Postmarketing surveillance of serious adverse events associated with the use of rofecoxib from 1999–2002

ABSTRACT

Objective: The authors performed a postmarketing database analysis to evaluate the incidence of cardiovascular and other serious adverse events (SAEs) reported to the US Food and Drug Administration's Adverse Event Reporting System (AERS) involving the use of rofecoxib.

Research design and methods: The authors studied all adverse events involving rofecoxib reported to the AERS from inception of the drug until 2002. An emphasis was placed on cardiovascular and other SAEs of interest categorized using the high level group terms hemorrhage, edema, thrombosis, embolism, and death.

Results: There were 31,024 reports of SAEs associated with the use of rofecoxib, which was considered primary suspect in 97.8% of these reports. There were 3915, 3677, 1653, 1917, and 233 reports of hemorrhage, edema, death, thrombosis, and embolism, respectively. Relative to the overall population in this dataset, reports of hemorrhage, death, thrombosis, and embolism consisted of a greater proportion of males. The mean age for patients that reported hemorrhage, death, and thrombosis was older, whereas the mean age for embolism and edema was younger than the overall population in this dataset. Aspirin was the most commonly reported concomitant medication (7.4% of reports) followed by acetaminophen (5.4%). Reports containing concomitant use of anti-coagulants, Cox-1, and nonselective inhibitors (each p < 0.001) but not Cox-2 inhibitors were associated with increased age while only anti-coagulants and Cox-1 inhibitors (each p < 0.001) were associated with more males. Reports containing concomitant use of an anti-coagulant or an NSAID accounted for a disproportionate incidence of hemorrhage, edema, embolism, and death. Most notably, the odds of a hemorrhagic event for those reporting concomitant use of an anti-coagulant, Cox-1, non-selective, or Cox-2 inhibitor was 3.05 (p < 0.001), 3.07 (p < 0.001), 2.07 (p < 0.001), and 1.18 (p < 0.001), respectively. Some weaknesses of this type of analysis are the retrospective nature of such a study, the inability to find causality, and that the data can contains multiple reports from any one individual.

Conclusions: It can be postulated that in addition to the risk of heart attack and stroke, rofecoxib users were at increased risk of hemorrhage, in addition to other thrombotic and embolic adverse events, which was exacerbated in those taking blood thinners or NSAIDs.

Key words: :

• Cyclooxygenase-2
• Drug toxicity
• Hemorrhage
• Postmarketing
• Product surveillance
• Rofecoxib
• Thrombosis

Transparency

Declaration of funding

This trial was not financially supported by Duke University or any federal or corporate sponsors.

Declaration of financial/other relationships

F.G.B. has disclosed that he is owner of Scriptorium Medica Medical Writing, Inc.; he has declared no relevant financial relationships with regard to this study. W.T.B. has disclosed no relevant financial relationships.

All peer reviewers receive honoraria from CMRO for their review work. Peer Reviewer 1 has disclosed that he/she is a stockholder of Merck & Co. Inc and is an occasional member of the speaker bureau for Merck & Co. Peer Reviewer 2 has disclosed he/she has no relevant financial relationships.

Acknowledgments

No third-party editorial assistance was provided during the preparation of this manuscript. The authors thank Garret A. Fitzgerald, MD of the Institute for Translational Medicine and Therapeutics, University of Pennsylvania, Philadelphia, PA for his valuable comments and suggestions regarding this manuscript; and Peter Stephens of IMS Health, London, UK, for supplying the IMS Health data.

Notes

* Vioxx is a registered trademark of Merck and Co., Whitehouse Station, NJ, USA.