| Author, date and country | Patient group | Study type (level of evidence) | Outcomes | Key results | Study weaknesses |
| Rasmussen, 1976, USA | 32 Children with Stevens–Johnson syndrome. 17 were given 40–80 mg/m2 a day prednisolone or equivalent with supportive care. 15 were given the same supportive care only | Retrospective controlled case series | Duration of hospitalisation | Case versus control: 21 versus 13 days (p<0.01). If those who suffered complications were removed: 15 versus 13 days (not significant) | Small numbers, not randomised, no mention of duration of steroid therapy |
| Complications | Control group 0, steroid group 4 severe infections, 3 GI bleeds needing transfusion, 1 seizure, 1 Cushingoid face, 1 pleural effusion, 1 severe leucopenia | ||||
| Kakourou T et al, 1997, Greece | 16 Children with Stevens–Johnson syndrome within 3 days of developing rash. 10 were given 4 mg/kg a day of methylprednisolone for 3–7 days and supportive therapy. 6 were given supportive therapy only | Prospective randomised controlled trial | Period of fever | Steroid versus control: 4.0 (SD 1.9) days versus 9.5 (SD 4.2) days (p = 0.01) | No blinding |
| Period of acute eruption | Steroid versus control: 7.0 (SD 5.3) days versus 9.8 (SD 3) days (p = 0.08) | Each group managed at different centres | |||
| Complications | Control 0, steroid group 3 minor skin superinfections | Small numbers | |||
| Wide confidence intervals |
GI, gastrointestinal.