A rare clinical presentation of dermatitis herpetiformis with blisters principally on elbows: a report of three female cases

Introduction

Dermatitis herpetiformis (DH) is a chronic illness classified as autoimmune bullous disease (AIBD) with autoimmunity to enzymes [1] and a variable degree of enteropathy identical to that seen in coeliac disease [2]. Any 2 of the following 3 criteria are sufficient for a DH diagnosis: clinical picture, direct immunofluorescence (DIF), or anti-transglutaminase antibody tests [3]. DIF is the gold standard test for DH diagnoses, with a sensitivity of 90–95% and specificity of 95–100% [2, 4]. The three main DIF patterns in DH are (i) microgranular IgA deposits at the tips of the dermal papillae, (ii) microgranular-fibrillar IgA deposits at the tips of the dermal papillae, and (iii) microgranular IgA deposits along the dermal-epidermal junction (DEJ) [2]. Additionally, evaluation of serum IgA antibodies with a single substrate, namely tissue transglutaminase (TG2/tTG), epidermal transglutaminase (TG3/eTG) or nonapeptides of gliadin (npG) using the ELISA method is useful for diagnosing DH. In our clinical laboratory experience, although eTG is considered the main autoantigen in the cutaneous pathology of DH, analysis of diagnostic accuracy of different monoanalyte ELISA tests (eTG, tTG, npG IgA ELISAs) indicated anti-tTG IgA as the best choice for serological immunodiagnosis of DH [5]. Still, opinions on the diagnosis of DH are very divided and some believe that eTG antibodies should be regarded as primary diagnostic serology, while anti-tTG and other autoantibodies may be used to support a diagnosis and for disease monitoring [6]. A histopathological examination of lesional skin biopsy is not required for a DH diagnosis, but, in obscure cases, findings compatible with DH support the diagnosis [2]. A combined approach entailing a gluten-free diet (GFD) and dapsone remains the primary treatment option.
In our previous publication, we mentioned a 72-year-old woman with blisters and their evolutionary lesions on both elbows and 1/3 of the proximal part of extensor surfaces of forearms [7]. In this patient, skin eruptions in the form of blisters appeared following an influenza vaccination. Notably, there was a family history of autoimmunity as her daughter suffered from mutilating rheumatoid arthritis. In the DIF test, fine granular deposits of IgA (+) were observed at the tips of the dermal papillae and along the DEJ. Positive results were obtained for IgA antibodies against tTG (45.225 RU/ml, cut-off 20 RU/ml).
According to Reunala et al. [2], DH mainly affects adults. The oldest patients were more than 80 years old when diagnosed. According to the Finnish data [2], the mean age at onset is about 50 years, but our experience with ethnic Poles is that DH usually affects young males in the third or even second decade of life. DH is characterized by an intense itch and an inflammatory symmetrical rash, typically on the 1/3 of the proximal part of extensor surfaces of forearms, on the elbows, the knees, and the buttocks, including the sacral area. The eruption is usually polymorphic, consisting of typically barely visible vesicles as well as papules, small wheal-like lesions, and erythematous macules; however, because of the intense itch and associated scratching, erosions, crusts, and postinflammatory hyperpigmentation often dominate the clinical picture. The characteristic vesicles are often not apparent as they are destroyed by excoriation. The severity of the rash varies between individuals, and a more intense rash may also affect other sites, such as the scalp, the face, and the upper back. Mucosal involvement is rare in DH and may be accompanied by subjective complaints, such as dryness, soreness, or burning sensation, and lesions, such as vesicles, macules, erosions, and aphthous ulcers in the oral mucosa. Cottini described a clinical variant of DH where elbows and knees were involved but not the gluteal region [8]. Uncommonly, petechiae and purpura, particularly on the palms and soles (digital purpura), may present alongside classic manifestations or as the sole presenting feature of DH [9, 10]. As purpuric changes may be small and hard to detect, the performance of an acral dermoscopy during a physical examination [2] should be considered. In trichoscopy, DH displays extravasations, yellow hemorrhagic crusts and characteristically clustered dotted vessels, and white diffuse scaling [11]. Rare DH presentations include isolated facial involvement, exclusively macular lesions, leukocytoclastic vasculitis-like appearance, palmoplantar keratosis, urticarial lesions, or lesions mimicking prurigo. Some dental abnormalities have been described in patients with DH. The most common dental findings in patients with DH include horizontal grooves, pits, and discoloration [2]. Duhring defined several types of the disease: erythematous, bullous, vesicular, pustular, and multiform – all of them may be concomitant [12]. Today, bullae are rarely found in clinical manifestations of DH patients [2].

