Comparison of reporting of Stevens-Johnson syndrome and toxic epidermal necrolysis in association with selective COX-2 inhibitors

Abstract

Background: Stevens-Johnson syndrome and toxic epidermal necrolysis are closely related severe acute life-threatening, drug-induced skin disorders. The US FDA Adverse Events Reporting System (AERS) has received reports of Stevens-Johnson syndrome and toxic epidermal necrolysis associated with the use of the recently introduced selective cyclo-oxygenase (COX)-2 inhibitor NSAIDs, two of which are also sulfonamides.

Objective: The objective of this study is to review cases of Stevens-Johnson syndrome and toxic epidermal necrolysis reported to the FDA associated with the use of the selective COX-2 inhibitor NSAIDs celecoxib, rofecoxib and valdecoxib, and to compare reporting rates of the two conditions associated with these drugs to each other, meloxicam (an oxicam NSAID that came on the US market at a similar time) and the background incidence rate.

Methods: We reviewed all US cases of Stevens-Johnson syndrome and toxic epidermal necrolysis reported to the FDA AERS database associated with the use of celecoxib, rofecoxib, valdecoxib and meloxicam since these agents were first marketed. We utilised AERS and drug use data to calculate reporting rates for each drug after the first 2 years of marketing. We obtained the background rate from the medical literature.

Results: Up to the end of March 2004, there were 63 cases of Stevens-Johnson syndrome/toxic epidermal necrolysis reported with valdecoxib use, 43 with celecoxib, 17 with rofecoxib (the non-sulfonamide coxib) and none for meloxicam. In the first 2 years of marketing the reporting rate for Stevens-Johnson syndrome/toxic epidermal necrolysis with valdecoxib was 49 cases per million person-years of use, 6 cases per million person-years for celecoxib and 3 cases per million person-years for rofecoxib. The reporting rates for the sulfonamide coxibs were substantially higher than the background rate of 1.9 cases per million population per year, with the valdecoxib rate being 8-9 times that of celecoxib and approximately 25 times that of the background rate.

Conclusion: There is a strong association between Stevens-Johnson syndrome/toxic epidermal necrolysis and the use of the sulfonamide COX-2 inhibitors, particularly valdecoxib. Physicians should be aware of the possibility of this serious life-threatening event when prescribing these drugs and advise patients to discontinue use at the earliest possible sign or symptom.