Evaluation of the Factors Influencing Mortality in Patients with Stevens-Johnson Syndrome and Toxic Epidermal Necrolysis: A Multicenter Study of 166 Patients.

Abstract

Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN) are life-threatening acute mucocutaneous disorders usually triggered by drugs. In this study, we aimed to evaluate the factors affecting mortality in patients with SJS-TEN. Our study is a retrospective cohort study, analyzing data collected from a total of 12 tertiary care centers between April 2012 and April 2022. The study included 59 males and 107 females, a total of 166 patients, with an average age of 50.91 ± 21.25years. Disease classification was TEN in 50% of cases, SJS in 33.1%, and SJS-TEN overlap in 16.9%. The average SCORTEN within the first 24h was 2.44 ± 1.42. Supportive care was provided to 99.4% of patients. The most commonly used systemic immunomodulatory treatments were systemic steroids (84.3%), IVIG (intravenous immunoglobulin) (49.3%), and cyclosporine (38.6%). Plasmapheresis was administered to five patients. While 66.3% of patients were discharged, 24.1% resulted in exitus. Our comparative analysis of survivors and deceased patients found no effect of systemic steroids, IVIG, and cyclosporine treatments on mortality. Univariate analysis revealed that the SCORTEN scores on days 1 and 3 as well as the rates of detachment at the onset and during follow-up were significantly higher in deceased patients compared to survivors. The rates of fever, positive blood cultures, and systemic antibiotic use were higher in deceased patients compared to survivors. The presence of comorbidities, diabetes, and malignancy were significantly more common in deceased patients. Multivariate regression analysis indicated that over SCORTEN 2, the mortality risk exponentially rose with each SCORTEN increment, culminating in an 84-fold increase in mortality at SCORTEN 5-6 (odds ratio [95% confidence interval]: 13.902-507.537, p < 0.001) compared to SCORTEN 0-1. Additionally, the utilization of plasmapheresis was associated with a 22-fold increase in mortality (odds ratio [95% confidence interval]: 1.96-247.2, p = 0.012). Our study found that a high SCORTEN score within the first 24h and the use of plasmapheresis were related to increased mortality, while systemic steroids, IVIG, and cyclosporine treatments had no impact on mortality. We believe that data gathered from one of the most comprehensive studies which we conducted on SJS-TEN will enrich the literature, although additional research is warranted.