Stephen E. Kimmel, MD, MSCE; Jesse A. Berlin, ScD; Muredach Reilly, MB; Jane Jaskowiak, BSN, RN; Lori Kishel, MS; Jesse Chittams, MS; Brian L. Strom, MD, MPH
Acknowledgments: The authors thank Sandy Barile for editorial assistance, Max Herlim for programming assistance, and William H. Sauer, MD, for assistance with data coding.
Grant Support: By the National Institutes of Health (R01HL57312), Searle Pharmaceuticals, Inc. (now Pfizer, Inc.), and Merck & Co., Inc.
Potential Financial Conflicts of Interest: Consultancies: S.E. Kimmel (Pfizer Inc.); J.A. Berlin (Wyeth-Ayerst, Johnson and Johnson, and GlaxoSmithKline); B.L. Strom (Pfizer, Inc., Novartis, Whitehall Robins, Wyeth-Ayerst); Honoraria: M. Reilly (Merck Frosst Canada Ltd., GlaxoSmithKline); B.L. Strom (Abbott, Wyeth-Ayerst, Pfizer, Inc.); Expert testimony: B.L. Strom (Bayer, Parke-Davis, GlaxoSmithKline); Grants received: S.E. Kimmel (Pfizer, Inc., Merck & Co., Inc.); J.A. Berlin (McNeil, Pfizer, Inc., Merck & Co., Inc.); M. Reilly (Merck & Co., Inc., GlaxoSmithKline); B.L. Strom (Bayer, GlaxoSmithKline, Merck & Co., Inc., Pfizer, Inc., Wyeth-Ayerst, Pharmacia, Whitehall Robins).
Requests for Single Reprints: Stephen E. Kimmel, MD, MS, University of Pennsylvania School of Medicine, 717 Blockley Hall, 423 Guardian Drive, Philadelphia, PA 19104-6021.
Current Author Addresses: Dr. Kimmel: University of Pennsylvania School of Medicine, 717 Blockley Hall, 423 Guardian Drive, Philadelphia, PA 19104-6021.
Dr. Berlin: Research and Development, LLC, Johnson and Johnson Pharmaceutical, 1125 Trenton-Harbourton Road, Titusville, NJ 08560.
Dr. Reilly: Hospital University of Pennsylvania, 9 Founders, 3400 Spruce Street, Philadelphia, PA 19104.
Ms. Jaskowiak: University of Pennsylvania, 731 Blockley Hall, 423 Guardian Drive, Philadelphia, PA 19104.
Ms. Kishel: 220 Locust Street, Apt 9F, Philadelphia, PA 19106.
Mr. Chittams: University of Pennsylvania, 525GH Blockley Hall, 423 Guardian Drive, Philadelphia, PA 19104.
Dr. Strom: University of Pennsylvania, 824 Blockley Hall, 423 Guardian Drive, Philadelphia, PA 19104.
Author Contributions: Conception and design: S.E. Kimmel, J.A. Berlin, B.L. Strom.
Analysis and interpretation of the data: S.E. Kimmel, J.A. Berlin, M. Reilly, J. Jaskowiak, L. Kishel, J. Chittams.
Drafting of the article: S.E. Kimmel, J.A. Berlin, M. Reilly.
Critical revision of the article for important intellectual content: S.E. Kimmel, J.A. Berlin, M. Reilly, J. Jaskowiak, L. Kishel, B.L. Strom.
Final approval of the article: S.E. Kimmel, J.A. Berlin, M. Reilly, J. Jaskowiak, L. Kishel, J. Chittams, B.L. Strom.
Statistical expertise: J.A. Berlin, J. Chittams.
Obtaining of funding: S.E. Kimmel, J.A. Berlin.
Collection and assembly of data: J. Jaskowiak, L. Kishel.
Studies have postulated that cyclooxygenase-2 (COX-2) selective inhibitors affect cardiovascular risk through various mechanisms. Some of these mechanisms could increase risk (for example, inhibition of prostacyclin production), and some could decrease risk (for example, inhibition of inflammation).
To determine the effect of COX-2 inhibitors on risk for nonfatal myocardial infarction (MI).
36 hospitals in a 5-county area.
1718 case-patients with a first, nonfatal MI admitted to these hospitals and 6800 controls randomly selected from the same counties.
Self-reported medication use assessed through telephone interviews.
