A Critical Appraisal of Gluten in Skin Disorders, the Evidence So Far, and Updated Recommendations

Abstract

Gluten, a polypeptide hapten, found in many cereals such as barley, wheat, rye, oats, and others, has been recently implicated in a range of cutaneous disorders ranging from chronic plaque psoriasis through psoriatic arthritis, urticaria (chronic as well as paediatric onset), and angioedema to lichen planus, vitiligo, and rosacea. The evidence for them is still not well reviewed. To generate evidence for the causal role of gluten in various dermatological disorders. The Pubmed, MedLine, and EMBASE databases were searched using the keywords “Gluten” and one of the dermatoses, namely, “Atopic Dermatitis”, “Vasculitis”, “Psoriasis”, “Psoriatic Arthritis”, “Acne”, “Alopecia Areata”, and “Immunobullous disorders”. All articles published in English for which free full text was available were taken into consideration. The search strategy returned in a total of 1487 articles which were screened for relevance and elimination of duplicates. Ultimately, around 114 articles were deemed suitable. The data were extracted and presented in the narrative review format. A simple and cost-effective solution to many of these chronic and lifelong conditions is to restrict gluten in the diet. However, the dermatologist would do well to remember that in the vast majority of dermatological disorders including the ones listed here, gluten restriction is not warranted and can even lead to nutritional deficiencies. The evidence varied from Grade I for some disorders like psoriatic arthritis to Grade IV to most disorders like acne, vitiligo, vasculitis, and atopic dermatitis. Herein, we review the evidence for each of these conditions and make practical recommendations for gluten restriction in them.

Introduction

Gluten is a polypeptide found in wheat and related plants such as rye, oats, barley, and millets such as bajra, jawar, and ragi. The product, unfortunately, is antigenic – it acts like a hapten and gets bounded to intestinal enzymes while processing (specifically transglutaminase), and the enzyme–hapten complex initiates antibody production. This interaction has been implicated in many disorders ranging from dermatitis herpertiformis, classically manifested by celiac disease to urticaria and anaphylaxis.

The latest craze amongst nutritionists is a gluten-free diet (GFD). While such a diet is of undoubted value in gluten sensitivity or dermatitis herpetiformis, its merits and demerits for other skin conditions remain to be examined. Often the dermatologist is confronted by patients, who follow voluntarily a GFD. The Internet and media have played a huge role, with many websites openly promoting a GFD even among those without gluten sensitivity. However, the dermatologist must take an informed decision with regards to its purported benefits and risks involved. The GFD has its own risks, namely, nutritional deficiencies like iron, zinc deficiency, fibre deficiency, and various micro-nutrient deficiency.[1] Certainly, the whole population does not need a GFD, and it is imprudent to correctly advise the patients about the merits and demerits of a GFD.

Aim

To generate evidence in a casual role of gluten for various dermatoses.

Methodology

Search strategy

The Pubmed, MedLine, and EMBASE databases were searched using the keywords “Gluten” and one of the dermatoses, namely, “Atopic Dermatitis”, “Vasculitis”, “Psoriasis”, “Psoriatic Arthritis”, “Acne”, “Alopecia Areata”, and “Immunobullous disorders”. All articles published in English for which free full text was available were taken into consideration.

Results

The search strategy returned in a total of 1487 articles which were screened for relevance and elimination of duplicates. Ultimately, around 114 articles were deemed suitable. The data were extracted and presented in a narrative review format.

Discussion

Herein, a brief review of the progress made so far and updated recommendations for the involvement of gluten in various dermatoses have been presented:

1. Papulo-squamous disorders

   1. Psoriasis: The evidence for the presence of antigliadin antibodies argues for a trial of GFD in these patients. Of note, our experience suggests that IgA-positive psoriasis patients probably represent a different subset of patients who have different pathogeneses and different disease courses vis a vis idiopathic psoriasis (unpublished observations). The range of IgA-positive patients was from 0% to 16% across various studies.[2,3,4,5,6,7] There is also evidence that IgA-positive patients have a more severe PASI, need more PUVA doses, and tend to respond poorly in the absence of a GFD.[8] They are also 2.2 more likely to develop celiac disease later on. As a practical point, because IgA testing is only available in limited settings, a GFD for 3 months is worth a trial [Table 1].
   2. Psoriatic arthritis: No systematic review or meta-analysis exists. However, a case-only series where patients served as internal controls depending on the level of severity showed a correlation with disease severity scores and titres of IgA. Several other studies have found no association. However, a trial of GFD for 3 months as in psoriasis is worth consideration [Table 1].[9,10,11,12]
   3. Atopic Eczema: In large cohorts, the prevalence of celiac disease has been found to be 0.8% and interestingly, many of the exacerbations of atopic dermatitis can be controlled by exclusion of gluten from diet. This correlated with a 3-fold increased prevalence of atopic eczema among those who had IgA to gluten compared to those who did not have. Interestingly, both being disorders of barrier dysfunction, (celiac disease, a barrier dysfunction of gut, and atopic eczema, a barrier dysfunction of skin), a causal association is still yet to be proven. Hypotheses suggested include an alteration of gut flora, leading to microbial translocation in celiac disease patients leading to skin inflammation and barrier disruption. No specific recommendations can be made at this point.[13,14,15]
   4. Oral Lichen Planus: This is an overlooked association. An estimated 15% cases of lichen planus had IgA to gluten in one study. An 8-month trial of GFD improved oral lesions. Interestingly, oral lesions have also been reported in authentic celiac disease cases. IgA testing is recommended in oral lichen planus.[35] The association between oral aphthous ulcers and gluten sensitivity has now been proven in sensitive meta-analyses.[40]

