doi: 10.1161/CIRCULATIONAHA.105.568451
(Circulation. 2005;112:759-770.)
© 2005 American Heart Association, Inc.
Special Report |
Cyclooxygenase Inhibition and Cardiovascular Risk
From the Cardiovascular Division, Department of Medicine, Brigham and Women’s Hospital, Boston, Mass (E.M.A., J.L.); and Department of Biostatistics and Medical Informatics, University of Wisconsin, Madison (D.D.).
Correspondence to Dr Elliott M. Antman, Cardiovascular Division, Brigham & Women’s Hospital, 75 Francis St, Boston, MA 02115. E-mail eantman{at}rics.bwh.harvard.edu
Key Words: inflammation • pharmacology • platelets • prostaglandins • thromboxane
 |
Introduction |
|---|
 Top Introduction Biology of EicosanoidsReferences |
|---|
Over the past several months clinicians have been confronted at an escalating rate with reports describing the risks of COX-2 inhibitors (coxibs). Information on this class of drugs appears not only in traditional medical journals and textbooks, but also on the Web sites of regulatory agencies, professional societies, and pharmaceutical manufacturers.1–9 An unusual, but noteworthy, aspect of the state of affairs is the high rate of reporting new findings and opinions (at times voiced in a strident tone) in the lay press.10–13 Understandably, patients are concerned about the potential risks associated with their antiinflammatory medications, and now, either spontaneously or in response to a public health warning, call their healthcare professional with questions or arrive for an office visit armed with publicly accessed information they wish to discuss.
Owing to the widespread use of antiinflammatory drugs and the fact that the reported risks are cardiovascular in nature, we offer the readers of Circulation this special article. Our goals are to provide an overview of the relevant biology and pharmacology, to synthesize the data on the cardiovascular risks associated with antiinflammatory medications, to offer suggestions on strategies for prescribing these medications, and to make observations on the regulatory and research implications of the data and their interpretation.
|
Biology of Eicosanoids |
|---|
|
TopIntroductionBiology of EicosanoidsReferences |
|---|
Eicosanoids are oxidized derivatives of the polyunsaturated long-chain fatty acids, arachidonic acid and eicosapentaenoic acid, that serve many roles in cardiovascular biology and disease. Eicosanoid biosynthesis can be initiated by release of arachidonic acid from membrane phospholipids by lipases (predominantly of the phospholipase A2 type) (Figure 1). Once mobilized, arachidonic acid is oxygenated into eicosanoids along the following 4 pathways: (1) prostaglandin (PG) endoperoxide synthase (cyclooxygenase [COX]), (2) lipoxygenase (LO), (3) P450 epoxygenase, and (4) (nonenzymatic) isoprostane synthesis. Details of the pharmacology of products of the LO, epoxygenase, and isoprostane pathways, as well as the lipoxins, are reviewed elsewhere.14–16 Here, we focus on products of the COX pathway. These derivatives of arachidonic acid are collectively referred to as prostanoids and comprise the PGs and thromboxanes. COX, a key enzyme in eicosanoid metabolism, converts arachidonic acid liberated from membrane phospholipids into PGG2 and PGH2.
|
The fate of PGH2 and the distribution of prostanoids formed from it depend on the cell type in which it is synthesized and the tissue-specific isomerases found within that cell. For example, leukocytes, vascular smooth muscle cells, endothelial cells, and platelets express PGE synthase and, as a result, are all capable of generating the inflammatory prostanoid PGE2; platelets express thromboxane synthase and elaborate the prothrombotic, vasoconstrictor prostanoid thromboxane A2; and endothelial cells express PGI synthase and synthesize the antithrombotic, vasodilator prostanoid PGI2 or prostacyclin. In addition to cell-specific synthesis, the biological effects of prostanoids are governed by cell-specific receptor-dependent signaling pathways (the receptors DP, EP, FP, IP, and TP for PGD2, PGE2, PGF2alpha, PGI2, and TxA2, respectively) that define biological responses. Importantly, prostanoid intermediates can also undergo transcellular metabolism (ie, PGH2 can be produced in a platelet and then taken up by a leukocyte in which it is converted to PGE2).17
Pharmacology and Molecular Biology of Inhibition of Prostanoid Synthesis
The pharmacological inhibition of prostanoid synthesis has been the focus of drug development for >100 years. From aspirin and other nonsteroidal antiinflammatory drugs (NSAIDs) that inhibit COX, the field evolved rapidly in the 1980s to yield a wide range of agents with varying antiinflammatory and analgesic potencies and varying pharmacodynamics. These NSAIDs comprise a class of agents with heterogeneous structures (Figure 2). As a therapeutic class, they provide antipyretic, analgesic, and inflammatory activities, but the relative degree of these effects varies markedly among the compounds (eg, acetaminophen has antipyretic and analgesic effects but little antiinflammatory activity). In addition, as a therapeutic class, the NSAIDs share the common side effects of gastrointestinal ulceration, inhibition of uterine motility, inhibition of PG-mediated renal function, and hypersensitivity reactions. The relative frequencies of these side effects, however, vary markedly among the members of the class.
|
With the recognition that aspirin inhibits platelet function, the antithrombotic effect of these agents gained unique therapeutic emphasis and, of course, has proved to be a most important treatment strategy in atherothrombotic disease. Because the endothelial prostanoid prostacyclin has antithrombotic action while the platelet prostanoid thromboxane A2 has prothrombotic action, investigators realized in the 1970s that nonselective inhibition of COX pools in these 2 cell types could theoretically result in an attenuation of the antithrombotic effect of inhibition of platelet COX. At that time, investigators defined doses and conditions under which the suppression of thromboxane A2 synthesis by aspirin predominates over the suppression of prostacyclin synthesis; as a result of the irreversible inhibition (via acetylation) of COX by aspirin and the difference in half-lives of inhibited platelet and endothelial COX, low-dose aspirin was found to provide sufficient antithrombotic selectivity for primary and secondary prevention of atherothrombotic events.
In the early 1990s, a new wrinkle in the complex molecular events governing vascular prostanoid synthesis arose with the identification of 2 members of the COX family. COX-1 is constitutively expressed in many cells and is the only isozyme expressed in platelets; COX-2 is induced by inflammatory cytokines in many cell types (Table 1 and Figure 3).18,19 This distinction developed in the face of concern about the hemorrhagic risk of nonselective COX inhibition, especially gastrointestinal bleeding. The recognition that there are 2 different COXs led to the straightforward view that COX-2 is responsible for the adverse proinflammatory effects of prostanoids and that nonspecific COX inhibitors cause bleeding by inhibiting COX-1 in platelets. This working paradigm led the pharmaceutical industry to develop COX-2–selective inhibitors, the coxibs, arguing that these agents would provide adequate analgesia and antiinflammatory effects without hemorrhagic risk. The incidence of gastropathy accompanying the use of nonselective NSAIDs was felt to be of sufficient magnitude to warrant the development of coxibs; according to one meta-analysis,20 approximately one-third of patients taking NSAIDs had gastric or duodenal ulcers by endoscopy, but the risk of serious gastrointestinal bleeding is much lower,21 with 107 000 patients hospitalized each year for gastrointestinal complications of NSAID use.22 Despite their development to provide a therapeutic option for patients at risk of gastrointestinal bleeding, survey reports suggest that coxibs are prescribed with approximately equal frequency to patients who are at low or high risk of gastrointestinal bleeding.23
|
|
Unfortunately, this clean mechanistic distinction between the COXs is an oversimplification. As a class, the NSAIDs have a broad range of relative COX-1 and COX-2 selectivity (Figure 2). Furthermore, COX-2 is expressed in normal endothelial cells in response to shear stress,24 and inhibition of COX-2 is associated with significant suppression of prostacyclin synthesis in human subjects25,26 (Figure 4). Both COX-1 and COX-2 are detectable in human atherosclerotic lesions27,28; however, the specific effect of COX inhibition on lesion progression is currently controversial. Low-dose aspirin and selective COX-2 inhibitors have been shown to improve29,30 or worsen31 endothelial dysfunction in hypercholesterolemia and hypertension. COX-2 may also play a role in destabilizing plaques as suggested by its increased expression and colocalization with microsomal PGE synthase-1 and metalloproteinases-2 and -9 in carotid plaques from individuals with symptomatic disease before endarterectomy.32 Studies in hypercholesterolemic mice have also been inconsistent, with reports showing that COX-2 inhibitors worsen, retard, or fail to affect the course of atherosclerosis,33–35 and evidence exists to support the view that COX-2–derived prostacyclin is atheroprotective, but only in female mice.36 Possible explanations for these divergent outcomes include differences among the methods of suppressing COX-2 activity, the timing of administration of the inhibitor, and the genetic backgrounds of the animals studied. In addition, COX-2–derived prostanoid products have divergent effects on the molecular and cellular mechanisms underlying disease pathogenesis. For example, through its interaction with the E-prostanoid receptor EP4 on macrophages, COX-2–derived PGE2 suppresses chemokine production,37 whereas TP antagonism retards atherogenesis.38 Moreover, COX-2 is differentially expressed among cell types in the plaque, as are the predominant prostanoids derived from those cells.