The adjusted odds ratio for MI among celecoxib users, relative to persons who did not use nonaspirin nonsteroidal anti-inflammatory drugs (NSAIDs), was 0.43 (95% CI, 0.23 to 0.79) compared with 1.16 (CI, 0.70 to 1.93) among rofecoxib users. The use of rofecoxib was associated with a statistically significant higher odds of MI compared with the use of celecoxib (adjusted odds ratio for rofecoxib vs. celecoxib, 2.72 [CI, 1.24 to 5.95]; P = 0.01). Nonselective NSAIDs were associated with a reduced odds of nonfatal MI relative to nonusers. Comparisons of COX-2 inhibitors with nonselective NSAIDs were the following: rofecoxib versus naproxen (odds ratio, 3.39 [CI, 1.37 to 8.40]) and celecoxib versus ibuprofen or diclofenac (odds ratio, 0.77 [CI, 0.40 to 1.48]).
The possibility of recall bias and uncontrolled confounding in this observational study limit the ability to make definitive conclusions. The association of celecoxib with a lower odds of MI could have occurred by chance. Only about 50% of eligible participants completed telephone interviews.
Celecoxib and rofecoxib were associated with different odds of MI. Cardiovascular effects among the COX-2 inhibitors seem different, but further studies, preferably randomized trials, are needed to fully understand the spectrum of effects of COX-2 inhibitors and potential differences among them.
Various cyclooxygenase-2 (COX-2) selective inhibitors may affect cardiovascular risk differently.
This case–control study from 36 hospitals in a 5-county area found that reported recent use of a COX-2 selective inhibitor was similar among adults with a first, nonfatal myocardial infarction (MI) and randomly selected community controls with no history of MI. However, reported use of rofecoxib was associated with a 2.72 (95% CI, 1.24 to 5.95) higher odds of MI than was use of celecoxib.
About 50% of eligible participants completed interviews. Although analyses controlled for potential confounders, recall bias and unmeasured factors related to MI risk could have affected results.
Table 1. Characteristics of Nonsteroidal Drug Use among Controls
Table 2. Characteristics of Nonsteroidal Use among Case-Patients
Table 3. Association of Cyclooxygenase-2 Inhibitor and Nonselective Nonsteroidal Anti-Inflammatory Drug Use within the Index Week with the Odds of Nonfatal Myocardial Infarction
The In the Clinic® slide sets are owned and copyrighted by the American College of Physicians (ACP). All text, graphics, trademarks, and other intellectual property incorporated into the slide sets remain the sole and exclusive property of the ACP. The slide sets may be used only by the person who downloads or purchases them and only for the purpose of presenting them during not-for-profit educational activities. Users may incorporate the entire slide set or selected individual slides into their own teaching presentations but may not alter the content of the slides in any way or remove the ACP copyright notice. Users may make print copies for use as hand-outs for the audience the user is personally addressing but may not otherwise reproduce or distribute the slides by any means or media, including but not limited to sending them as e-mail attachments, posting them on Internet or Intranet sites, publishing them in meeting proceedings, or making them available for sale or distribution in any unauthorized form, without the express written permission of the ACP. Unauthorized use of the In the Clinic slide sets will constitute copyright infringement.
David A. Nardone
VHA and OH&SU Portland
February 1, 2005
COX-2 and Perceptions of Conflict of Interest
David A. Nardone 6714 NE Copper Beech Drive Hillsboro, Oregon firstname.lastname@example.org 503-648-6565
To the Editor:
As I read the recently published information regarding cyclooxygenase-2 inhibitors in the Annals I was heartened (1-2). Since the latter part of 2004 there has been a dearth of reliable information for physicians to offer objective recommendations to patients and for patients to make informed decisions. However, when I reviewed the sections for grant support and conflicts of interest, I became dismayed. Four of the seven authors of the original article, one of the two editorial authors, and both lead authors offered at least one potential affiliation with a drug company. I think it odd that the Annals felt compelled to rely, to this extent, on these experts for addressing such an important clinical issue. Surely there are consultants, who do not possess these potentially negative characteristics and who are qualified to perform critical appraisals of the pharmaceutical literature. At this point I am conflicted. The analyses and advice rendered appear relevant and reliable, but I must wonder if the majority of these authors could be biased. Publishing these potential conflicts is an excellent idea, but I am not certain it is sufficient to allay the concerns of physicians and patients, who desire and deserve an unbiased opinion. Perceptions sometimes speak volumes.
