Skip to main content
Open Access Publications from the University of California

Dermatology Online Journal

Dermatology Online Journal bannerUC Davis

Dermatitis herpetiformis: Potential for confusion with linear IgA bullous dermatosis on direct immunofluorescence

Main Content

Dermatitis herpetiformis: Potential for confusion with linear IgA bullous dermatosis on direct immunofluorescence
Livia Van MD, John C Browning MD, Ravi S Krishnan MD, Brandi M Kenner-Bell MD, Sylvia Hsu MD
Dermatology Online Journal 14 (1): 21

Department of Dermatology, Baylor College of Medicine, Houston, Texas

Dermatitis herpetiformis (DH) and linear IgA bullous dermatosis (LABD) are two distinct cutaneous bullous diseases that are traditionally differentiated by direct immunofluorescence (DIF). Classically, in DH there is granular IgA deposition along the dermoepidermal junction with concentration at the papillary tips. In contrast, LABD has linear IgA deposition along the cutaneous basement membrane zone [1, 2]. However, DIF alone offers little guidance in ambiguous cases, particularly when IgA deposition appears both granular and linear at the same time. In our experience, DH does not always have the classic granular deposition of IgA, but can be more linear, like LABD (Fig. 1). We recently saw a patient with clinical DH but the DIF was interpreted by the dermatopathologist as LABD, which helps to illustrate our point.

Clinical synopsis

A 35-year-old man presented to his dermatologist with a 6-month history of intensely pruritic vesicles on his face, posterior neck, elbows, and buttocks. He denied any gastrointestinal symptoms. His dermatologist informed him that her clinical impression was DH and performed two 4-mm punch biopsies, one for histopathology and one for DIF. The histopathology revealed a subepidermal vesicle with neutrophils and the DIF was interpreted as LABD by a dermatopathologist. With these results, his dermatologist informed the patient that he could eat all the gluten he wanted and started him on dapsone. The patient continued to develop new blisters despite being on dapsone 100 mg daily for several months. He was then referred to one of us (SH).

We performed another DIF, which was interpreted as DH by the same dermatolopathologist and tissue transglutaminase antibody testing was positive at 117 U (negative < 20 U, weak positive 20-30 U, strong positive > 30 U). The patient was then referred to a gastroenterologist who confirmed the diagnosis of celiac disease.


Figure 1
Figure 1. Direct immunofluorescence of skin specimen taken from a patient with DH showing granular-linear deposition of IgA along the dermoepidermal junction. (Courtesy of Robert E Jordon MD)

While both DH and LABD skin lesions show dramatic improvement with dapsone or sulfapyridine treatment, initiating drug treatment without correctly distinguishing between DH and LABD can result in negative consequences for the patient. For instance, a patient with DH may not be started on a gluten-free diet and would be at risk for intestinal lymphoma [1]. Dermatitis herpetiformis patients should also be screened for thyroid disease, a co-morbid condition. A patient with LABD may be maintained on an offending medication that could easily be discontinued, reversing the manifestations of the disease.

Dermatitis herpetiformis is characterized by severely pruritic papules and vesicles, frequently presenting as erosions secondary to scratching, in a herpetiform configuration on symmetric extensor surfaces of the elbows, knees, shoulders, buttocks, and posterior neck. The cutaneous manifestation of gluten sensitive enteropathy (GSE), DH has an estimated incidence of 1 case per 10,000 people [3]. Most DH patients have intestinal histopathological changes even though only a minority will experience overt gastrointestinal symptoms [1, 4]. In fact, the severity of DH is inversely proportional to the severity of the intestinal symptoms (i.e., severe DH is associated with mild intestinal disease). Far rarer in occurrence, LABD is characterized by grouped pruritic papules and vesicles (and sometimes larger bullae that mimic bullous pemphigoid) that appear on symmetric extensor surfaces but do not necessarily favor the knees, elbows, buttocks, and neck. LABD patients more frequently present with mucosal lesions than do patients with DH [2]. Both diseases demonstrate an increased incidence of patients with histocompatibility locus antigens B8, DR3 and DQ2 [5].

