Comparison of effectiveness of interventions in reducing mortality in patients of toxic epidermal necrolysis: A network meta-analysis

Background: Limited evidence is available about effectiveness and choice of immunomodulating treatment modalities for toxic epidermal necrolysis (TEN). Aims: To compare the effectiveness of interventions to reduce mortality in patients of toxic epidermal necrolysis through network metaanalysis. Methods: Studies were retrieved using PubMed, Google Scholar and Cochrane Database of Systematic Reviews from inception to September 18, 2018. Only English language articles were considered. Observational and randomized controlled studies having >= 5 TEN patients in each intervention arm were included. Two investigators independently extracted study characteristics, intervention details and mortality data. Bayesian network meta-analysis was performed using the Markov chain Monte Carlo (MCMC) approach through the random effect model. The ranking analysis was done to provide a hierarchy of interventions. The consistency between direct and indirect evidence was assessed through node spit analysis. The primary outcome was to compare the mortality [Odds ratio OR (95% credibility interval CrI)] among all treatment modalities of TEN. Results: Twenty- four studies satisfying the selection criteria were included. The network analysis showed improved survival with cyclosporine as compared to supportive care [OR- 0.19 (95% CrI: 0.05, 0.59)] and intravenous immunoglobulin [OR- 0.21 (95% CrI: 0.05, 0.76)]. The hierarchy of treatments based on "surface under the cumulative ranking curves" (SUCRA) value were cyclosporine (0.93), steroid+intravenous immunoglobulin (0.76), etanercept (0.59), steroids (0.46), intravenous immunoglobulin (0.40), supportive care (0.34) and thalidomide (0.02). No inconsistencies between direct and indirect estimates were observed for any of the treatment pairs. Limitations: Evidence is mainly based on retrospective studies. Conclusion: The use of cyclosporine can reduce mortality in TEN patients. Other promising immunomodulators could be steroid+intravenous immunoglobulin combination and etanercept.