Dermatitis herpetiformis is a chronic, extremely pruritic disease that is characterized by clusters of vesicles distributed symmetrically over the extensor surfaces of the knees and elbows, scalp, and the back of the neck. The face is affected infrequently and the oral mucosa only rarely. The disease responds dramatically to sulfones and sulfapyridine. The illness may begin at any age, but it most frequently appears in the second to fourth decades. Although the disease appears to be lifelong in the most individuals, it tends to become milder after it has been present for about 10 years. Spontaneous remissions persisting for several years or spontaneous cures have been recorded.
The histology of the skin lesions in the earliest phase is characterized by collections of neutrophils at the tips of the dermal papillae (microabscesses), often in association with separation of the papillary tips from overlying epidermis. Most patients with dermatitis herpetiformis diagnosed on the basis of clinical and histologic features and their response to therapy have granular deposits of IgA identified by direct immunofluorescence in the dermal papilae of their cutaneous lesions, in a 3-5 mm perilesional zone, and sometimes in more distant normal skin. It is now possible to make the diagnosis on formalin-fixed tissue with the use of an avidin-biotin-peroxidase technique that is more convenient than direct immunofluorescence on fresh tissue. Immunofluorescent markers of IgA deposition are neccessary for the diagnosis of dermatitis herpetiformis.
Dermatitis herpetiformis is associatted with mucosal changes in the small bowel that are indistinguishable from those of celiac sprue. The atrophic villous pattern in dermatitis herpetiformis ranges from mild to severe. It is more prominent in the proximal small bowel than in the distal portion as observed in celiac sprue, but the atrophy occurs in a spotty fashion as opposed to more universal involvement in sprue. The differences in extent and degree of mucosal involvement in dermatitis herpetiformis have been proposed as the reason for the rarity of clinical signs of malabsorption.
Data from several series have shown that chemical evidence of steatorrhea was present in 22 of 69 patients, but in only 6 were there clinical signs and symptoms of malabsoption. Katz and Strober found only 2 of 51 patients had symptoms of malabsoption. Dermatitis herpetiformis preceded or followed malabsoption by as long as 3 to 8 years. Ten percent to 33% of patients have had abnormalities of D-xylose excretion, and anemia secondary to iron and folate deficiencies. Abnormalities in glucose, bicarbonate and water absorption similar to those found in sprue also have been present.
Sulfones can produce remission of the skin lesions but have no effect on the enteropathy. However, a gluten-free diet in dermatitis herpetiformis will not only cause the intestinal lesions to revert to normal as in sprue, but in most patients it will induce a remission in the skin lesions or significantly reduce the dose of sulfones necessary to keep the skin clear. It requires 6 to 12 months to achieve these cutaneous effects with a gluten-free diet. Return to a gluten-containing diet is followed by a relapse of skin disease in 1 to 3 weeks. Some patients with dermatitis herpetiformis do not respond to a gluten-free diet. Curiously, IgA and C3 still can be found in the normal skin of patients who have responed to a gluten-free diet.
The pathogenesis of dermatitis herpetiformis is belived to involve the deposition of IgA at the dermal-epidermal jonction of the skin with subsequent activation of complement by the alternative pathway. Neutrophils are attracted to the site and the histologic lesion is generated. However, the presence of IgA alone is insufficient to initiate the formation of the lesions, and other factors are necessary although they still are undefined. It has been established that patch tests with anionic I, Cl, Br, F and SCN will initiate local lesions within 24 to 48 hours. Oral potassium iodide also can induce lesions and exacerbate the disease. It is not understood how the halogens bring this about. The significance of IgA deposition in the skin is likewise unclear. It has been proposed that the IgA is directed against gluten protein or cross-reacting antigens, and that it originates in the gastrointestinal tract. There is no evidence that circulating immune complexes, which are detectable in some patients, play a pathogenetic role in this disease.
Other evidence links dermatitis herpetiformis to sprue. Relatives of patients with dermatitis herpetiformis have been shown to have a high prevalence of villous artrophy assumed to be caused by celiac sprue. HLA-B8/DRw3/DQw2 are present with increased frequency in both sprue and dermatitis herpetiformis.
Fry. et al. and Doran et al. have reported that absorption studies with D-xylose and folic acid yielded false-positive results for malabsoption in patients with extensive eczema and psoriasis. The abnormal results from D-xylose testing have been attributed to decreased renal clearance, not malabsoption. The reason for the decreased renal clearance has not been studied. . Serum folate and D-xylose levels, which are common screening tests for malabsoption, cannot be used for this purpose in patients with extensive skin disease. Correction of the serum folate levels has not benefited the eruptions in these patients.