2. Disorders of Foliculo-Pilosebaceus units

   1. Rosacea: A hazard ratio of around 3 has been reported across various studies. With the exact pathogenesis being unclear, it has been postulated that shared haplotypes between rosacea and celiac disease susceptibility genes might play a role. At present, no specific dietary recommendation is needed.
   2. Alopecia Areata: A population hazard ratio of around 2 has been found in sufferers of gluten sensitivity.[40]

3. Vasculitis and related conditions

   Many cases of allergic cutaneous vasculitis have developed on gluten challenge, which subsequently resolved following omission of gluten from diet. Interestingly, number of episodes and severity of each episode correlated nicely with IgA titres. Urticarial vasculitis has also been reported in this context. Specifically, food and exercise-induced anaphylaxis is a well-known association of gluten. The risk of chronic urticaria in children is estimated to be in the order of 5%, while the absolute risk of development of urticaria in someone with celiac disease is around 144/100000 population. Erythema elevatum diutinum, a specific type of IgA-mediated vasculitis (now reclassified as IgA dermatoses), has also occurred in association with gluten sensitivity. Whether it is a chance association remains to be proven [Table 1].

4. Disorders of Subcutaneous Fat: The association between erythema nodosum and IgA levels to gluten is well known. IgA testing in erythema nodosum is recommended if lesions persist >8 weeks or multiple episodes occur in one year.

5. Disorders of pigment: Interestingly, vitiligo has been associated with gluten sensitivity, although the magnitude of risk is currently unknown. The list of the conditions with the evidence for a gluten free diet has been summarized in Table 2.

Table 1.