|
Inhibition of COX-2 also has as a theoretical side effect an increase in the flux of arachidonate through the LO pathways, which may be especially important in the setting of inflammation in the atheromatous plaque. The 12-,15-, and 5-LOs all have key roles in inflammation, and the role of each in atherosclerosis has been examined. Although 12-LO and 15-LO appear to contribute to LDL oxidation, the data supporting the proatherogenic role of these enzymes are inconsistent.39 Data suggest that 15-LO products may be antiinflammatory.40 Furthermore, work from Serhan’s group shows that acetylation of COX-2 by low-dose aspirin leads to its biosynthesis of 15R-hydroxyeicosatetraenoic acid.40 This intermediate is then converted by transcellular metabolism to the antiinflammatory lipoxin 15-epi-lipoxin A4 in leukocytes.41
Mehrabian and colleagues42 have demonstrated convincingly that 5-LO is a critical determinant of atherogenesis in mouse models of the disease, even in the setting of profound hypercholesterolemia. The inflammatory eicosanoids derived from increased 5-LO expression in plaque–leukotriene B4 and the cysteinyl-leukotrienes–are active in the atherothrombotic vasculature, having been shown to promote inflammatory cell activation, cell proliferation, and vasoconstriction. In human subjects, Dwyer and colleagues43 showed that variant 5-LO genotypes–tandem promoter repeats of Sp-1 binding motifs–identify a subpopulation of individuals with increased atherosclerosis (determined as carotid intima-media thickness). Helgadottir and colleagues44 showed that a promoter haplotype comprising 4 linked polymorphisms in the 5-LO activating peptide (an accessory protein that facilitates presentation of substrate arachidonate to 5-LO) confers an approximately 2-fold increased risk of myocardial infarction (MI) and stroke in an Icelandic population. Thus, the potential importance of shifting the flux of arachidonate through the LO pathway by inhibiting COX activity bears consideration as we attempt to dissect the vascular consequences of coxib use.
To appreciate the complexity of interactions among the small-molecule vascular mediators in the system, we also need to consider nitric oxide (NO) and superoxide anion. NO activates prostacyclin synthase and suppresses thromboxane synthase, likely by nitrosylating bound heme.45 In addition, NO potentiates the vascular effects of prostacyclin, likely via the cGMP-dependent inhibition of cAMP phosphodiesterase.46 This potentiation of prostacyclin by NO has also been demonstrated to account for the synergistic inhibition of platelets by these vascular effectors.47,48 Niwano and colleagues49 have shown that a stable prostacyclin analogue (beraprost) increases endothelial NO synthase (eNOS) expression by activating a cAMP-dependent transcriptional element in the eNOS promoter. In the setting of an inflammatory stimulus that induces expression of inducible NO synthase (iNOS) and a source of superoxide [such as NAD(P)H oxidase], peroxynitrite generation ensues and leads to 3-nitration of tyrosine 430 in prostacyclin synthase, inactivating the enzyme,50,51 and activation of the TxA2 receptor TP.50 TxA2, in turn, induces gp91phox expression and NAD(P)H oxidase–dependent superoxide generation,52 increasing oxidant stress in the inflamed vasculature. NO derived from iNOS also increases expression and activity of COX-2.53,28 In addition, other inflammatory mediators may modulate these interactions; eg, evidence suggests that C-reactive protein decreases prostacyclin release from endothelial cells.54
Consideration of these interactions is essential for understanding the full spectrum of activities of COX-2–dependent eicosanoid synthesis in the context of their interaction with NO. For example, COX-2 not only has been recognized as a key source of prostacyclin under normal laminar flow conditions in the vasculature but also is cardioprotective in ischemia-reperfusion injury55 and has antiproliferative effects toward vascular smooth muscle cells in conjunction with NO.56 NO can also inhibit 5-LO, likely by peroxynitrite-dependent S-nitrosation and/or 3-nitrotyrosination.57,58 Induction of iNOS by endotoxin leads to inhibition of 5-LO activity without an effect on expression,59 likely via a peroxynitrite-dependent mechanism.
The interrelationships among COX-2, 5-LO, and NO in the endothelium can best be analyzed when considered under 2 sets of conditions: in the normal state of laminar flow and in an inflammatory state (Figure 5). Under normal conditions, laminar flow induces COX-2 in the endothelial cell to promote the synthesis of prostacyclin, and stimulates elaboration of NO by eNOS. NO derived from eNOS, in turn, stimulates prostacyclin synthase activity and suppresses thromboxane synthase activity; NO also activates guanylyl cyclase to increase cGMP and acts synergistically with prostacyclin to increase cAMP levels in target cells (eg, platelets). Taken together, the net effect of these actions is to impair platelet activation, as summarized in Figure 4 (left) and Figure 5A.
|
In states of vascular inflammation, COX-2, iNOS, and NAD(P)H oxidase are induced in endothelial cells; these enzymes, together with 5-LO, are also expressed in inflammatory leukocytes. High-flux production of NO (from iNOS) together with superoxide anion [from NAD(P)H oxidase, COXs, LOs, and uncoupled NO synthases, among other sources] leads to the synthesis of peroxynitrite (OONO–), which inhibits prostacyclin synthase, activates TP-dependent signaling, and promotes additional COX-2 activity. The COX pathways also promote NAD(P)H oxidase activation via TxA2,33 whereas 5-LO promotes NAD(P)H oxidase activation via leukotriene B460 and the cysteinyl-leukotrienes. Moreover, PGE2, the synthesis of which is enhanced by COX2-derived PGH2 owing to kinetic selectivity and compartmentalization,61 promotes platelet activation by increasing intraplatelet calcium flux and decreasing cAMP via its interaction with the platelet surface EP3 receptor.62 (For a review of the effects of NO-derived reactive nitrogen species in inflammatory states on COXs, LOs, and peroxidases, see Coffey and colleagues.63) Taken together, the net effect of these actions is to promote platelet activation, as summarized in Figure 4 (right) and Figure 5B.
