Correspondence 479 7 Klotz SA, Drutz DJ, Huppert M, Johnson JE. Pityrosporum folliculitis. Its potential for confusion with skin lesions of systemic candidiasis. Arch Intern Med 1982; 142:2126–9. 8 Rosen T, Schell JB, Orengo I. Anti-inflammatory activity of antifungal preparations. Int J Dermatol 1997; 36:788–92. 9 Faergemann F, Bergbrant IM, Dohse M et al. Seborrhoeic dermatitis and Pityrosporum (Malassezia) folliculitis: characterization of inflammatory cells and mediators in the skin by immunohistochemistry. Br J Dermatol 2001; 144:549–56. 10 Hill MK, Goodfield JD, Rodgers FG et al. Skin surface electron microscopy in Pityrosporum folliculitis: the role of follicular occlusion in disease and the response to oral ketoconazole. Arch Dermatol 1990; 126:181–4.
Conflicts of interest: none declared.
Linear IgA disease presenting as prurigo nodularis DOI: 10.1111/j.1365-2133.2006.07315.x SIR, A 57-year-old white woman was referred to our clinic with a 7-year history of recurrent episodes of severe itching accompanied by eruption of nodular lesions on the trunk and limbs. Oral steroids and antihistamines and topical steroids had not produced any improvement of symptoms. In the preceding year, the eruption had been continuous without remission. The patient had a 4-year history of carpal tunnel syndrome and had a cholecystectomy at the age of 35 years. Examination showed many round, hyperpigmented nodules, studded with erosions or small serosanguineous crusts and distributed on the limbs, abdomen and lower back (Fig. 1). Laboratory tests revealed increased erythrocyte sedimentation
rate (78 mm in the first hour; normal 2–30), C-reactive protein (2Æ0 mg dL)1; normal < 0Æ5), fibrinogen (596 mg dL)1; normal 200–400), c-glutamyltransferase (70 U L)1; normal 5–36) and cholesterol (262 mg dL)1; normal 140–240). Chronic hepatitis B was diagnosed via detection of anti-HBc antibodies. Histopathology of a nodular lesion showed a subepidermal unilocular blistering dermatosis characterized by hyperkeratosis, focal parakeratosis and marked irregular acanthosis of the epidermis with central ulceration (Fig. 2a,b). The dense dermal inflammatory infiltrate was localized at the floor of the blister, in perivascular and interstitial sites (Fig. 2c). It was mainly composed of neutrophils, eosinophils and lymphohistiocytes. The number of small blood vessels and fibroblasts was increased. Direct immunofluorescence assay (DIF) on perilesional skin demonstrated linear deposits of IgA (Fig. 2d), C3 and fibrinogen along the basement membrane zone (BMZ). Indirect immunofluorescence assay (IIF), using monkey oesophagus and human healthy skin as substrates, was negative for serum IgG, IgA and IgM. Salt-split skin (SSS)DIF revealed linear deposits of IgA along the epidermal side of split perilesional skin, while SSS-IIF showed a weak, linear discontinuous deposit of IgA along the epidermal side of normal human skin. Immunoblot analysis (IB) of epidermal extracts detected several IgA bands, including one at 120 kDa (Fig. 2e). IB of dermal extracts was negative. On the basis of such findings, we made the diagnosis of linear IgA disease (LAD) with clinical features of prurigo nodularis (PN). Having excluded glucose-6-phosphate dehydrogenase deficiency, oral therapy was started with dapsone 50 mg daily, prednisone 25 mg daily and hydroxyzine 10 mg daily. During the following months, the patient reported reduction of itching and there was a progressive improvement in the clinical picture, with long-term remission of the cutaneous lesions. The patient described here presented with typical symptoms and signs of PN. Skin biopsy showed some features compatible with PN (i.e. hyperkeratosis, irregular acanthosis
(a)
(b)
(c)
Fig 1. Many prurigo nodularis-like lesions grouped on the lower back (a) and on the dorsum of the right foot (b); nodular lesion of recent onset (c). 2006 British Association of Dermatologists • British Journal of Dermatology 2006 155, pp477–500
480 Correspondence
(a)
(b)
(c)
(e)
(d)
Fig 2. (a) An in toto section of a nodular lesion shows hyperkeratosis, acanthosis, ulceration of the epidermis, and a dense infiltrate in the dermis (haematoxylin and eosin; objective · 2.5). (b) The edge of a subepidermal cleft that shows hyperkeratosis and marked irregular hyperplasia of the superficial epidermis. In the upper dermis there is a moderate increase of small vessels in a background of an inflammatory infiltrate in perivascular and interstitial sites (haematoxylin and eosin; objective · 10). (c) The inflammatory infiltrate is constituted by neutrophils, eosinophils and lymphohistiocytes in the interstitial dermis (haematoxylin and eosin; objective · 100). (d) Direct immunofluorescence assay shows linear deposition of IgA at the dermoepidermal junction of perilesional skin. (e) Immunoblotting analysis with the epidermal extract. Lane 1, standard molecular weights (150 and 250 kDa); lane 2, serum reactivity (IgA) at 120 kDa, slightly above 130 kDa, slightly below 180 kDa and at 250 kDa; lane 3, desmoglein 3 (130 kDa) control serum (IgG) from a patient with pemphigus vulgaris; lane 4, BP180 control serum (IgG) from a patient with bullous pemphigoid.