Case report

We present 3 female DH patients with skin eruptions in the form of blisters right on the elbows (exclusively there for 2 of the patients), clinically mimicking epidermolysis bullosa acquisita. Two patients were elderly, and 1 patient was 34 years old. For a 66-year-old patient with a cluster of blisters (fig. 1 A), a DIF examination revealed fine granular deposits of IgA (++) (fig. 1 B) and C3 (+) at the tips of dermal papillae and an examination of IgA antibodies against tTG was negative. A multiplex ELISA test for blistering diseases (DSG1, DGS2, BP180, BP230, envoplakin, type VII collagen) was also negative. For another 73-year-old patient, a DIF examination revealed granular deposits of IgA (++) both at the tips of dermal papillae and along the DEJ. A histopathologic examination showed a subepidermal separation with an accumulation of neutrophils forming neutrophil microabscesses at the tips of dermal papillae (fig. 1 C). The IgA tTG ELISA test was negative. Still, type IV collagen (the lamina densa marker) was visualized by immunohistochemical technique (immunoperoxidase staining) at the bottom of the subepidermal blister, indicating its localization within the lamina lucida space. The multiplex ELISA test for blistering diseases was negative. For the youngest patient, 34 years old, blisters appeared after an abrupt cessation of dapsone therapy (fig. 1 D). A DIF examination revealed fine granular deposits of IgA (++) and IgM (++) at the tips of dermal papillae. A positive result for IgA antibodies against tTG was obtained (> 200 RU/ml, cut-off 20 RU/ml).