We can use the models shown in Figures 4 and 5
to construct working hypotheses about the use of coxibs in the normal state and in states of vascular inflammation. Central to this model is the balance between prostacyclin (PGI2) and thromboxane A2 in normal and diseased vessels.27,64,65 The use of a coxib under normal (ie, noninflammatory) conditions would be expected to have limited effects on platelet activation in that NO production by eNOS is relatively unimpaired, and COX-1–dependent generation of prostacyclin would still be maintained. In contrast, the use of a coxib in vascular inflammatory states would lead to a decrease in antithrombotic prostacyclin made by arachidonate flux through COX-2 and would, therefore, make available more arachidonate for leukotriene synthesis. Leukotrienes, especially leukotriene B4 and the cysteinyl-leukotrienes, would increase reactive oxygen species generation by leukocytes, especially superoxide, thereby consuming antithrombotic NO through the synthesis of peroxynitrite. Peroxynitrite, in turn, would further limit prostacyclin synthesis via synthase nitration.47 Coxibs may also increase reactive oxygen species generation via uncoupling of mitochondrial oxidative phosphorylation.66 Thus, the net result of coxib action in diseased vessels is an increase in the amount of TxA2 relative to PGI2 (see Figure 4).
In addition to the considerations at the molecular level discussed above, it should be noted that manipulation of the relative balance of COX-1 and COX-2 activity may alter important cardiorenal responses in patients.19 COX-1 and COX-2 are colocalized in the macula densa. In elderly patients or under conditions of sodium or fluid depletion, selective COX-2 inhibitors cause sodium retention and may result in edema formation.67 Administration of COX-2 inhibitors has also been associated with a reduction in glomerular filtration rate and exacerbation of hypertension.68–70 Increases in blood pressure have been proposed as a mechanism by which COX-2 inhibitors may promote an increased risk of cardiovascular events.71
Synthesis of Data on Risk of Cardiovascular Events
The emerging data on the potential for an increased risk of cardiovascular events with coxib use present a disturbing and confusing message for clinicians and patients. The concept of assessing the risk-to-benefit ratio of any medication is familiar to clinicians but is difficult to operationalize for the coxibs. Although the benefits of coxibs (eg, a convenient form of analgesia for arthritic and musculoskeletal discomfort) are more apparent, the risks associated with their use are less clear. Critical questions remain unanswered: What is the magnitude of the increased risk of cardiovascular events with coxib use? Does the risk vary among the coxibs, or is it a true class effect? Is a similar increased risk of cardiovascular events seen with the nonselective NSAIDs?
The evidence base with regard to the risk of cardiovascular events consists of randomized controlled trials (RCTs), new drug application databases submitted to regulatory authorities, and analyses from community- and hospital-based prescription practices. Many of the individual RCT and prescription practice reports can be found in the medical literature, although some are unpublished reports that have been made available on Web sites.72 Meta-analyses of the available RCT data have also been published.73 Rather than reiterate the details of previous reports, we offer potential explanations for the nature of the data and present a model that synthesizes the information in a clinically meaningful fashion.
We begin the discussion of the RCT data by a worked example summarizing the cardiovascular risk associated with celecoxib in a clinical trial (APC) for prevention of colorectal adenoma (Figure 6).74 A total of 2035 patients with a history of colorectal neoplasia were randomized to receive either placebo or 1 of 2 doses of celecoxib (200 or 400 mg twice daily). After 2.8 to 3.1 years of follow-up, an independent Data Safety Monitoring Board concluded that continued exposure to celecoxib placed patients at increased risk for serious cardiovascular events. The APC study was designed for a noncardiovascular purpose–prevention of colorectal adenoma–yet an increased risk of cardiovascular events was identified. The situation surrounding the APC study and cardiovascular risk is representative of many of the coxib trial data to date.
|
What kind of observations might have concerned the Data Safety Monitoring Board? Of the 679 patients randomized to placebo, 4 patients (0.6%) either died or sustained a nonfatal MI. In contrast, of the 1356 patients assigned to either of the celecoxib arms, 27 (2.0%) died or had a nonfatal MI. After the data are entered into a 2x2 table, several statistical tests of the treatment effect (harm in this case) of celecoxib can be performed. Regardless of whether one uses a 
2 test (or the equivalent binomial comparison of proportions) or Fisher’s exact test, the difference in event rates in the combined celecoxib group is statistically significantly higher than that in the placebo group (Figure 6).
Another important message is also illustrated by the APC trial findings. The event rates are relatively low compared with those seen in trials enrolling acute coronary syndrome patients in which the death/MI rate may be 10% at 1 year, depending on the level of risk at the time of enrollment. Statements about the relative risk (RR) and odds ratio (OR) convey information about the proportionate increase in risk with celecoxib use versus placebo (3.4-fold increased risk in this example) (Figure 6). (For low event rates, as in this example, the OR approximates the RR.) Such ratios are often the only statistic quoted in the lay press but do not present the full range of clinically relevant information. For example, the absolute risk difference (ARD) in event rates is 1.4%. To interpret the information from the APC trial in the context of data from other trials, it is useful to calculate the number of patients who need to be treated with celecoxib to observe 1 death or MI event, ie, the number needed to harm (NNH). In this case, NNH=1/ARD=71 patients (Figure 6).
In contrast to the APC trial in which there was a straightforward comparison of a coxib (celecoxib) with placebo, the Vioxx Gastrointestinal Outcome Study (VIGOR) compared the occurrence of gastrointestinal toxicity with a different coxib (rofecoxib 50 mg/d) and another NSAID (naproxen 1000 mg/d) in patients with rheumatoid arthritis.75 Aspirin use was not permitted in VIGOR. A signal of increased risk (RR, 2.38; 95% CI, 1.39 to 4.00; P<0.001) of serious thrombotic cardiovascular events (MI, unstable angina, cardiac thrombus, resuscitated cardiac arrest, sudden or unexplained death, ischemic stroke, and transient ischemic attacks) was found in a report submitted to the US Food and Drug Administration (FDA) subsequent to the publication of the original VIGOR manuscript. This increased risk of cardiovascular events with rofecoxib was observed whether or not the patient would (RR, 4.89; 95% CI, 1.41 to 16.88; P=0.01) or would not (RR, 1.89; 95% CI, 1.03 to 3.45; P=0.04) have been eligible to receive aspirin.76
The active comparator naproxen in VIGOR was a confounding factor that made interpretation of the RR associated with rofecoxib more difficult to assess. It was initially argued that naproxen had a protective antithrombotic effect that reduced the event rate (compared with a putative placebo) and led to an apparent increased RR with rofecoxib. However, more compelling data consistent with a true increase in the RR with rofecoxib emerged from reports such as a cumulative meta-analysis of RCTs comparing rofecoxib with placebo,73 a case-control study of 54 475 Medicare beneficiaries,77 a case-control study of 8518 patients from a database of 36 hospitals in 5 counties,78 a retrospective case-control study of 113 927 patients in Quebec’s administrative health databases,79 and a nested case-control study from the Kaiser Permanente database in California.80
Eventually, when the Adenomatous Polyp Prevention on Vioxx trial (APPROVe), comparing rofecoxib 25 mg/d with placebo in patients with a history of colorectal adenomas, reported an excess risk of thrombotic cardiovascular events in the rofecoxib group (RR, 1.92; 95% CI, 1.19 to 3.11; P=0.008), Merck, the manufacturer of rofecoxib, voluntarily withdrew the drug from the market in September, 2004.1 Between September, 2004, and December, 2004, reports of increased risk of thrombotic cardiovascular events had accumulated not only for rofecoxib and celecoxib but also for the third coxib available in the United States, valdecoxib, to the point that on December 23, 2004, the FDA issued a public health advisory concerning the use of all coxibs.81,82 Of additional concern was a report from the NIH on December 20, 2004 about the Alzheimer’s Disease Anti-Inflammatory Prevention Trial (ADAPT), which was designed to assess the potential benefit of long-term use of naproxen (220 mg twice daily) or celecoxib (200 mg twice daily) versus placebo in decreasing the risk of developing Alzheimer’s disease in subjects 
70 years of age.83 There was an apparent increase (quantitative details not specified) in cardiovascular and cerebrovascular events in the naproxen group compared with placebo in this trial. Thus, while continuing to review the data, the FDA issued an advisory in December, 2004, as an interim measure, sensitizing clinicians and patients to the emerging data on increased risk of thrombotic events not only with coxibs but NSAIDs in general.81,82
In February, 2005, a special 3-day advisory committee meeting was convened by the FDA to provide a forum for full discussion of the issues. The advisory panel recommended that celecoxib and valdecoxib remain on the market but advocated that "black box" warnings be added to the label. At the same time, the European Medicines Agency imposed strong warnings on coxibs, recommending that they not be prescribed to patients who have coronary heart disease or who have had a stroke and that they should be used with caution in patients at risk for vascular disease.84,85
Additional information bearing on the issue of cardiovascular risk comes from the Therapeutic Arthritis Research and Gastrointestinal Event Trial (TARGET), which compared the most selective coxib, lumiracoxib (see Figure 2), 400 mg once daily with either naproxen 500 mg twice daily or ibuprofen 800 mg 3 times daily for 1 year in 18 000 patients with osteoarthritis.86 Low-dose aspirin (75 to 100 mg daily) was permitted in TARGET. There were only 109 cardiovascular or cerebrovascular events reported, of which 59 (0.65%) occurred in the lumiracoxib group and 50 (0.55%) occurred in the NSAID groups (hazard ratio, 1.14; 95% CI, 0.78 to 1.66; P=0.51). Although these findings might be interpreted as showing that lumiracoxib is as safe as either naproxen or ibuprofen, in the absence of a placebo group, the results are also consistent with the possibility that all 3 drugs are associated with increased risk of events with little difference among them.
