Neutrophilic eccrine hidradenitis in actinic reticuloid syndrome

ARTICLE

Neutrophilic eccrine hidradenitis (NEH) is an acute inflammatory reaction that occurs most commonly in patients receiving cytarabine induction chemotherapy for acute myelogenous leukemia (AML) [1, 2]. It is characterized clinically by erythematous papules, plaques, or solitary nodules and histologically by neutrophilic infiltrates of eccrine glands and degeneration of eccrine epithelial cells. Although NEH was first described in a patient with AML, it was found subsequently that NEH might occur in patients with various types of malignancies who were receiving chemotherapy. It has been speculated that NEH is due to some toxic effect of chemotherapeutic agents excreted and perhaps even concentrated in the eccrine glands. However, NEH also occurred in HIV-infected patients treated with zidovudine [3] or stavudine [4] and more rarely a similar condition might be occur spontaneously [5] or in conjunction with a bacterial infection [6-9]. Since NEH sometimes precedes the onset of leukemia [10, 11] or of a solid tumor [12, 13], it is postulated that degenerative changes in the eccrine apparatus might result in the accumulation of neutrophils. Therefore, NEH might be a paraneoplastic syndrome characterized by neutrophilic infiltration such as Sweet's syndrome and atypical pyoderma gangrenosum in patients with hematological abnormalities.

We report the first patient with NEH associated with actinic reticuloid syndrome (AR), who developed a skin rash two weeks after cessation of a four-month course of methotrexate therapy.

Case report

A 54-year-old male with a five-year history of pruritus and dermatitis on sun-exposed areas had been treated with topical steroids and oral antihistamines without benefit. He had no history of systemic exposure to photosensitizers or oral administration of drugs capable of causing photoallergic and/or phototoxic reactions or pseudolymphoma. He had not noticed any reaction to sun exposure. When he visited our clinic, he presented with erythroderma with plaque lesions, and lymphadenopathies on his neck, axillae, and groin. A biopsy from one of the plaques showed histopathologic features consistent with those of mycosis fungoides (MF); diffuse hyperkeratosis with pronounced acanthosis, subepidermal band-like infiltrate consisting mainly of lymphocytes with a few atypical cells and exocytosis of lymphocytes with Pautrier's microabscess formation (Fig. 1). However, unlike typical MF, CD8⁺ lymphocytes were predominantly infiltrating in the epidermis (data not shown). A lymph node biopsy from the inguinal area showed characteristics of dermatopathic lymphadenitis. Neither skin nor lymph node biopsies demonstrated clonal rearrangement of T cell receptor (TCR) genes by Southern blotting using a TCR-Cbeta probe. Laboratory results revealed a white blood cell count of 10,800/mul, with 30% lymphocytes, 18% eosinophils, and 2% atypical lymphocytes. Biochemical parameters were within normal limits except for elevated levels of lactate dehydrogeneas (1,050 IU/l, normal range 250-410 IU/l) and C-reactive protein (3.3 mg/dl, normal range < 0.3 mg/dl). Serum sIL-2R (3,170 U/ml, standard value 246~742 U/ml) and IgE (1,880 U/ml, standard value 27.5-138.3 U/ml) were markedly elevated. Serum anti HTLV-1 antibody was negative. Immunophenotyping of peripheral blood lymphocytes revealed a normal CD4/CD8 ratio (1.33) and a slightly increased number of CD8 positive cells (36.2%, normal value 19~32%). We considered both MF and erythrodermic AR, although we could not evaluate the results of phototesting because of the severe erythrodermic lesions. Since he was not aware of photosensitivity, he was treated with oral prednisolone (15 mg/day) and a trial of 8-MOP-mediated photochemotherapy (PUVA) since we suspected erythrodermic MF. The PUVA-treated lesions became exacerbated after three trials, and PUVA treatment was discontinued. A diagnosis of erythrodermic AR was considered. Further diagnostic phototesting was impossible because of severe desquamative erythroderma. The treatment was changed to intravenous interferon-gamma therapy, (2 x 10⁶ IU, five times weekly) with oral prednisolone (15 mg/day), followed by a low-dose of oral methotrexate (7.5-15 mg/week) together with a maintenance dose of oral prednisolone of 15 mg. Because the erythroderma gradually improved, we began to taper the dosage of methotrexate.

