The classic clinical presentation of patients with DH is that of a papulovesicular eruption that is present on the extensor surfaces of the extremities, sacrum, and scalp (Fig. 10.1). Pruritus is significant, often dramatically affecting quality of life, and can be confined to the specific skin lesions or generalized. Due to the degree of pruritus that is present, the classic skin lesions of papulovesicles are often difficult to find as secondary changes of scratching such as excoriations or dermatographism may be the predominant clinic findings (Fig. 10.2). With more advanced cutaneous disease, the papulovesicles can coalesce, forming large, scaling patches or plaques with serous crusts and vesicles present at the periphery of the lesions. If the presence of the vesicles is not appreciated in these larger lesions, the location and appearance of the plaques can mimic psoriasis (Fig. 10.3). Other presentations can include urticarial plaques, bullous lesions, and larger erosions (Fig. 10.4). Occasionally, the presentation can be nonspecific, with only faint erythema and minimal scaling present on clinical examination (Fig. 10.5).

An unusual presentation of DH is that of isolated digital or palmar petechiae, purpura, and microvesicles. This is a more common presentation in children [27], but adult cases have been described as well [28]. As the digital lesions can be the only manifestation of disease in these patients, a high degree of clinical suspicion is required.

There are several diagnostic procedures that can be considered in making a diagnosis of DH, starting with a skin biopsy. Tissue can be taken from a vesicle or excoriated papule and sent for standard H&E analysis, with an additional biopsy from normal peri-lesional skin to be sent for immunofluorescence studies. Classic H&E findings in DH include a subepidermal bullae with sterile microabscesses present within one or more dermal papillae [29], although nonspecific histologic findings may be present in up to 22–37 % of cases [18, 30]. A much more sensitive test is that of direct immunofluorescence [DIF] of perilesional normal skin, which will demonstrate granular deposition of IgA in the dermal papillae in 97 % of cases [18]. Rare cases have been reported in the literature in which both the histology and DIF studies of patients with DH were negative [31]. In cases without classic H&E and immunofluorescence findings, the presence of circulating IgA anti-tissue transglutaminase antibodies and/or IgG/IgA anti-gliadin antibodies, in association with a pruritic eruption of erythematous papules and plaques, may raise the possibility of an atypical form of DH, and patients may respond to DH therapy. Serum testing for gliadin or transglutaminase antibodies is not completely sensitive, and the widely available IgA anti-tissue transglutaminase test will be positive in approximately 70 % of patients [18, 25]. Tests which can measure IgA anti-epidermal transglutaminase are more sensitive [95 %] [25], but are not widely commercially available.