Patients with DH produce IgA antibodies against endomysium [6]. A number of studies have since shown anti-endomysial antibody to be a highly sensitive (90%) and specific (96%) marker for detecting untreated DH and celiac disease [1, 6]. Another useful clinical marker in DH is anti-tissue transglutaminase IgA antibody. Tissue translutaminase (tTG) is the major target antigen of the anti-endomysial antibodies. Enzyme-linked immunosorbent assay (ELISA) for IgA antibodies against tTG yield a sensitivity and specificity comparable to EMA testing [7].

Anti-tTG IgA levels correlate with anti-endomysial antibodies and tTG-ELISA has the added benefit of lower cost ($100 instead of $130), faster results (1 instead of 2 days) and no need for an animal substrate [1, 7, 8]. As an investigator-independent, easy to use, noninvasive, and reproducible method, some experts have suggested tTG-ELISA should replace DIF as a first step in DH diagnosis [7, 8, 9].

As our case and others demonstrate [10, 11], one should question DIF results that conflict with clinical findings and supplement such cases with serologic testing. Dermatologists should not mistakenly change their clinical diagnosis of DH to LABD if DIF studies are read as consistent with LABD, because DH may appear linear under DIF.


1. Nicolas ME, Krause PK, Gibson LE, Murray JA. Review: Dermatitis herpetiformis. Int J Dermatol 2003; 42: 588-600.

2. Hall RP. "Linear IgA Dermatosis and Chronic Bullous Disease of Childhood." Fitzpatrick's Dermatology in General Medicine, 5th ed. 1999; 680-685. McGraw-Hill.

3. Bickle KM, Roark RM, Hsu S. Autoimmune Bullous Dermatoses: A Review. Am Family Physician 2002; 65(9): 1861-1870.

4. Heil PM, Volc-Platzer B, Karlhofer F, et al. Transglutaminases as diagnostically relevant autoantigens in patients with gluten sensitivity. J Dtsch Dermatol Ges 2005; 3:974-978.

5. Collier PM, Wojnarowska F, Welsh K, et al. Adult linear IgA disease and chronic bullous disease of childhood: the association with human lymphocyte antigens Cw7, B8, DR3 and TNF influences disease expression. Brit J Dermatol 1999; 141: 867-875.

6. Chorzelski TP, Beutner EH, Sulej J, et al. IgA antiendomysium antibody. A new immunological marker of dermatitis herpetiformis and coeliac disease. Brit J Dermatol 1984; 111(4): 395-402.

7. Dieterich W, Laag E, Bruckner-Tuderman L, et al. Antibodies to Tissue Transglutaminase as Serologic Markers in Paitents with Dermatitis Herpetiformis. J Invest Dermatol 1999; 13(1): 133-136.

8. Rose C, Dieterich W, Brocker EB, et al. Circulating autoantibodies to tissue transglutaminase differentiate patients with dermatitis herpetiformis from those with linear IgA disease. J Am Acad Dermatol 1999; 41(6): 957-961.

9. Desai AM, Krishnan RS, Hsu S. Medical Pearl: Using tissue transglutaminase antibodies to diagnose dermatitis herpetiformis. J Am Acad Dermatol 2005; 53:867-869.

10. Beutner EH, Baughman RD, et al. A case of dermatitis herpetiformis with IgA endomysial antibodies but negative direct immunofluorescent findings. J Am Acad Dermatol 2000; 43(2): 329-332.

11. Sousa L, Bajanca R, Cabral J, Fiadeiro T. Dermatitis Herpetiformis: Should Direct Immunofluorescence Be the only diagnostic Criterion? Pedi Dermatol 2002; 19(4): 336-339.

© 2008 Dermatology Online Journal