The next major event in this rapidly evolving story occurred on April 7, 2005.87 The FDA concluded that the overall risk-to-benefit profile for valdecoxib was unfavorable and that valdecoxib lacked any demonstrable advantage compared with other NSAIDs. The agency requested that Pfizer voluntarily withdraw valdecoxib from the market, which Pfizer agreed to do. While permitting celecoxib to remain on the market, the FDA requested revision to the labeling of celecoxib and 18 other nonselective NSAIDs to highlight the increased risk for cardiovascular events and stated that all NSAID prescriptions must be accompanied by a medication guide to inform patients. In support of this decision by the FDA is a report from a registry experience in Denmark of 10 280 cases of first-time hospitalization for MI and 102 797 controls.88 Current and new users of all classes of non-aspirin NSAIDs had elevated RR estimates for MI.
How might the seemingly contradictory evidence about the increased risk of cardiovascular events with COX-2 inhibitor use (and NSAID use in general) be explained? Figure 7 summarizes the interplay of a variety of factors that influence the ability of investigators to detect a signal of increased risk of cardiovascular events associated with COX-2 inhibition. Certain variables are set by investigators during the design of a clinical trial. These include (1) the definition of events that constitute the trial end points (eg, a hard end point such as death is infrequent, resulting in fewer events observed compared with a composite end point), (2) the duration of follow-up (short-term follow-up limits the time during which events may occur and reduces the likelihood of detecting harm), and (3) sample size (an inadequate sample size places investigators at the risk of a large type II error and failure to conclude correctly that evidence of harm from COX-2 inhibition exists).
|
Variables related to both the patient and the drug being investigated influence the relative difference in events in the treatment groups and may minimize or magnify the signal of increased risk of events. Variables of note include (1) the risk of events in the control group (the impact of COX-2 inhibition may be less evident in healthier subjects in whom relatively few events occur in the control group), (2) the RR of events in the treatment group (related to both the intrinsic properties of the drug being investigated and the choice of the comparator arm—eg, a signal of harm with a coxib is more easily detected if the comparator arm is placebo and less readily detected in trials with an active comparator unless that active comparator actually is cardioprotective, thus magnifying the RR with a coxib), 3) and drug interactions (concomitant medications may exacerbate the risk with COX-2 inhibitors by such mechanisms as worsening of renal function or diminishing the cardioprotective effects of drugs such as aspirin [low dose]).89 To complicate the situation further, the patient and drug variables may change over the course of exposure to the drug. For example, development of diabetes or worsening hypertension may culminate in disruption of a high-risk or vulnerable plaque with the development of a superimposed thrombus. As the acute situation evolves, the risk in the control arm may change, and the RR associated with the drug may also change, both in an adverse direction.
Building on these concepts, one may depict the relationship among the risk of events in the control group (control event rate, CER), the RR of events with a particular drug, and the NNH (critically related to the ability to detect a signal of harm), which are related to each other by the formula: NNH=1/[(RR–1)xCER]. The surface shown in Figure 8 rises steeply to a high NNH (difficulty in detecting harm) with a low rate of events in the control group and/or low RR in the treatment group. The ability to detect harm improves as the NNH drops with increasing rates in the control arm and/or increasing RR in the treatment arm (Figures 7 and 8 
).
|
Strategies for Use of NSAIDs
Although the evidence base is incomplete, one can use the concepts articulated in this report to develop strategies for the use of NSAIDs.7 The general goal is to operate on the steep portion of the surface in Figure 8, thereby minimizing patient risk. This can be accomplished by preferentially prescribing NSAIDs only to patients at low risk of thrombotic events (ie, moving to lower rates of events in the control group in Figure 8). Selecting drugs with a lower risk of thrombotic events and minimizing the dose and duration of treatment are also advisable (ie, moving to a lower RR in Figure 8). If clinical circumstances call for using an NSAID in a higher-risk patient and/or for extended periods, concomitant prescription of low-dose aspirin (81 mg/d) may help to mitigate the tendency to thrombotic events but may not eliminate the risk entirely90 (Figure 4). A suggested strategic plan for the use of NSAIDs is shown in Table 2.
|
Regulatory and Research Implications
The coxib story has brought into sharp focus issues related to the drug approval process and surveillance for adverse events for approved drugs. It has been argued that the passage of the Prescription Drug User Fee Act (PDUFA) in response to pressure from the pharmaceutical industry, medical community, and patients to hasten the speed of approval of new drugs set the stage for the tsunami of events between September, 2004, and April, 2005.91 PDUFA provided a mechanism for sponsors of new drugs to pay a fee to the FDA to cover the costs of accelerating the review and approval process. Although concern has been raised by some authors, it is not clear whether acceleration of the review and approval process under PDUFA resulted in less attention to the safety database for the coxibs.92 Another element in the coxib story was the introduction of direct-to-consumer advertising.93 In the case of the coxibs, aggressive direct-to-consumer advertising rapidly generated a large number of requests by patients for prescriptions.
What lessons can be learned from the early signals of harm that were detected with coxib use so that the drug approval process can be improved? We propose that for all new drugs under review, a minimum number of patient-years of safety experience be obtained before approval for use. These data may be acquired in a few very large trials or in a series of smaller trials provided that the cumulative patient-years of safety data are sufficient, which appears not to have been the case for the coxibs before their approvals. Exactly what the minimum number of patient-years of safety observation should be is not clear, but this is an important topic for future biostatistical and regulatory research efforts. A reasonable starting point is to examine the compiled number of patient-years of safety experience before approval of drugs that were later withdrawn from clinical use because of toxicity. Depending on the rate of adverse events, it can be argued that such an analysis would help frame the minimum requirement for patient-years of safety experience to acquire a comfort level (from a regulatory perspective) with a new drug.94
Once drugs are approved, continued surveillance for safety issues is needed. An initial step at augmenting postmarketing surveillance was the announcement by Health and Human Services Secretary Michael Leavitt that a new Drug Safety Oversight Board would be established within the FDA.95 This organizational and oversight strategy is a welcome start, but it is likely that more will be required, including efforts independent of the FDA.91,96,97 One possible modification is to require a mandatory registry of the first several thousand patients exposed to a newly approved drug and followed up for a period of several years. While this mandatory registry would not have an ideal control, it at least would provide a more accurate estimate of adverse events. Also, given the potential for ascertainment bias (ie, overreporting or underreporting of adverse events) in the present postmarketing surveillance system, efforts to standardize protocols for adverse event reporting are needed.
