Stevens-Johnson syndrome and toxic epidermal necrolysis: a 20-year single-center experience

Background Stevens–Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) are life‐threatening diseases that are most frequently caused by drugs. Objectives The purpose of this study was to summarize 20 years of experience with SJS and TEN in the largest dermatology clinic in Serbia. Methods The study included 38 patients treated during the period 1993–2012. The patients were classified into three groups according to whether they were diagnosed with SJS, a condition representing an overlap of SJS and TEN (SJS/TEN), or TEN. Patients with TEN were also divided into three groups according to the modality of therapy: supportive therapy (ST) only (n = 3); ST plus systemic corticosteroids (SC) (n = 8); and ST plus SC plus IV immunoglobulins (IVIG) (n = 6). Results The study population included 13 SJS patients, eight SJS/TEN patients, and 17 TEN patients. The disease had started at a mean ± standard deviation (SD) of 7.1 ± 3.5 days after the commencement of treatment with the offending drug. The disease resulted in three lethal outcomes, all of which occurred in TEN patients. However, the predicted mortality for the whole group was 5.6 in 38 patients, whereas that for the TEN group was 3.97 in 17 patients. The differences between actual and predicted rates of mortality were not significant. Among the three groups of TEN patients, there were no significant differences in the commencement of re‐epithelialization or the duration of hospitalization. Conclusions In the present study, nonsteroidal anti‐inflammatory and anti‐infective drugs were the most frequent causative agents (eight patients in each group). In the group of SJS and SJS/TEN patients treated with ST and SC, the mortality rate was 0%. In TEN patients, the mortality rate was 17.6% (three of 17 patients). There were no significant differences in mortality rate among the three TEN treatment groups, but the results may have been biased by the small number of patients.