BackgroundThe SCORTEN score is a specific predictor of mortality for patients with Stevens–Johnson syndrome (SJS)/toxic epidermal necrolysis (TEN). There is little evidence in support of the common immunomodulating therapies for SJS/TEN.ObjectivesTo systematically assess the effectiveness of several therapies for SJS/TEN through the SCORTEN score.MethodsDatabases were searched for original studies on the use of SCORTEN. Six meta‐analyses were carried out on patients with SJS/TEN who received supportive care only or in combination with immunomodulating drugs: corticosteroids, cyclosporine, etanercept, immunoglobulins or a combination of corticosteroids with immunoglobulins. A multivariate meta‐regression and a network meta‐analysis were also performed.ResultsOf 3893 studies identified, fifty‐two involving 2466 patients with SJS/TEN were preselected. Data from thirty‐eight of these studies (1827 patients) were finally pooled, and results [log(SMR)] from meta‐analyses were as follows: −0.13 (95% CI, −0.42,0.16) for corticosteroids, −0.39 (95% CI, −0.87,0.09) for immunoglobulins, 0.13 (95% CI, −0.15,0.40) for supportive treatment, −0.88 (95% CI, −1.47, −0.29) for cyclosporine, −0.95 (95% CI, −1.82, −0.07) for etanercept and − 0.56 (95% CI, −0.94, −0.19) for immunoglobulins plus corticosteroids. The meta‐regression analysis confirmed that cyclosporine and immunoglobulins plus corticosteroids were associated with less deaths than predicted by SCORTEN. In the network meta‐analysis, no treatment achieved a significant reduction in the SMR.LimitationsHeterogeneity and quality of the included studies.ConclusionsSome treatments for SJS/TEN show a better performance, but there is not sufficient evidence to recommend its widespread use in all patients.