Evidence for and against gluten in various dermatoses

Disease	Evidence for	Evidence Against	Reccomendation
Psoriasis	1. 16% of psoriasis patients have IgA to gliadin[2] In patients with antigliadin antibodies, 30 out of 33 patients (i.e., 90%) had significant reductions in PASI after 3 months of GFD psoriatic Patients with no IgA to gliadin did NOT improve after GFD In those with IgA to gliadin, they again relapsed after cessation of GFD 2. Michaëlsson et al.[6] did a case-control study on 28 psoriasis patients with anti tTG and found improvements in tTG after 3 months of GFD 3. Wu et al.[7] found an odds ratio of 2.2 to develop celiac disease in patients of psoriasis in a retrospective cohort of 25,341 patients. 4. Anti IgA to gliadin may correlate with need for PUVA, systemic therapy in psoriasis, and more severe disease[8]	1. Sultan et al.[3] did case-control study in Kashmir that found NO difference in prevalence of IgA antigliadin antibodies between psoriasis patients and controls 2. Kia et al.[4] found NO statistical difference in IgA antigliadin prevalence between psoriasis (18%) and controls (19%) 3. Zamani et al.[5] did a case-control study in 328 psoriatic patients in Iran and found no significant elevation of markers of gluten sensitivity. Also, a 3-month trial of GFD did not improve symptoms or PASI	1. GFD may be of benefit in a subset of psoriatic patients who have markers of celiac disease that is, IgA antibodies to gliadin, tTG, etc. 2. Because such patients form approximately only 16% of all psoriatic patients, routine GFD need not be advocated for all psoriatic patients. 3. No evidence to suggest GFD in those psoraisis patients without markers of gluten sensitivity. Following GFD in these patients may predispose to nutritional deficiency.
Psoriatic Arthritis	1. A case-only study showed psoriatic arthritis patients with higher IgA Antigliadin levels had significantly higher ESR and CRP values and longer durations of morning stiffness as compared to patients with lower IgA AGA levels.[9] 2. Isolated case reports suggest associations of psoriatic arthritis with celiac disease.[10] 3. Anti IgA to gliadin correlates with higher ESR, higher CRP, and longer morning stiffness	1. In Nurses Health Study II, compiling data from 1991 to 2015, involving more than 85000 nurses, women who ate the highest amounts of gluten were no more likely to develop psoriasis, psoriatic arthritis, compared to those who ate very little gluten.[11]	1. It may be prudent to evaluate for IgA antigliadin antibodies and tTG even in the absence of GI symptoms in psoriatic patients. 2. In those with positive serologic markers of gluten sensitivity, a GFD should be followed. 3. In the rest, it is still premature to recommend GFD, especially in IgA to gliadin-negative patients.[12] It is important to note that none of the major guidelines on Psoraitic Arthritis recommend GFD
Atopic Eczema	1. A cross-sectional study in 2020 among 1,16,816 patients showed prevalence of celiac disease 0.8% in moderate to severe AD compared to controls[13] 2. Incidence of AE is 3 times more common in celiac disease patients compared to controls.[14] 3. Non-celiac gluten sensitivity has skin manifestations that resemble atopic eczema[15] 4.In a 2017 piece published in the Journal of Dermatological Treatment, 169 people with eczema were surveyed. More than half of those who cut gluten from their diets reported seeing an improvement in their eczema symptoms.[16] 5. 2013 study found that among 149 participants, 80% saw an improvement of their eczema symptoms when following a hypoallergenic diet (which included cutting out gluten).[17]	1. In Nurses Health Study II, compiling data from 1991 to 2015, involving more than 85000 nurses, women who ate the highest amounts of gluten were no more likely to develop atopic eczema, compared to those who ate very little gluten[11] 2. Anti Gliadin IgA is considered to be marker of gluten sensitivity in atopic eczema. In its absence, GFDs are unlikely to work. 3. GFD has been proven not to prevent atopic eczema or atopic march in one large study[11] 4. Possibly Fructan and NOT gluten is more closely linked to Asthma.[18]	1. Gliadin antibodies may be requested in atopic eczema. In its absence, following a GFD is not recommended 2. Too premature now to conclude if GFD also prevents I an Atopic March and subsequent manifestations of Asthma/Allergic Rhinitis 3. In absence of laboratory facilities of immunoglobulin measurement, a trial of GFD may be considered.
Rosacea	1. In a nationwide cohort study, the prevalence of CD was higher among patients with rosacea when compared to control subjects (HR=1.46, 95%CI=1.11–1.93)[19] 2. GWAS (Genome Wide Association Studies) found significant links with Roasecea and Celiac disease due to shared Haplotypes[20,21] 3. Women with Rosacea may be twice more likely to have Gluten sensitivity than controls[20,21]	1. No formal RCT on dietary interventions in Roasecea has focussed on GFD 2. The associations of Gluten sensitivity into subsets of Rosacea and how it affects prognosis is still not known	1. In the absence of GI intolerance, a serum level of antigliadin antibodies may be requested. If positive, a trial of GFD should be considered 2. Not proper to recommend GFD for all roasecea patients as % of Gluten Sensitivity among rosacea patients is likely small.
Vasculitis	1. Isolated Case reports suggestive of Gluten Sensitivity exists. Eg: Meyers et al. reported a case of Allergic Cutaneous Vasculitis (LCV) that resolved on switching to GFD[22] 2. Vascultis is a known manifestation of Celiac disease.[23,24,25,26]	1. Most cases of Vasculitis are not related to Gluten Sensitivity 2. No formal RCT or Case Control Studies on Vasuclitis and Gluten Sensitivity.	1. Antigliadin Immunoglobulins may be considered where history may be suggestive: Eg Concomitant celiac sprue, recalcitrant vascultis. If Positive, a GFD may be considered.
Urticaria and Angiedema	1. Food-dependent exercise-induced anaphylaxis is widely known to be in the spectrum of Gluten-Sensitive disorder[27] 2. A Case Controlled study found 5% cases of Chronic Urticaria in children had concomitant Gluten sensitivity as opposed to only 0.67% of controls[28] 3. GFD led to rapid disease control in these children, Serologic markers took a longer time (5–9 months) to normalise.[28] 4. A large population-based cohort study was conducted in Chornic Urticaria. The corresponding HRs for Gluten Sensitivity were 1.51 for any urticaria (95%CI=1.36–1.68) and 1.92 for chronic urticaria (95%CI=1.48–2.48).[29] 5. The absolute risk for urticaria in CD was 140/100,000 person-years (excess risk=47/100,000 person-years). Corresponding figures for chronic urticaria were 24/100,00 person-years and 12/100,000 person-years. Patients with CD were also at increased risk of having both urticaria (odds ratio, OR=1.31; 95%CI=1.12–1.52) . and chronic urticaria (OR=1.54; 95%CI=1.08–2.18) prior to the CD diagnosis[29]	1. Most inducible Urticarias are not related to Gluten sensitivity, only a minority is related 2. Large RCTs needed to confirm any association 3. Evidence is now limited to Case Control studies and isolated case reports.	1. In Food-dependent Exercise-Induced Anaphylaxis, consideration of GFD should be given 2. In Chronic Urticaria, a trial of GFD may be worthwhile as a part of Low Pseudoallergen diet.
Pemphigus	1. Case reports suggest association of Epilepsy, Pemphigus, and CD. The association is however exceptional.[30] 2. Similarly, there are case reports of pemphigus improving on GFD.[31]	1. Very rare 2. No large case series/RCTs	1. Too premature to recommend routine celiac disease screening in all pemphigus patients. May be considered on select cases.
Erythema Nodosum	1. Reported at least 3 times since 1991 in association with CD. 2. Indicators to suspect underlying Gluten sensitivity may be duration >8 weeks and unusual persistence	Rare case reports only	1. All patients with persistent EN should be screened for CD and gluten sensitivity[32]
Erythema Elevatum Diutinum	1. Isolated case reports 2. Rodriguez et al.[33] first reported case of EED in a case of CD. 3. Pathologically both CD and EED are IgA mediated disease	Rare case reports	1. Routine screening not usually performed worldwide for gluten sensitivity in EED.
Oral Aphthae	1. Case reports suggest association. 2. Meta-analyses of Controlled studies have found association; however, causal effect remains to be determined.[34]	1. Not known if Apthae are part and parcel of CD or related to underlying hematinic deficiency that frequently occurs in CD	1. All patients with Reccurent Aphtae should undergo diagnostic procedures for CD[34,35]
Vitiligo	1. Shahmoradi et al. performed a case controlled study of 64 Vitiligo patients and 64 controls. Serum markers of Gluten sensitivity was seen in 2/64 patients and 0/64 controls[36] 2. Both CD and Vitiligo have similar pathogenesis.	1. Evidence not sufficient to support routine scrrening	Screening only in suspect associations.
Oral Lichen Planus	Cigic et al.[35] did a Case control study where, AGA and anti-tTG, were measured in 56 lichen planus patients. CD was diagnosed in eight OLP patients (14.29%) Additionally, 6 LP patients (10.71%) were positive for IgA Ttg.	1. Limited evidence – More case control/RCTs needed.	Screening may be considered, particularly for recalcitrant cases.
Dermatomyositis	Several case Reports suggest association	No large RCTs/Case Controlled studies	1. Given the ramifications for a diagnosis of DM, screening should be considered.[37,38,39,40]
Alopceia Areata	1. Numerous Case Controlled studies suggest an association.[41,42] 2. In several Case reports, following a GFD led to improvement[41,42]	1. Mokhtari et al.[41] measured antibodies in Alopecia areata and found no difference with their prevalence from that of controls	1. Screening may be performed especially in Alopecia Universalis and Totalis.
Cutaneous Malignancies	1. Cutanous Lymphomas are the best known malignancy known to occur in skin of patients of CD.[43] 2. Incidence of BCC and Melanomas also raised compared to controls, although somewhat marginally,	Only isolated reports exist. Had it been a common association, more studies would have reported it.	Not sufficient to make recommend.
Acquired Generalized Cutis Laxa	Only 1 Case report exists[44]	-	Not sufficient evidence