1. Kimmel SE, Berlin JA, Reilly M, Jaskowiak J, Kishel L, Chittams J, Strom BL. Patients Exposed to Rofecoxib and Celecoxib Have Different Odds of Nonfatal Myocardial Infarction. Ann Intern Med 2005: 142: 157- 164
2. Finckh A, Aronson MD. Cardiovascular Risks of Cyclooxygenase-2 Inhibitors: Where Do We Stand Now? Ann Intern Med 2005;142: 212-214.
Sincerely, David A. Nardone, MD (Retired) Clinical Director Primary Care Staff Physician VHA Medical Center Professor Emeritus Oregon Health & Sciences University
"Civico e Benfratelli" National Relevance Hospital Trust - Palermo, Italy
March 5, 2005
Is Medicine based on evidences or merely on opinion? An example: cardiotoxicity of COX-2 inhibitors.
The cardiotoxicity of COX-2 inhibitors is becoming a topic of great interest, and cardiotoxicity as class effect or not is the question raised. The paper of Kimmel et al. (1) deals with this topic by examining whether the risk for cardiotoxicity differs among the COX-2 inhibitors celecoxib and rofexocib and nonselective nonaspirin NSAIDs. Seemingly, it is a work academically perfect with a large sample size. Moreover, I appreciated cautions on biases the editors expressed, but I believe all that is not enough. I previously remembered dangers of both misuse of statistics and study design biases before jumping to biologically implausible conclusions (2). In this case, there are no evident matters on plausibility, but this could be itself a confounding factor able to lower attention to important issues. Indeed, all patients and clinicians want to know whether cardiotoxicity is a class effect, and thereby applicable to any COX-2 inhibitor, or whether cardiotoxicity is limited to certain drugs in the class. Moreover, the editorial of Finckh et at (3) just affirmed that "the work of Kimmel and colleagues shed some light on this question". Another time, the danger of strongly biased conclusions looms up to the horizon, and the possibility to mold physician opinion likewise. This study was a case-control one based on a telephone interview, and all we know limitations of this kind of design. Nevertheless, there are almost other two main methodological problems that invalidate results of Kimmel et al work. Firstly, control-patients and case-ones did not match for principal characteristics: control-patients were younger, leaner, with different race and gender distribution etc.. But a heavy selection bias gets over matching failure: only about 50% of eligible participants completed interviews. We do not know anything about the second half of the selected sample, and so results could be totally reversed. Secondly, patients taking COX-2 inhibitors were scantly represented, particularly in the case-patient group (18 and 27 respectively for Celcoxib and Rofecoxib), and multivariable logistic regression assumptions were not entirely satisfied (4). In conclusion, the actual question I want to raise is: what is the role of relevant journals to oversee true methodological quality of papers? The question is not of little importance since strongly biased studies alike of sound ones published on relevant journals can influence physician opinion and prescription behaviour. Is there the danger of a clinical practice increasingly based on opinion dressed up as evidence?
1) Kimmel SE, Berlin JA, Reilly M, Jaskowiak J, Kishel L, hittams J, Strom BL. Patients exposed to rofecoxib and celecoxib have different odds of nonfatal myocardial infarction. Ann Intern Med. 2005 Feb 1;142(3):157- 64.
2) Corrao S. Protective effect of smoking: misleading use of statistics. Radiology. 2004 Dec;233(3):934
3) Finckh A, Aronson MD. Cardiovascular risks of cyclooxygenase-2 inhibitors: where we stand now. Ann Intern Med. 2005 Feb 1;142(3):212-4.
4) Ottenbacher KJ, Ottenbacher HR, Tooth L, Ostir GV. A review of two journals found that articles using multivariable logistic regression frequently did not report commonly recommended assumptions. J Clin Epidemiol. 2004 Nov;57(11):1147-52.
Salvatore Corrao, MD, Director
Clinical Methodology, Epidemiology and Statistics Department
"Civico e Benfratelli" National Relevance Hospital Trust Piazza Nicola Leotta, 2
90127 Palermo "“ Italy
Kimmel SE, Berlin JA, Reilly M, Jaskowiak J, Kishel L, Chittams J, et al. Patients Exposed to Rofecoxib and Celecoxib Have Different Odds of Nonfatal Myocardial Infarction. Ann Intern Med. ;142:157–164. doi: 10.7326/0003-4819-142-3-200502010-00005
Download citation file:
Published: Ann Intern Med. 2005;142(3):157-164.
Acute Coronary Syndromes, Cardiology, Coronary Heart Disease, Emergency Medicine.
Results provided by:
Copyright © 2019 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use