To the Editor: In August 2009, a 52-year-old man presented with chronic cough and multifocal lymphadenopathy of the left neck. After sputum culture and computerized tomography scan, he was diagnosed with pulmonary tuberculosis. He was initially treated with isoniazid, ethambutol, rifampin, and pyrazinamide; however, sputum cultures showed resistance to isoniazid, so isoniazid was changed to levofloxacin. Two weeks after initiation of antituberculous drugs, an erythematous to violaceous eruption developed. After cervical lymph node biopsy, the patient was further diagnosed with nasopharyngeal carcinoma and treated with radiotherapy and cisplatin-based chemotherapy. The eruption persisted for 5 months, at which time the patient was referred to the dermatology department. Physical examination revealed generalized dusky purpuric to hyperpigmented macules and some lichenoid papules and his buccal mucosa and tongue showed hyperpigmented lesions (Fig. 1). The clinical diagnosis included drug eruption and erythema dyschromicum perstans. A punch biopsy showed diffuse basal cell vacuolar degeneration and satellite cell necrosis in the epidermis (Fig. 2) and in the follicular epithelium. Superficial perivascular lymphocytes and eosinophils intermingled with abundant melanophages were observed. Lichenoid drug eruption was diagnosed, and antihistamines and topical corticosteroid were prescribed. Cisplatin-based chemotherapy was completed in January 2010, but the patient's skin eruption and mouth burning persisted. After successful treatment of his tuberculosis, the patient discontinued antituberculous drugs in April 2011 and the skin lesions rapidly healed.FIGURE 1: Lichenoid, hyperpigmented macules, and plaques on the trunk and oral mucosa.FIGURE 2: A and B, Hyperkeratosis, parakeratosis, acanthosis, and basal vacuolization in the epidermis. The underlying dermis contains perivascular infiltration of lymphocytes and eosinophils intermingled with abundant melanophages (hematoxylin–eosin–saffron stain, ×100). C and D, These findings also seemed in the follicular epithelium (hematoxylin–eosin–saffron stain; C: ×100; D: ×200).Unlike idiopathic lichen planus, which typically involves flexural surfaces, lichenoid drug eruption is characterized by an extensive symmetric eruption of flat-topped violaceous plaques involving the trunk and extremities.1,2 Lichenoid drug eruption does not exhibit classic Wickham striae and commonly implicated drugs include gold, β-blockers, antimalarials, thiazide diuretics, and penicillamine.1,2 Mucous membrane involvement is less common and is often associated with specific drugs, including allopurinol, angiotensin converting enzyme inhibitors, cyanamide, gold, ketoconazole, nonsteroidal antiinflammatory drugs, penicillamines, sulphonylurea, and carbamazepine.1,2 The literature includes only a few cases of lichenoid eruption induced by ethambutol and pyrazinamide, and none have shown mucosal involvement.3,4 Although our patient had been receiving many drugs, the rapid response to withdrawal of antituberculous drugs supports that this treatment caused the eruption. The histopathologic findings in our patient were also consistent with lichenoid drug eruption. Clues to the drug-induced etiology include epidermal parakeratosis, absence of wedge-shaped hypergranulosis, transepidermal necrotic keratinocytes, deeper mid-dermal perivascular and periadnexal infiltrates, and the presence of eosinophils.1,2,5 The finding of follicular dyskeratosis is typical in graft versus host disease (GVHD) but is very rare in lichenoid drug eruption. The histological differential diagnosis includes erythema dyschromicum perstans, GVHD, and chemotherapy-induced interface dermatitis. Erythema dyschromicum perstans is characterized by mild vacuolar degeneration and perivascular infiltrates followed by melanophages. Follicular involvement is frequent in GVHD; however, GVHD develops after hematopoietic stem cell transplantation or transfusion. Although follicular involvement can be seen in chemotherapy-induced eruption, this diagnosis was excluded because the patient's skin eruption persisted despite withdrawal of chemotherapy. To the best of our knowledge, this is the first report of antituberculous drug treatment being associated with lichenoid drug eruption with oral involvement and follicular dyskeratosis.
Read full abstract