Table 2.

Evidence and Recommendations for Gluten Restriction in various conditions

Condition	Level of Evidence	Recommendation for GFD
Pemphigus	4	Not Recommended
Urticaria (Spontaneous)	2B	Recommended if IgA positive
Atopic Dermatitis	4	Recommended only if clear history
Psoriasis	1A	Strongly Recommended
Aphthous Ulcers	2A	Strongly Recommended
Oral Lichen Planus	3B	Can be tried
Cutanous Vascultis and EED	4	Not recommended
Dermatolmyositis	4	Not recommended
Erythema Nodosum	4	Explained in text
NME	4	Not recommnended

1A: Systematic Review of RCTs.

1B: Single RCT.

2A: Systematic Reviews of Cohort Studies.

2B: Single Cohort Studies.

3A: Systematic Reviews of Case Control Studies.

3B; Individual Case Control Studies.

4: Case Series and Poor Quality Case Control and Cohort Studies.

5: Expert Opinions.

Concluding Thoughts

Gluten, the polypeptide systemic allergen, found in wheat, rye, oats, barley, and many other cereals, has been implicated in a number of disorders, although the evidence for them is bare minimum. The role of a GFD has been proven through meta-anlayses in dermatitis herpetiformis and oral aphthae. However, evidence pointing to the possibility of such association has been seen in uticarias, psoriasis, psoriatic arthritis, vitiligo, erythema nodosum, and many other conditions. In some of these conditions, a specific subset of patients seem to have antibodies to gluten, and in them, a GFD is warranted. In the rest of the conditions, however, little evidence exists and a call as to whether a GFD is worth trying should be taken on a case-to-case basis.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.