Finally, the panic of some patients after the withdrawal of rofecoxib and valdecoxib from the marketplace when these were the drugs that relieved their symptoms most effectively underscores the need for continued pharmacological research. Investigators need to explore whether other compounds may be identified that truly treat inflammation without an adverse impact on the course of atherothrombosis. Nevertheless, it should be noted that it is unrealistic to expect any pharmaceutical entity to be completely devoid of side effects and that the decision-making process during evaluation of new agents remains complicated even for experienced investigators and regulatory authorities.98
Note Added in Proof
Additional support for the concept of an increased risk of MI with both conventional NSAIDs and coxibs is found in a report from the United Kingdom that analyzed 9218 cases and 86 349 controls in a clinical database containing records from 468 practices. 99 Further discussion of the marketing influences surrounding the coxibs and regulatory considerations can be found in a perspective written by Henry Waxman, a US Representative from California.100
|
Acknowledgments |
|---|
We thank Drs Joseph Vita, Jane Leopold, and Charles Serhan for helpful comments.
|
References |
|---|
|
TopIntroductionBiology of EicosanoidsReferences |
|---|
1. Bresalier RS, Sandler RS, Quan H, et al. Cardiovascular events associated with rofecoxib in a colorectal adenoma chemoprevention trial. N Engl J Med. 2005; 352: 1092–102.
2. Pratico D, Dogne J. Selective COX-2 inhibitors development in cardiovascular medicine. Circulation. In press.
3. Brophy J. Hot Topic: Cyclooxygenase-2 inhibition and coronary risk. Available at: McGraw-Hill’s AccessMedicine, Harrison’s Online Updates, 2005. Accessed March 25, 2005.
4. Statement of Sandra Kweder, MD, Deputy Director, Office of New Drugs, Center for Drug Evaluation and Research, US Food and Drug Administration. Washington, DC: Committee on Finance, United States Senate; November 18, 2004. Available at: http://www.fda/gov/ola/2004/vioxx1118.html. Accessed December 31, 2004.
5. US Food and Drug Administration, Center for Drug Evaluation and Research, Public Health Advisory. Non-steroidal anti-inflammatory drug products (NSAIDS). Available at: http://www.fda.gov/cder/drug/advisory/nsaids.htm. Accessed January 19, 2005.
6. American College of Rheumatology. American College of Rheumatology offers guidance for assessing arthritis pain medication usage. Press release, 2005. Available at: http://www.rheumatology.org/press/2004/cox_2_news.asp. Accessed January 23, 2005.
7. Bennett JS, Daugherty A, Herrington D, et al. The use of nonsteroidal anti-inflammatory drugs (NSAIDs): a science advisory from the American Heart Association. Circulation. 2005; 111: 1713–1716.
8. Crawford L. Letter to FDA revealing heart dangers in an unpublished clinical trial of Celebrex (HRG Publication #1721). Public Citizen, citizen.org, 2005. Available at: http://www.citizen.org/publications/release,cfm?ID=7359. Accessed on February 1, 2005.
9. Summary of Prepared Testimony, Raymond V. Gilmartin, President, Chairman and Chief Executive Officer, Merck and Co, Inc, before the US Senate Committee on Finance, November 18, 2004. Merck News Item, 2005. Available at: http://www.merck.com/newsroom/vioxx_withdrawal/11182004.html. Accessed April 12, 2005.
10. Rubin R. Painkillers hang in the balance. USATODAY.com, 2005. http://www.usatoday.com/news/health/2005–02–10-painkillers_x.htm. Accessed February 11, 2005.
11. Schmid R. FDA scientist warns of COX-2 drug dangers. USA TODAY. February 18, 2005: 4A.
12. Hla T. A balanced look at COX-2 drugs. Boston Globe. March 25, 2005.
13. Henderson D, Rowland C. Once "too slow," FDA approvals called "too fast." Boston Globe. Available at: http://www.boston.com/business/articles/2005/04/10/fda_criticized_as_too_quick_to_ok_drugs/. Accessed on April 10, 2005.
14. Foegh M, Ramwell P. The eicosanoids: prostaglandins, thromboxanes, leukotrienes, and related compounds. In: Katzung BG, ed. Basic and Clinical Pharmacology. 9th ed. New York, NY: Lange Medical Books/McGraw-Hill; 2004.
15. Wagner W, Khanna P, Furst DE. Nonsteroidal anti-inflammatory drugs, disease-modifying antirheumatic drugs, nonopioid analgesics, and drugs used in gout. In: Katzung BG, ed. Basic and Clinical Pharmacology. 9th ed. New York, NY: Lange Medical Books/McGraw-Hill; 2004.
16. Samuelsson B, Dahlen SE, Lindgren JA, et al. Leukotrienes and lipoxins: structures, biosynthesis, and biological effects. Science. 1987; 237: 1171–1186.
17. Marcus AJ. Transcellular metabolism of eicosanoids. Prog Hemost Thromb. 1986; 8: 127–142.[Medline] [Order article via Infotrieve]
18. Hawkey CJ. COX-2 inhibitors. Lancet. 1999; 353: 307–314.[CrossRef][Medline] [Order article via Infotrieve]
19. Warner TD, Mitchell JA. Cyclooxygenases: new forms, new inhibitors, and lessons from the clinic. FASEB J. 2004; 18: 790–804.
20. Huang JQ, Sridhar S, Hunt RH. Role of Helicobacter pylori infection and non-steroidal anti-inflammatory drugs in peptic-ulcer disease: a meta-analysis. Lancet. 2002; 359: 14–22.[CrossRef][Medline] [Order article via Infotrieve]
21. Ofman JJ, MacLean CH, Straus WL, et al. A metaanalysis of severe upper gastrointestinal complications of nonsteroidal antiinflammatory drugs. J Rheumatol. 2002; 29: 804–812.
22. Singh G. Recent considerations in nonsteroidal anti-inflammatory drug gastropathy. Am J Med. 1998; 105: 31S–38S.[CrossRef][Medline] [Order article via Infotrieve]
23. Kaufman DW, Kelly JP, Rosenberg L, et al. Are cyclooxygenase-2 inhibitors being taken only by those who need them? Arch Intern Med. 2005; 165: 1066–1067.
24. Gimbrone MA, Jr., Topper JN, Nagel T, et al. Endothelial dysfunction, hemodynamic forces, and atherogenesis. Ann N Y Acad Sci. 2000; 902: 230–239;discussion 239–240.
25. McAdam BF, Catella-Lawson F, Mardini IA, et al. Systemic biosynthesis of prostacyclin by cyclooxygenase (COX)-2: the human pharmacology of a selective inhibitor of COX-2. Proc Natl Acad Sci U S A. 1999; 96: 272–277.
26. Hansson GK. Inflammation, atherosclerosis, and coronary artery disease. N Engl J Med. 2005; 352: 1685–1695.
27. Belton O, Byrne D, Kearney D, et al. Cyclooxygenase-1 and -2-dependent prostacyclin formation in patients with atherosclerosis. Circulation. 2000; 102: 840–845.
28. Schonbeck U, Sukhova GK, Graber P, et al. Augmented expression of cyclooxygenase-2 in human atherosclerotic lesions. Am J Pathol. 1999; 155: 1281–1291.
29. Monobe H, Yamanari H, Nakamura K, et al. Effects of low-dose aspirin on endothelial function in hypertensive patients. Clin Cardiol. 2001; 24: 705–709.[Medline] [Order article via Infotrieve]
30. Widlansky ME, Price DT, Gokce N, et al. Short- and long-term COX-2 inhibition reverses endothelial dysfunction in patients with hypertension. Hypertension. 2003; 42: 310–315.
