Western diets may see rising celiac/DH rates if they carry susceptibility alleles

Environmental factors beyond genetics: - Wheat consumption levels - Vitamin D status / sunlight exposure - Gut microbiome composition (influenced by local diet, sanitation, antibiotic use) - Viral infection patterns - Hygiene hypothesis considerations (reduced microbial exposure in developed nations may contribute)

Sources: - StatPearls - DH Epidemiology - PMC - DH: An Update on Diagnosis, Disease Monitoring, and Management - Frontiers - Dermatitis Herpetiformis: Novel Perspectives - PMC - Lower Prevalence of Celiac Disease in Southern vs Northern US Latitudes - ScienceDirect - Latitude and Celiac Disease Prevalence Meta-Analysis - PMC - Role of Vitamin D in Celiac Disease and IBD - PMC - Possible Role of Vitamin D in Celiac Disease Onset - PMC - Distinct Characteristics in Japanese Dermatitis Herpetiformis - PMC - DH: From Genetics to Development of Skin Lesions

Summary of Key Research Gaps#

Topic	Evidence Level	Status
Oat safety in DH	Strong	Safe for ~90%+; small subset reacts
Casein-driven enteropathy in CD	Moderate	Multiple mechanisms proposed; clinical validation limited
Corn/zein cross-reactivity	Weak	Single flawed study; no replication
Coffee cross-reactivity	Very weak	Methodological concerns; likely not immune-mediated
Microbial transglutaminase	Growing	Biologically plausible; human outcome studies needed
Carrageenan/maltodextrin	Moderate (for IBD)	No DH-specific studies
DH skin microbiome	Absent	No published studies
S. aureus in DH	Absent	Extrapolated from atopic dermatitis
Gut-skin axis in DH	Limited	Dysbiosis confirmed; mechanisms theoretical
Viral triggers for DH specifically	Absent	All data from celiac disease studies
Antibiotics and celiac risk	Conflicting	Meta-analyses disagree
Early life factors and DH	Absent	No DH-specific data
Seasonal DH patterns	Absent	No systematic studies
Geographic/vitamin D patterns	Moderate	Latitude gradient confirmed; mechanisms debated

Last updated: 2026-02-14