Gluten sensitivity plays a pathogenic role in both skin and gastrointestinal manifestations of DH [22, 23]. A gluten-free diet (GFD) offers the possibility of control of the skin disease without the use of medications [32]. Dietary modification alone can alleviate the skin lesions [27, 29], and adherence to a GFD is essential for preventing recurrence in the long term [16, 31, 32]. However, it often takes months for complete remission to occur on GFD alone, and many patients may not be able to comply with a GFD either because of personal preference, financial considerations, or availability of gluten-free foods.

Referral to a dietician with expertise in gluten-free nutrition is essential to optimize patient understanding and compliance. A study by Fry et al. found that only 50 % of patients who thought they were adhering to a strict gluten-free diet (GFD) were actually doing so [27]. Participation in national or regional Celiac disease associations (e.g., Celiac Disease Foundation [33], Celiac Support Association [34]) that provide additional information about gluten-free meal options as well as psychosocial support can significantly improve compliance and quality of life.

One study found that following of a GFD enabled a tapering of dapsone in up to 96 % of patients, while 28 % of DH patients were able to cease dapsone altogether [35]. Similar rates were reported in comparable studies [32]. Although the goal in diet modification is strict gluten avoidance, even partial dietary compliance may reduce the requirement for systemic therapy [25]. Any benefit achieved is proportional to the strictness of dietary gluten restriction and the patient and physician must understand that total gluten avoidance must be adhered to in order to significantly reduce the requirement for pharmacologic therapy [27]. Maintaining a GFD for at least 5 months is required to achieve a relapse-free reduction of dapsone dosage and some 8–48 months before it can be ceased altogether [22, 23, 32]. Our clinical experience recently has suggested that institution of a strict gluten-free diet when patients first develop symptoms may result in a quicker response to the diet than in patients who have had symptoms or been treated with dapsone alone for a year or more (Hall, RP unpublished data). Over several years, the amount of IgA deposition in the skin appears to improve on GFD [27, 36, 37], although complete clearance was uncommon even in patients with good clinical response who were followed up to 43 months [37]. Moreover, there was no significant difference in the quantity of IgA deposited, regardless of whether patients were controlled with a GFD alone, GFD and dapsone, dapsone alone, or in those in clinical remission [36]. It is important to understand that the disappearance of IgA is certainly related to the degree of compliance and the duration of the GFD. In the minority of patients who showed a clearance of cutaneous IgA deposition, gluten challenge resulted in the reappearance of cutaneous IgA deposits within 1 month; however, this may vary considerably depending in part upon the extent of the gluten challenge [31]. Leonard et al. reported that patients in whom IgA was no longer present in the skin had been on a strict GFD for an average of 7.6 years [31]. A more direct immunologic effect of a GFD is reflected in elevation of serum IL-8, a putative pathogenic mediator, in DH patients on gluten-containing regular diets. Adherence to a GFD has been shown to reduce or even normalize serum IL-8 levels in these patients, suggesting that the gut is the source of this circulating cytokine [38].

Gluten avoidance has been demonstrated to reverse the macroscopic and histological changes in the small intestinal tissue among patients with DH [27]. While the reported increased incidence of lymphoma in patients with DH has been recently questioned [39]. Lewis and coworkers have demonstrated that patients with DH had no increased risk of any malignancy, gastrointestinal malignancy, or lymphoproliferative cancer when compared to gender and age-matched controls [39]. Others, however, have hypothesized that adherence to a GFD may play a role in protecting patients from lymphomas [40, 41].

A small gluten challenge study showed that both skin and gastrointestinal manifestations recur upon reintroduction of gluten in diet [31]. However, gluten restriction need not be lifelong in all patients, as one study reported that 7 out of 38 patients who reverted to a normal, gluten-containing diet after an average of 8 years of gluten restriction continued to experience remission of their skin and gut disease after a mean follow-up of 12 years [42].

Apart from gluten, other dietary proteins may play a pathogenic role in the development of DH [43]. The consumption of an elemental diet has been shown to control the eruption of DH in as quickly as 2–4 weeks [44].

A GFD should be considered as the first-line treatment in the management of DH although many patients may choose dapsone therapy alone. Long-term compliance with a GFD is essential, although remission can occur in as many as 10–12 % of patients despite being on a normal, gluten-containing diet [24, 25]. The possibility of unintentional or hidden gluten exposure should always be considered if relapse does occur. Serological studies for IgA anti-tissue transglutaminase and IgA epidermal transglutaminase may be helpful in assessing dietary compliance in some patients.

Dapsone is usually the first-line agent in the management of DH. It offers rapid and satisfactory control of cutaneous symptoms with minimal side effects at its lowest effective dose [16, 28, 29, 45, 46]. While this efficacy is documented by decades of clinical efficacy, no randomized control trials have yet been performed for the use of dapsone in DH.

A detailed discussion of its pharmacokinetics and mechanism of action, dosing, and monitoring is provided in Chap.​ 49. It is thought that dapsone exerts its therapeutic action by mediating inflammatory and immunomodulatory processes within the skin [47, 48]. Dapsone impairs neutrophil function, including chemotaxis [49], neutrophil adherence to tissue-bound antibody [50], and myeloperoxidase release [51]. It also thought to decrease the release of IL-8 from keratinocytes [52]. Which of these mechanisms is most important in patients is however not known. Dapsone does not alter the histopathologic features of the gastrointestinal mucosa [27, 30] nor does it affect the deposition of complement in the skin of DH patients even when their cutaneous lesions are controlled [53].

Treatment is usually commenced at a dose of 25–50 mg per day, with upward titration measured against clinical response and patient tolerance. The exact dosage required for DH is variable; however, most patients respond within 24–36 h of initiating treatment. Excellent initial control can be achieved at doses between 50 and 200 mg per day, and some patients only require 25 mg per day [26, 54].

In patients who adhere to a GFD, tapering of dapsone should be done gradually to prevent disease flare-ups and may take as long as 2 years [35]. Tapering regimens are largely physician or institution dependent; however, it should only be considered after the patient has been on a gluten-free diet for at least 2 months [55].

Sulfa-containing medications such as sulfapyridine, sulfasalazine, and sulfamethoxypyridazine are reserved for patients who cannot tolerate dapsone. These are not all readily available and are generally less effective than dapsone. Experience with sulfa medications in DH is based on case reports [56–60].

Patients have been reported to respond to 2–4 g per day of sulfasalazine [57, 59]. Sulfasalazine is cleaved by intestinal bacteria into sulfapyridine and 5-aminosalicylic acid (5-ASA). Sulfapyridine is absorbed, but 5-ASA remains in the large intestines and may exert a local anti-inflammatory effect in the colon. Thus the 5-ASA component has little utility in GSE [61, 62].

While sulfapyridine has a more predictable absorption profile, it usually requires compounding. Patients are usually controlled with 1–2 g of sulfapyridine daily.