31. Bulut D, Liaghat S, Hanefeld C, et al. Selective cyclo-oxygenase-2 inhibition with parecoxib acutely impairs endothelium-dependent vasodilatation in patients with essential hypertension. J Hypertens. 2003; 21: 1663–1667.[CrossRef][Medline] [Order article via Infotrieve]
32. Cipollone F, Prontera C, Pini B, et al. Overexpression of functionally coupled cyclooxygenase-2 and prostaglandin E synthase in symptomatic atherosclerotic plaques as a basis of prostaglandin E(2)-dependent plaque instability. Circulation. 2001; 104: 921–927.
33. Burleigh ME, Babaev VR, Oates JA, et al. Cyclooxygenase-2 promotes early atherosclerotic lesion formation in LDL receptor-deficient mice. Circulation. 2002; 105: 1816–1823.
34. Olesen M, Kwong E, Meztli A, et al. No effect of cyclooxygenase inhibition on plaque size in atherosclerosis-prone mice. Scand Cardiovasc J. 2002; 36: 362–367.[CrossRef][Medline] [Order article via Infotrieve]
35. Rott D, Zhu J, Burnett MS, et al. Effects of MF-tricyclic, a selective cyclooxygenase-2 inhibitor, on atherosclerosis progression and susceptibility to cytomegalovirus replication in apolipoprotein-E knockout mice. J Am Coll Cardiol. 2003; 41: 1812–1819.
36. Egan KM, Lawson JA, Fries S, et al. COX-2-derived prostacyclin confers atheroprotection on female mice. Science. 2004; 306: 1954–1957.
37. Takayama K, Garcia-Cardena G, Sukhova GK, et al. Prostaglandin E2 suppresses chemokine production in human macrophages through the EP4 receptor. J Biol Chem. 2002; 277: 44147–44154.
38. Cayatte AJ, Du Y, Oliver-Krasinski J, et al. The thromboxane receptor antagonist S18886 but not aspirin inhibits atherogenesis in apo E-deficient mice: evidence that eicosanoids other than thromboxane contribute to atherosclerosis. Arterioscler Thromb Vasc Biol. 2000; 20: 1724–1728.
39. Funk CD, Cyrus T. 12/15-lipoxygenase, oxidative modification of LDL and atherogenesis. Trends Cardiovasc Med. 2001; 11: 116–124.[CrossRef][Medline] [Order article via Infotrieve]
40. Serhan CN, Jain A, Marleau S, et al. Reduced inflammation and tissue damage in transgenic rabbits overexpressing 15-lipoxygenase and endogenous anti-inflammatory lipid mediators. J Immunol. 2003; 171: 6856–6865.
41. Chiang N, Bermudez EA, Ridker PM, et al. Aspirin triggers antiinflammatory 15-epi-lipoxin A4 and inhibits thromboxane in a randomized human trial. Proc Natl Acad Sci U S A. 2004; 101: 15178–15183.
42. Mehrabian M, Allayee H, Wong J, et al. Identification of 5-lipoxygenase as a major gene contributing to atherosclerosis susceptibility in mice. Circ Res. 2002; 91: 120–126.
43. Dwyer JH, Allayee H, Dwyer KM, et al. Arachidonate 5-lipoxygenase promoter genotype, dietary arachidonic acid, and atherosclerosis. N Engl J Med. 2004; 350: 29–37.
44. Helgadottir A, Manolescu A, Thorleifsson G, et al. The gene encoding 5-lipoxygenase activating protein confers risk of myocardial infarction and stroke. Nat Genet. 2004; 36: 233–239.[CrossRef][Medline] [Order article via Infotrieve]
45. Wade ML, Fitzpatrick FA. Nitric oxide modulates the activity of the hemoproteins prostaglandin I2 synthase and thromboxane A2 synthase. Arch Biochem Biophys. 1997; 347: 174–180.[CrossRef][Medline] [Order article via Infotrieve]
46. Zellers TM, Wu YQ, McCormick J, et al. Prostacyclin-induced relaxations of small porcine pulmonary arteries are enhanced by the basal release of endothelium-derived nitric oxide through an effect on cyclic GMP-inhibited-cyclic AMP phosphodiesterase. Acta Pharmacol Sin. 2000; 21: 131–138.[Medline] [Order article via Infotrieve]
47. Koga T, Az-ma T, Yuge O. Prostaglandin E1 at clinically relevant concentrations inhibits aggregation of platelets under synergic interaction with endothelial cells. Acta Anaesthesiol Scand. 2002; 46: 987–993.[CrossRef][Medline] [Order article via Infotrieve]
48. Stamler JS, Vaughan DE, Loscalzo J. Synergistic disaggregation of platelets by tissue-type plasminogen activator, prostaglandin E1, and nitroglycerin. Circ Res. 1989; 65: 796–804.
49. Niwano K, Arai M, Tomaru K, et al. Transcriptional stimulation of the eNOS gene by the stable prostacyclin analogue beraprost is mediated through cAMP-responsive element in vascular endothelial cells: close link between PGI2 signal and NO pathways. Circ Res. 2003; 93: 523–530.
50. Bachschmid M, Thurau S, Zou MH, et al. Endothelial cell activation by endotoxin involves superoxide/NO-mediated nitration of prostacyclin synthase and thromboxane receptor stimulation. FASEB J. 2003; 17: 914–916.
51. Schmidt P, Youhnovski N, Daiber A, et al. Specific nitration at tyrosine 430 revealed by high resolution mass spectrometry as basis for redox regulation of bovine prostacyclin synthase. J Biol Chem. 2003; 278: 12813–12819.
52. Muzaffar S, Shukla N, Lobo C, et al. Iloprost inhibits superoxide formation and gp91phox expression induced by the thromboxane A2 analogue U46619, 8-isoprostane F2
, prostaglandin F2, cytokines and endotoxin in the pig pulmonary artery. Br J Pharmacol. 2004; 141: 488–496.[CrossRef][Medline]
[Order article via Infotrieve]
53. Perez-Sala D, Lamas S. Regulation of cyclooxygenase-2 expression by nitric oxide in cells. Antioxid Redox Signal. 2001; 3: 231–248.[CrossRef][Medline] [Order article via Infotrieve]
54. Venugopal SK, Devaraj S, Jialal I. C-reactive protein decreases prostacyclin release from human aortic endothelial cells. Circulation. 2003; 108: 1676–1678.
55. Bolli R, Shinmura K, Tang XL, et al. Discovery of a new function of cyclooxygenase (COX)-2: COX-2 is a cardioprotective protein that alleviates ischemia/reperfusion injury and mediates the late phase of preconditioning. Cardiovasc Res. 2002; 55: 506–519.
56. Haider A, Lee I, Grabarek J, et al. Dual functionality of cyclooxygenase-2 as a regulator of tumor necrosis factor-mediated G1 shortening and nitric oxide-mediated inhibition of vascular smooth muscle cell proliferation. Circulation. 2003; 108: 1015–1021.
57. Coffey MJ, Phare SM, Peters-Golden M. Interaction between nitric oxide, reactive oxygen intermediates, and peroxynitrite in the regulation of 5-lipoxygenase metabolism. Biochim Biophys Acta. 2002; 1584: 81–90.[Medline] [Order article via Infotrieve]
58. Velardez MO, Ogando D, Franchi AM, et al. Role of nitric oxide in the metabolism of arachidonic acid in the rat anterior pituitary gland. Mol Cell Endocrinol. 2001; 172: 7–12.[CrossRef][Medline] [Order article via Infotrieve]
59. Coffey MJ, Phare SM, Peters-Golden M. Prolonged exposure to lipopolysaccharide inhibits macrophage 5-lipoxygenase metabolism via induction of nitric oxide synthesis. J Immunol. 2000; 165: 3592–3598.
