Evaluation of serum cytokine levels in toxic epidermal necrolysis and Stevens–Johnson syndrome compared with other delayed‐type adverse drug reactions

Dear Editor, Toxic epidermal necrolysis (TEN) and Stevens–Johnson syndrome (SJS) are life-threatening disorders characterized by destruction of the epidermis and mucosal epithelium. They are mainly observed as consequences of adverse drug reactions (ADR) and are considered variants of the same disorder differentiated by the extent of the body surface area involved. It has been suggested that several pathways are implicated in the widespread apoptosis of keratinocytes in SJS ⁄TEN. Involvement of cytotoxic T cells and the molecular cytotoxicity of Fas and cytotoxic enzymes including granzyme B, perforine and granulysin have been shown in SJS ⁄TEN. Cytokines including tumor necrosis factor (TNF)-a and interferon (IFN)-c were found to be overexpressed in the lesional skin. It has also been shown that IFN-c can activate production of TNF-a by keratinocytes in the skin. Plasma exchange has been reported as a useful method of TEN treatment. In TEN patients, plasma exchange can effectively remove from patients’ serum not only drugs and drug metabolites, but also increased amounts of soluble Fas ligand and cytokines. However, in patients with SJS ⁄TEN, production of these cytokines were shown mainly in the skin lesions and blister fluids and only a few reports have shown an increase of these cytokines in the patient serum. In addition, there are few reports published about cytokine production other than IFN-c and TNF-a in SJS ⁄TEN patients. The aim of the present study was to evaluate a possible role of serum cytokines in SJS ⁄TEN by comparing their concentration levels among delayedtype ADR. Eight SJS (male : female ratio, 4:4; mean age, 52.9 years; age range, 23–74 years) and three TEN (three males; mean age, 51.0 years; age range, 4–75 years) patients were included in this study. These patients were admitted to either one of our two hospitals and treated with corticosteroid therapy with or without plasma exchange. Maximum epidermal detachment was 30–70% of body surface area in the TEN patients. All of these patients recovered. Sera were obtained from these patients at the onset (within 3 days of hospitalization) before starting treatment with high-dose systemic corticosteroids and plasma exchange. One SJS patient with systemic lupus erythematodes had been treated with daily 20 mg prednisolone for 1 year before initiating SJS treatment. In order to compare the cytokine levels among ADR types, 34 patients with other types of generalized delayed-type ADR were also included in this study. Their final diagnoses were as follows; macropapular type (MP), 15 patients (male : female ratio, 7:8; mean age, 58.4 years), erythema multiforme type (EM), 12 patients (male : female ratio, 6:6; mean age, 60.7 years), and drug-induced hypersensitivity syndrome (DIHS) or drug rash with eosinophilia and systemic symptoms (DRESS), which is another systemic life-threatening ADR that is mostly associated with reactivation of human herpesvirus 6, six patients (male : female ratio, 3:3; mean age, 48.7 years). Obtained sera were stored at )80 C until cytokine measurement. The cytokine levels were measured using the BioPlex suspension array system (Bio-Rad, San Francisco, CA, USA). Results were expressed as the mean ± standard error of the mean. Statistical analysis was performed using Mann–Whitney’s U-test, with P < 0.05 considered significant. Serum cytokine levels in normal volunteers (n = 13) were as follows: IFN-c, 54.9 ± 16.8 pg ⁄mL; TNF-a, 13.6 ± 10.0 pg ⁄mL; interleukin (IL)-10, 3.8 ± 4.4 pg ⁄mL; IL-1