and a moderate increase in the number of small blood vessels and fibroblasts in the upper dermis). However, there was also the unexpected findings of dermoepidermal separation and a dense dermal infiltrate of neutrophils and eosinophils, suggesting the diagnosis of pemphigoid nodularis. Surprisingly, DIF revealed linear deposits of IgA and C3 at the BMZ. IIF was negative for both IgG and IgA, while SSSIIF stained serum IgA along the roof of the split. IB showed that such IgA autoantibodies were specific for an epidermal antigen of 120 kDa, i.e. LAD-1,1 that is frequently targeted in LAD. We interpreted the other bands identified by IgA serum as an expression of the ‘epitope spreading’ phenomenon.1 Histopathological and immunopathological features were consistent with the diagnosis of LAD, which was further supported by a good response to oral dapsone and prednisone, a combination that is effective in most cases of LAD.2 We hypothesize that the atypical presentation of this case of LAD was heavily influenced by itching and scratching, triggered by LAD itself and possibly by chronic hepatitis B, which is likely to have been present for several years, as testified by elevation of the anti-HBc titre only. We have reviewed the literature on the immunopathological features of other autoimmune subepidermal bullous dermatoses presenting as PN. We found two cases of pemphigoid nodularis showing linear deposits of IgA at the BMZ,3,4 but both featured, in addition, linear deposits of IgG and C3 in DIF and circulating anti-BMZ IgG at a high titre in IIF. Thus, these two reports can, in retrospect, be confirmed as ‘true’ bullous pemphigoid with clinical features of PN. Two possible cases of dermatitis herpetiformis presenting as PN have been reported,5,6 although the diagnosis was not confirmed by DIF.
Thus, to the best of our knowledge this is the first reported case of LAD presenting as PN. Department of Dermatological Sciences, University of Florence, Via della Pergola 58/60, 50121 Florence, Italy *DiSEM, Section of Dermatology, University of Genoa, Genoa, Italy Department of Human Pathology and Oncology, University of Florence, Florence, Italy Correspondence: Paolo Fabbri. E-mail:
[email protected]
D. TORCHIA M. CAPRONI E. DEL BIANCO E. COZZANI* S. KETABCHI P. FABBRI
References 1 Allen J, Wojnarowska F. Linear IgA disease: the IgA and IgG response to the epidermal antigens demonstrates that intermolecular epitope spreading is associated with IgA rather than IgG antibodies, and is more common in adults. Br J Dermatol 2003; 149:977–85. 2 Zone JJ. Clinical spectrum, pathogenesis and treatment of linear IgA bullous dermatosis. J Dermatol 2001; 28:651–3. 3 Tani M, Murata Y, Masaki H. Pemphigoid nodularis. J Am Acad Dermatol 1989; 21:1099–104. 4 Borradori L, Rybojad M, Verola O et al. Pemphigoid nodularis. Arch Dermatol 1990; 126:1522–3. 5 Rowland Payne CM, Wilkinson JD, McKee PH et al. Nodular prurigo – a clinicopathological study of 46 patients. Br J Dermatol 1985; 113:431–9. 6 Delfino M, Nino M, Delfino G et al. Dermatitis herpetiformis and gluten-sensitive enteropathy in a patient with nodular prurigo. J Eur Acad Dermatol Venereol 2002; 16:88–9.
Conflicts of interest: none declared.
2006 British Association of Dermatologists • British Journal of Dermatology 2006 155, pp477–500