60. Luchtefeld M, Drexler H, Schieffer B. 5-Lipoxygenase is involved in the angiotensin II–induced NAD(P)H-oxidase activation. Biochem Biophys Res Commun. 2003; 308: 668–672.[CrossRef][Medline] [Order article via Infotrieve]
61. Penglis PS, Cleland LG, Demasi M, et al. Differential regulation of prostaglandin E2 and thromboxane A2 production in human monocytes: implications for the use of cyclooxygenase inhibitors. J Immunol. 2000; 165: 1605–1611.
62. Ma H, Hara A, Xiao CY, et al. Increased bleeding tendency and decreased susceptibility to thromboembolism in mice lacking the prostaglandin E receptor subtype EP(3). Circulation. 2001; 104: 1176–1180.
63. Coffey MJ, Coles B, O’Donnell VB. Interactions of nitric oxide-derived reactive nitrogen species with peroxidases and lipoxygenases. Free Radic Res. 2001; 35: 447–464.[CrossRef][Medline] [Order article via Infotrieve]
64. Kobayashi T, Tahara Y, Matsumoto M, et al. Roles of thromboxane A(2) and prostacyclin in the development of atherosclerosis in apoE-deficient mice. J Clin Invest. 2004; 114: 784–794.[CrossRef][Medline] [Order article via Infotrieve]
65. FitzGerald GA, Smith B, Pedersen AK, et al. Increased prostacyclin biosynthesis in patients with severe atherosclerosis and platelet activation. N Engl J Med. 1984; 310: 1065–1068.[Abstract]
66. Moreno-Sanchez R, Bravo C, Vasquez C, et al. Inhibition and uncoupling of oxidative phosphorylation by nonsteroidal anti-inflammatory drugs: study in mitochondria, submitochondrial particles, cells, and whole heart. Biochem Pharmacol. 1999; 57: 743–752.[CrossRef][Medline] [Order article via Infotrieve]
67. Rossat J, Maillard M, Nussberger J, et al. Renal effects of selective cyclooxygenase-2 inhibition in normotensive salt-depleted subjects. Clin Pharmacol Ther. 1999; 66: 76–84.[CrossRef][Medline] [Order article via Infotrieve]
68. Aw TJ, Haas SJ, Liew D, et al. Meta-analysis of cyclooxygenase-2 inhibitors and their effects on blood pressure. Arch Intern Med. 2005; 165: 490–496.
69. Sowers JR, White WB, Pitt B, et al. The effects of cyclooxygenase-2 inhibitors and nonsteroidal anti-inflammatory therapy on 24-hour blood pressure in patients with hypertension, osteoarthritis, and type 2 diabetes mellitus. Arch Intern Med. 2005; 165: 161–168.
70. Swan SK, Rudy DW, Lasseter KC, et al. Effect of cyclooxygenase-2 inhibition on renal function in elderly persons receiving a low-salt diet: a randomized, controlled trial. Ann Intern Med. 2000; 133: 1–9.
71. Singh G, Miller JD, Huse DM, et al. Consequences of increased systolic blood pressure in patients with osteoarthritis and rheumatoid arthritis. J Rheumatol. 2003; 30: 714–719.
72. Pfizer Inc. Protocol No. IQ5–97–02–001: a double-blind, randomized, placebo-controlled, comparative study of Celecoxib (SC–58635) for the inhibition of progression of Alzheimer’s disease; Protocol No. EQ5–98–02–002: a placebo-controlled evaluation of the long-term efficacy and safety of Celecoxib (SC-58635) in Alzheimer’s disease. ClinicalStudyResults.org, Available at: http://www.clinicalstudyresults.org/drugdetails/?inn_name_id=55&sort=c.company_name&. Accessed February 1, 2005.
73. Juni P, Nartey L, Reichenbach S, et al. Risk of cardiovascular events and rofecoxib: cumulative meta-analysis. Lancet. 2004; 364: 2021–2029.[CrossRef][Medline] [Order article via Infotrieve]
74. Solomon SD, McMurray JJ, Pfeffer MA, et al. Cardiovascular risk associated with celecoxib in a clinical trial for colorectal adenoma prevention. N Engl J Med. 2005; 352: 1071–1080.
75. Bombardier C, Laine L, Reicin A, et al. Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis: VIGOR Study Group. N Engl J Med. 2000; 343: 1520–1528.
76. Mukherjee D, Nissen SE, Topol EJ. Risk of cardiovascular events associated with selective COX-2 inhibitors. JAMA. 2001; 286: 954–959.
77. Solomon DH, Schneeweiss S, Glynn RJ, et al. Relationship between selective cyclooxygenase-2 inhibitors and acute myocardial infarction in older adults. Circulation. 2004; 109: 2068–2073.
78. Kimmel SE, Berlin JA, Reilly M, et al. Patients exposed to rofecoxib and celecoxib have different odds of nonfatal myocardial infarction. Ann Intern Med. 2005; 142: 157–164.
79. Levesque LE, Brophy JM, Zhang B The risk for myocardial infarction with cyclooxygenase-2 inhibitors: a population study of elderly adults. Ann Intern Med. 2005; 142: 481–489.
80. Graham DJ, Campen D, Hui R, et al. Risk of acute myocardial infarction and sudden cardiac death in patients treated with cyclo-oxygenase 2 selective and non-selective non-steroidal anti-inflammatory drugs: nested case-control study. Lancet. 2005; 365: 475–481.[Medline] [Order article via Infotrieve]
81. Furberg CD, Psaty BM, FitzGerald GA. Parecoxib, valdecoxib, and cardiovascular risk. Circulation. 2005; 111: 249.
82. US Food and Drug Administration. FDA issues Public Health Advisory recommending limited use of Cox-2 inhibitors: agency requires evaluation of prevention studies involving Cox-2 selective agents. FDA Talk Paper, 2004. Available at: http://www.fda.gov/bbs/topics/ANSWERS/2004/ANS01336.html. Accessed January 19, 2005.
83. National Institutes of Health, NIH Office of Communications and Public Liaison. Use of non-steroidal anti-inflammatory drugs suspended in large Alzheimer’s disease prevention trial. Available at: http://www.nih.gov/news/or/dec2004/od-20.htm. Accessed April 12, 2005.
84. Whalen J. Stronger warnings are ordered for Cox-2 inhibitors in Europe. The Wall Street Journal Europe. February 18–February 20, 2005:A4.
85. Adetunji L, Bowe C Watchdog imposes curbs on dosages of Cox-2 pain drugs. London Financial Times. February 18, 2005:15.
86. Farkouh ME, Kirshner H, Harrington RA, et al. Comparison of lumiracoxib with naproxen and ibuprofen in the Therapeutic Arthritis Research and Gastrointestinal Event Trial (TARGET), cardiovascular outcomes: randomised controlled trial. Lancet. 2004; 364: 675–684.[CrossRef][Medline] [Order article via Infotrieve]
87. US Food and Drug Administration. FDA announces series of changes to the class of marketed non-steroidal anti-inflammatory drugs (NSAIDS). FDA News. Available at: http://www.fda.gov/bbs/topics/news/2005/NEW01171.html. Accessed April 7, 2005.
88. Johnsen SP, Larsson H, Tarone RE, et al. Risk of hospitalization for myocardial infarction among users of rofecoxib, celecoxib, and other NSAIDs: a population-based case-control study. Arch Intern Med. 2005; 165: 978–984.
89. Catella-Lawson F, Reilly MP, Kapoor SC, et al. Cyclooxygenase inhibitors and the antiplatelet effects of aspirin. N Engl J Med. 2001; 345: 1809–1817.
