Drug-induced Stevens-Johnson syndrome: a disproportionality analysis from the pharmacovigilance database of the World Health Organization

Stevens-Johnson syndrome is a rare but serious skin condition, which can lead to death. Stevens-Johnson syndrome is usually attributed to drug-induced reactions, thus making it vital for clinicians to prevent its occurrence by knowing the trigger drugs. The objective of this study was to comprehensively and systematically excavate the drugs that cause SJS to provide references for clinician.
This is an observational, retrospective study, conducting a disproportionality analysis. Where the Information Component (IC) method and Reporting odds ratio (ROR) are used to mine the drugs that cause SJS.
A total of 17,787,905 reports were extracted from VigiBase database, of which 25,051 reports were related to SJS. The 18–44 age group had the largest number of cases (N=7,973, 31.83%). A total of 295 drugs was detected as signals. Allopurinol (IC025/ROR025=5.86/69.84), phenytoin (IC025/ROR025=5.60/57.65) and carbamazepine (IC025/ROR025=5.25/43.88) were the top 3 strongest signals. Our study only considered the possibility of SJS caused by a single drug.        
Allopurinol, phenytoin and carbamazepine were three strongest signals. Garenoxaci, carbocisteine and dimetindene were strong signals, but there are no relevant cases reported on PubMed or specific SJS in labels, which need further study to verify.