Two weeks after the discontinuation of oral methotrexate with a cumulative dose of 195 mg, multiple, slightly tender, erythematous papules appeared suddenly on his upper and lower extremities and trunk (Fig. 2). The differential diagnoses included lymphomatoid papulosis, folliculitis, deep mycosis, miliaria profunda, and NEH.

A biopsy from the erythematous papules on his right forearm was obtained. Microscopic examination of the serially sectioned specimen revealed a mixed inflammatory infiltrate composed of many neutrophils around the eccrine sweat apparatus. The eccrine epithelial cells showed vacuolar degeneration and cytoplasmic eosinophilia associated with exocytosis of neutrophils (Fig. 3). Although acanthosis and elongation of rete ridges remained in the epidermis, dense lymphocytic infiltrates that had been observed in the initial erythroderma had decreased. There was no evidence of vasculitis. Neither fungi nor bacteria were detected by PAS and Grocott staining. There was no growth of bacteria or fungi in several cultures obtained from the papules.

Based on the clinical and histological findings, we made a diagnosis of NEH. The eruptions resolved rapidly within five or six days, but new lesions appeared two weeks later. Similar episodes occurred three times and the eruptions resolved without any additional treatment, except for the maintenance dose of prednisolone.

Both erythroderma and the recurrence of NEH improved in response to oral prednisolone, leaving chronic lichenified lesions on the sun-exposed areas including the face, neck, and hands. Since there was the presence of photosensitivity dermatitis, we performed phototesting on the normal-appearing skin. The phototesting demonstrated a shortening of the minimal erythema dose to UV-B (15 mJ/cm², normal value 50-200 mJ/cm²) and UV-A induced erythema (2.52 J/cm²). He was finally diagnosed as having erythrodermic AR associated with NEH.

Discussion

The final diagnosis of erythrodermic AR was made based on the following clinical and histological findings. Exacerbation of the eruptions following PUVA treatment, the development of chronic lichenified lesions on the sun-exposed skin, photosensitivity to UV-B and UV-A, and the predominance of CD8⁺ cells in a band-like infiltration in the upper dermis [14-16].

During the clinical course, the patient had transient, multiple similar-sized papules on his trunk and extremities (Fig. 2). Microscopic examination revealed striking alterations in the eccrine sweat gland apparatus, such as epithelial necrosis and infiltration by neutrophils (Fig. 3). These features were consistent with the clinicopathologic features of NEH. Although there was no evidence of pathogenic organisms or provocative factors, the cessation of methotrexate might have been a significant factor.

Since he had been treated with low doses of systemic methotrexate and oral prednisolone, these may have had a pathogenic link to the development of NEH. Most previously reported cases of NEH occurred in patients with an underlying malignancy who were receiving a variety of chemotherapeutic agents. Such patients mainly had leukemia [1, 2] or lymphoma [17, 18], and sometimes they had a genitourinary tumor [12, 20], breast cancer [21], or osteosarcoma [13]. The causative chemotherapeutic agents might be cytarabine, doxorubicin and its related compounds (daunorbicin and daunoblastin), mitoxantrone, bleomycin, cyclophosphamide [1, 2, 12, 13, 17-20], minocycline [21], zidovudine [3], stavudine [4], granulocyte colony-stimulating factor [22] or acetaminophen [5]. Several reports described in the absence of an underlying malignancy [5], having a bacterial origin [6-9] or HIV [3, 4] infection. Although the true pathogenic mechanism of NEH remains unknown, the T cell proliferative condition together with methotrexate-induced immunosuppression may have led to the development of NEH in our patient.

Article accepted on 6/01/02

REFERENCES

1. Harrist TJ, Fine JD, Berman RS, et al. Neutrophilic eccrine hidradenitis: a distinctive type of neutrophilic dermatosis associated with acute myelogeneous leukemia and chemotherapy. Arch Dermatol 1982; 118: 263-6.

2. Flynn TC, Harrist TJ, Murphy GF, et al. Neutrophilic eccrine hidradenitis: a distinctive rash associated with cytarabine therapy and acute leukemia. J Am Acad Dermatol 1984; 11: 584-90.