Discussion

In human skin, DEJs unite and sustain the connection between two distinctly diverse tissue regions: the cellular epidermis and the connective tissue-rich dermis. A DEJ comprises four distinct compartments, each marked by a consistent presence of specific protein components. Starting from the epidermal side, these compartments are (i) cytoskeleton, hemidesmosomes, cytoplasmic membranes of basal keratinocytes, (ii) lamina lucida, (iii) lamina densa, and (iv) sublamina densa (fig. 2).
Immunohistochemical examination of skin biopsy for DH patients shows an elevated concentration of B granzyme within DEJs as compared to the skin of healthy people. Its presence leads to the degradation of key parts of this combination (α6/β4 integrin, collagen VII and collagen XVII) which are substrates for this protease, inducing proteolysis and DEJ separation. It is also found in microabscesses, mainly in the upper layer of papillary skin adjacent to a DEJ, which suggests involvement of neutrophils and lymphocytes in the secretion of this enzyme [13]. Subepidermal vesicles with an accumulation of neutrophils at the tips of dermal papillae (papillary microabscesses) are typical histopathological findings of DH [6].
The dermal-epidermal junction of human skin exhibits age-related remodeling, resulting in a flattened appearance and reduced surface area. Changes at the DEJ during intrinsic aging involve not only morphological alterations but, more significantly, shifts in protein composition as well. Diminution of protein also occurs within all four zones of the DEJ in intrinsically aged skin. A significant reduction in the protein distribution of collagen IV, collagen VII, collagen XVII, integrin α6/β4, and laminin-332 is observed in aging skin. A functional implication of this altered protein composition appears to be loss of structural integrity, which may partly explain the increased fragility of aged skin. In particular, reduced deposition of collagen XVII at the DEJ in intrinsically aged skin may have functional consequences similar to those seen in the AIBD, especially in bullous pemphigoid [14]. That is probably why we observe blisters in DH more frequently in elderly patients. Additionally, with age, women exhibit a significant decrease in estrogen levels. Estrogens act on the DEJ through interaction with anchoring fibers, which consist of collagen IV and VII produced by dermal fibroblasts and keratinocytes of the basal layer of the epidermis [15]. Additionally, in pathogenesis, we should take into consideration the increased susceptibility of the skin to the Koebner phenomenon. The potential for DH to demonstrate the Koebner phenomenon would explain its predilection for cutaneous eruptions on selective body areas such as knees and elbows [16]. Data show that eTG in the deposits may be of cutaneous origin. Under the influence of microtraumas (Koebner phenomenon), it is displaced to the papillae, which may explain different locations of the active enzyme and that present in the deposits [12]. In younger patients suffering from DH, a rebound phenomenon could be caused by an abrupt cessation of dapsone therapy. Dapsone works quickly, suppressing neutrophils and resolving symptoms of pruritus in several hours and new blister formation in 24–36 hours, making it an effective option for acute inflammatory phases of the disease and early stages of gluten-free diet initiation. DH symptoms and rash usually return within 24 to 48 hours of halting medication if the DH has not yet been adequately controlled without medication [6, 17].
Drugs and chemical compounds are known to exacerbate or cause DH. These include iodine compounds, biological drugs, such as infliximab, adalimumab, interferon α, gonadoliberin (GnRH) analogues, non-steroidal anti-inflammatory drugs including ibuprofen, indomethacin, diclofenac and acetylsalicylic acid as well as cytostatics used in the treatment of proliferative diseases. A case of DH triggered by propafenone has been described [18, 19]. Generally speaking, drugs that activate neutrophils can exacerbate or even trigger DH in genetically prone individuals.
It has been proven that hormones like estrogen and testosterone modulate the activity of inflammatory diseases. Men are more prone to a severe inflammatory reaction whereas women tend to be more at risk when it comes to autoimmune diseases due to production of the Xist ribonucleoprotein complex. Xist (X-inactive specific transcript) is a long non-coding RNA (lncRNA) that plays a crucial role in the process of X-chromosome inactivation in female mammals. Xist inactivates one of the X chromosomes in female cells, protecting them from excessive expression of genes located on X chromosomes. Still, alongside DNA, it also forms autoantigen complexes characteristic of autoimmune diseases. Estrogens also influence the tendency of women to acquire autoimmune diseases by increasing antibodies production [20–27].
Two of our patients had a negative result for IgA antibodies against tTG. Circulating tTG antibodies support the diagnosis, but their absence does not exclude DH [28, 29]. However, DH can be excluded for individuals who are negative for both DIF and anti-tTG [29]. According to the multicenter study, the highest diagnostic accuracy in DH serological diagnosis can be achieved with IIF on the monkey esophagus for IgA endomysial antibodies combined with IgA TG2/tTG ELISA [30]. Perhaps then an evaluation of serum IgA autoantibodies against endomysium analyzed by IIF on monkey esophagus would corroborate the diagnosis. Still, IIF is less convenient than ELISA because it requires subjective evaluation of images. In cases where the clinical picture suggests a diagnosis of DH and IgA antibodies to transglutaminases are undetectable, IgA deficiency should be suspected, blood levels of this immunoglobulin should be determined, and IgG antibody tests should be used [31].

Acknowledgments

We thank Ms Barbara Jastrzębska, a laboratory technician, for her skillful technical assistance.
A part of this work was presented as an oral communication: Bowszyc-Dmochowska M. Opryszczkowate zapalenie skóry w spektrum schorzeń z nietolerancji glutenu/Dermatitis herpetiformis within the spectrum of gluten intolerance disorders. Dermatologia Praktyczna 2023. XVII Ogólnopolska Konferencja Edukacyjna/Practical Dermatology 2023. XVII Nationwide Educational Conference. Łódź, Poland, 17-18 November 2023.
This work, after editing, was presented as an e-poster: Jałowska M, Spałek M, Bowszyc-Dmochowska M, Dmochowski M. Blisters in females with dermatitis herpetiformis: a study of four cases. 8^th Continental Congress of Dermatology International Society of Dermatology with 12^th Controversies in Dermatology Wroclaw, Poland, 23-25 May 2024.
In memory of Ewa Neumann MD, PhD, from whom Marian Dmochowski caught the autoimmune bullous diseases bug.

Funding

No external funding.
Ethical approval
Not applicable.
Conflict of interest
The authors declare no conflict of interest.

References