gms | German Medical Science

13. Jahrestagung der Gesellschaft für Arzneimittelforschung und Arzneimittelepidemiologie

Gesellschaft für Arzneimittelforschung und Arzneimittelepidemiologie

02.11. bis 03.11.2006, Berlin

Cox-2 inhibitors, other NSAIDs, and the risk of myocardial infarction in three populations in three countries (UK, Canada, and the US)

Meeting Abstract

  • corresponding author M.A. Lewis - EPES Epidemiology, Pharmacoepidemiology and Systems Research GmbH, Berlin
  • D.R. Miller - Center for Health Quality, Outcomes, and Economic Research, VA Medical Center, Bedford, MA, USA
  • E. Rahme - Division of Clinical Epidemiology, The Montreal General Hospital, McGill University Health Centre Research Institute, Montreal, Canada
  • D. Kühl-Habich - EPES Epidemiology, Pharmacoepidemiology and Systems Research GmbH, Berlin
  • F. Cunningham - Pharmacy Benefits Management Strategic Health Care Group, Hines VAMC, Chicago, Il, USA
  • A.T. Monfared - Division of Clinical Epidemiology, The Montreal General Hospital, McGill University Health Centre Research Institute, Montreal, Canada
  • D.W. Kaufman - Slone Epidemiology Center, Boston University, Boston, MA, USA
  • L. Rosenberg - Slone Epidemiology Center, Boston University, Boston, MA, USA
  • J. LeLorier - Centre de recherché de l'Hôtel-Dieu du CHUM, Unité de recherche en pharmaco-économie et pharmaco-épidémiologie, Montréal, Quebec, Canada.

Gesellschaft für Arzneimittelanwendungsforschung und Arzneimittelepidemiologie e.V. (GAA). 13. Jahrestagung der Gesellschaft für Arzneimittelforschung und Arzneimittelepidemiologie. Berlin, 02.-03.11.2006. Düsseldorf: German Medical Science GMS Publishing House; 2006. Doc06gaa16

The electronic version of this article is the complete one and can be found online at: http://www.egms.de/en/meetings/gaa2006/06gaa16.shtml

Published: October 30, 2006

© 2006 Lewis et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by-nc-nd/3.0/deed.en). You are free: to Share – to copy, distribute and transmit the work, provided the original author and source are credited.


Outline

Text

Background: Selective COX-2 inhibitors are associated with increased risk of myocardial infarction (MI). The effect of meloxicam, a semi-selective COX-2, and other NSAIDs, is not established.

Methods: Case-control studies were conducted with a uniform protocol to assess the risk of first MI among users of COX-2 inhibitors, meloxicam and other NSAIDs compared with non-users and with users of diclofenac in the General Practitioner’s Research Database (GPRD; UK), the Quebec Hospital and Prescription database (RAMQ; Canada), and the Veterans administration database (VA; US). Subjects were age 50 or older and had received one or more prescriptions for an NSAID. Information of 7,037 cases and 40,619 age and sex-matched controls in the GPRD, 34,574 cases and 138,068 controls in RAMQ, and 58,553 cases and 221,885 controls in the VA was analysed using conditional (matched) logistic regression with a standard set of adjustment factors for all populations.

Results: At baseline, the populations show moderate (GPRD), intermediate (RAMQ), and high (VA) levels of comorbidity and drug use. The adjusted matched odds ratio (OR) of MI for meloxicam compared with non-use is 0.99 (95% CI: 0.55 to 1.78) in the GPRD, 0.74 (0.60 to 0.91) in the RAMQ, and 0.96 (0.74 to 1.25) in the VA study. For rofecoxib, statistically significant increases are shown in the VA (1.46; 1.36 to 1.58) and in the RAMQ studies (1.09; 1.04 to 1.15), and an elevated point estimate in the GPRD (1.14; 0.76 to 1.69). A small but statistically significant increase is found for celecoxib (1.89; 1.23 to 2.89) in the GPRD and in the VA (1.24; 1.16 to 1.33), but not in the RAMQ (1.02; 0.98 to 1.07). The estimates for diclofenac against non-use are 1.22 (0.97 to 1.52) in the GPRD, 1.07 (0.99 to 1.14) in the RAMQ, and 1.23 (1.14 to 1.33) in the VA study. Reduced risk estimates are found for naproxen in the RAMQ (0.74; 0.68 to 0.80) and the VA (0.74; 0.68 to 0.81).

The adjusted estimate for the risk of MI for meloxicam use compared with diclofenac use is 0.46 (95% CI: 0.15 to 1.48) in the GPRD, 0.63 (0.49 to 0.81) in the RAMQ (22,352 cases, 89,408 controls), and 0.81 (0.50 to 1.32) in the VA (30,218 cases, 118,701 controls). A small increase in risk of MI is found for rofecoxib use compared with diclofenac use in the VA (1.23; 1.06 to 1.44) and in RAMQ (1.11; 1.01 to 1.21), but not in the GPRD (0.91; 0.35 to 2.33).

Conclusion: Despite a highly coordinated approach, small variations in results are seen across these studies. Rofecoxib use shows an increased MI risk relative to diclofenac use and to non-use in two studies. Compared with non-use, two of the populations each show small increases for celecoxib and reduced risks for naproxen. In all three populations, meloxicam shows no increased risk of MI against non-use, and a lower risk against diclofenac use.