3. Smith KJ, Skelton HG, James WD, et al. Neutrophilic eccrine hidradenitis in HIV-infected patients. J Am Acad Dermatol 1990; 23: 945-7.

4. Krischer J, Rutschmann O, Vollenweider RS, et al. Neutrophilic eccrine hidradenitis in a patient with AIDS. J Dermatol 1998; 25: 199-200.

5. Kuttner BJ, Kurban RS. Neutrophilic eccrine hidradenitis in the absence of an underlying malignancy. Cutis 1988; 41: 403-5.

6. Moreno A, Barnadas MA, Ravella A, de Moragas JM. Infectious eccrine hidradenitis in a patient undergoing hemodialysis. Arch Dermatol 1985; 121: 1106-7.

7. Allegue F, Rocamoro A, Martin-Gonzales M, et al. Infectious eccrind Dermatol 1990; 22: 1119-20.

8. Taira JW, Gerber HB. Eccrine hidradenitis. Int J Dermatol 1992; 31: 433-4.

9. Combemale P, Faisant M, Azoulay-Petit C, et al. Neutrophilic eccrine hidradenitis secondary to infection with Serratia marcescens. Br J Dermatol 2000; 142: 784-8.

10. Pierson JC, Helm TN, Taylor JS, et al. Neutrophilic eccrine hidradenitis heralding the onset of acute myelogenous leukemia. Arch Dermatol 1993; 129: 791-2.

11. Saada V, Aractingi S, Leblond V. Hidradenite neutrophile eccrine associée à une rechute de leucémie aiguë myéloblastique. Ann Dermatol Venereol 1998; 125: 420-2.

12. Fitzpatrick JE, Bennion SD, Reed OM, Wilson T, Reddy VVB, Golitz L. Neutrophilic eccrine hidradenitis associated with induction chemotherapy. J Cutan Pathol 1987; 14: 272-8.

13. Scallan PJ, Kettler AH, Levy ML, Tschen JA. Neutrophilic eccrine hidroadenitis: evidence implicating bleomycin as a causative agent. Cancer 1988; 62: 2532-6.

14. Toonstra J, Henquet CJ, van Weelden H, van der Putte SC, van Volten WA. Actinic reticuloid: a clinical, photobiologic, histopathologic, and follow-up study of 16 patients. J Am Acad Dermatol 1989; 21: 205-14.

15. Rijlaarsdam JU, Scheffer E, Meijer CJLM, Willemze R. Cutaneous pseudo-T-cell lymphomas: a clinicopathologic study of 20 patients. Cancer 1992; 69: 717-24.

16. Ploysangam T, Breneman DL, and Mutasim DF. Cutaneous pseudolymphomas. J Am Acad Dermatol 1998; 38: 877-905.

17. Beutner KR, Packman CH, and Markowitch W. Neutrophilic eccrine hidradenitis associated with Hodgkin's disease and chemotherapy. Arc Dermatol 1986; 122: 809-11.

18. Bailey DL, Barron D, and Lucky AW. Neutrophilic eccrine hidradenitis: a case report and review of the literature. Pediatr Dermatol 1989; 6: 33-8.

19. Greenbaum BH, Heymann WR, Reid CS, Travis SF, Donaldson MH. Chemotherapy-associated eccrine hidradenitis: neutrophilic eccrine hidradenitis reevaluated: the role of neutrophilic infiltration. Med Pediatr Oncol 1988; 16: 351-5.

20. Burg G, Bieber T, Langecker P. Localized neutrophilic eccrine hidradenitis in mitoxantrone therapy: a typical side-effect of cytostatic drugs. Hautarzt. 1988; 39: 233-6.

21. Mouly F. Vignon-Pennamen MD, Morel P. Dermatose neutrophilique aigue associant des aspects d'hidradenite neutrophile eccrine et de syndrome de Sweet après prise de minocycline. Ann Dermatol Venereol 1997; 124: S51.

22. Okcu A, Soyer HP, Kaddu S. Ekkrine hidradenitis. Hautarzt. 1997; 48: 743-8.