90. Weksler BB, Pett SB, Alonso D, et al. Differential inhibition by aspirin of vascular and platelet prostaglandin synthesis in atherosclerotic patients. N Engl J Med. 1983; 308: 800–805.[Abstract]
91. Okie S. What ails the FDA? N Engl J Med. 2005; 352: 1063–1066.
92. Hawthorne F. How powerful is industry? In: Inside the FDA: The Business and Politics Behind the Drugs We Take and the Food We Eat. Hoboken, NJ: John Wiley & Sons; 2005: 143–177.
93. Hawthorne F. The FDA meets Madison Avenue. In: Inside the FDA: The Business and the Politics Behind the Drugs We Take and the Food We Eat. Hoboken, NJ: John Wiley & Sons; 2005: 253–272.
94. Coronary Drug Project Research Group. Practical aspects of decision making in clinical trials: the coronary drug project as a case study. Control Clin Trials. 1981; 1: 363–376.[CrossRef][Medline] [Order article via Infotrieve]
95. US Department of Health and Human Services. Reforms will improve oversight and openness at FDA: Secretary Leavitt meets with employees and announces a new day at FDA. FDA Press Office, 2/15/05. Available at: www.fda.gov/cder/drugsafety.htm, http://www.hhs.gov/news. Accessed April 12, 2005.
96. Hawthorne F. The next 100 years. In: Inside the FDA: The Business and Politics Behind the Drugs We Take and the Food We Eat. Hoboken, NJ: John Wiley & Sons; 2005: 285–308.
97. Bennett CL, Nebeker JR, Lyons EA, et al. The Research on Adverse Drug Events and Reports (RADAR) project. JAMA. 2005; 293: 2131–2140.
98. Roden DM, Temple R. The US Food and Drug Administration Cardiorenal Advisory Panel and the drug approval process. Circulation. 2005; 111: 1697–1702.
99. Hippisley-Cox J, Coupland C. Risk of myocardial infarction in patients taking cyclo-oxygenase-2 inhibitors or conventional non-steroidal anti-inflammatory drugs: population based nested case-control analysis. Br Med J. 2005; 330: 1366–1343.
100. Waxman HA. The lessons of Vioxx: drug safety and sales. N Engl J Med. 2005; 352: 2576–2578.
This article has been cited by other articles:
 |
|
 |
|
  G. H. Gislason, J. N. Rasmussen, S. Z. Abildstrom, T. K. Schramm, M. L. Hansen, E. L. Fosbol, R. Sorensen, F. Folke, P. Buch, N. Gadsboll, et al. Increased Mortality and Cardiovascular Morbidity Associated With Use of Nonsteroidal Anti-inflammatory Drugs in Chronic Heart Failure Arch Intern Med, January 26, 2009; 169(2): 141 - 149. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. L. Bhatt, J. Scheiman, N. S. Abraham, E. M. Antman, F. K.L. Chan, C. D. Furberg, D. A. Johnson, K. W. Mahaffey, E. M. Quigley, R. A. Harrington, et al. ACCF/ACG/AHA 2008 Expert Consensus Document on Reducing the Gastrointestinal Risks of Antiplatelet Therapy and NSAID Use: A Report of the American College of Cardiology Foundation Task Force on Clinical Expert Consensus Documents J. Am. Coll. Cardiol., October 28, 2008; 52(18): 1502 - 1517. [Full Text] [PDF]
|
|
|
|
 |
|
 Writing Committee Members, D. L. Bhatt, J. Scheiman, N. S. Abraham, E. M. Antman, F. K.L. Chan, C. D. Furberg, D. A. Johnson, K. W. Mahaffey, and E. M. Quigley ACCF/ACG/AHA 2008 Expert Consensus Document on Reducing the Gastrointestinal Risks of Antiplatelet Therapy and NSAID Use: A Report of the American College of Cardiology Foundation Task Force on Clinical Expert Consensus Documents Circulation, October 28, 2008; 118(18): 1894 - 1909. [Full Text] [PDF]
|
|
|
|
 |
|
 H. Sun, E. Sheveleva, B. Xu, H. Inoue, T. G. Bowden, and Q. M. Chen Corticosteroids induce COX-2 expression in cardiomyocytes: role of glucocorticoid receptor and C/EBP-{beta} Am J Physiol Cell Physiol, October 1, 2008; 295(4): C915 - C922. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Li, M. Hortmann, A. Daiber, M. Oelze, M. A. Ostad, P. M. Schwarz, H. Xu, N. Xia, A. L. Kleschyov, C. Mang, et al. Cyclooxygenase 2-Selective and Nonselective Nonsteroidal Anti-Inflammatory Drugs Induce Oxidative Stress by Up-Regulating Vascular NADPH Oxidases J. Pharmacol. Exp. Ther., September 1, 2008; 326(3): 745 - 753. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. L. Roumie, E. F. Mitchel Jr, L. Kaltenbach, P. G. Arbogast, P. Gideon, and M. R. Griffin Nonaspirin NSAIDs, Cyclooxygenase 2 Inhibitors, and the Risk for Stroke Stroke, July 1, 2008; 39(7): 2037 - 2045. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. D. M. Haag, M. J. Bos, A. Hofman, P. J. Koudstaal, M. M. B. Breteler, and B. H. C. Stricker Cyclooxygenase Selectivity of Nonsteroidal Anti-inflammatory Drugs and Risk of Stroke Arch Intern Med, June 9, 2008; 168(11): 1219 - 1224. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Jelic, M. Padeletti, S. M. Kawut, C. Higgins, S. M. Canfield, D. Onat, P. C. Colombo, R. C. Basner, P. Factor, and T. H. LeJemtel Inflammation, Oxidative Stress, and Repair Capacity of the Vascular Endothelium in Obstructive Sleep Apnea Circulation, April 29, 2008; 117(17): 2270 - 2278. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. D. Solomon, J. Wittes, P. V. Finn, R. Fowler, J. Viner, M. M. Bertagnolli, N. Arber, B. Levin, C. L. Meinert, B. Martin, et al. Cardiovascular Risk of Celecoxib in 6 Randomized Placebo-Controlled Trials: The Cross Trial Safety Analysis Circulation, April 22, 2008; 117(16): 2104 - 2113. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. E. Fredenburgh, O. D. Liang, A. A. Macias, T. R. Polte, X. Liu, D. F. Riascos, S. W. Chung, S. L. Schissel, D. E. Ingber, S. A. Mitsialis, et al. Absence of Cyclooxygenase-2 Exacerbates Hypoxia-Induced Pulmonary Hypertension and Enhances Contractility of Vascular Smooth Muscle Cells Circulation, April 22, 2008; 117(16): 2114 - 2122. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. H. Hennekens and S. Borzak Cyclooxygenase-2 Inhibitors and Most Traditional Nonsteroidal Anti-inflammatory Drugs Cause Similar Moderately Increased Risks of Cardiovascular Disease Journal of Cardiovascular Pharmacology and Therapeutics, March 1, 2008; 13(1): 41 - 50. [Abstract] [PDF]
|
|
|
|
 |
|
 G. Letts and J. Loscalzo Frontiers in Nephrology: Targeting Inflammation Using Novel Nitric Oxide Donors J. Am. Soc. Nephrol., November 1, 2007; 18(11): 2863 - 2869. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Anderson, C. D. Adams, E. M. Antman, C. R. Bridges, R. M. Califf, D. E. Casey Jr, W. E. Chavey II, F. M. Fesmire, J. S. Hochman, T. N. Levin, et al. ACC/AHA 2007 Guidelines for the Management of Patients With Unstable Angina/Non ST-Elevation Myocardial Infarction Executive Summary: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines for the Management of Patients With Unstable Angina/Non ST-Elevation Myocardial Infarction) Developed in Collaboration with the American College of Emergency Physicians, the Society for Cardiovascular Angiography and Interventions, and the Society of Thoracic Surgeons Endorsed by the American Association of Cardiovascular and Pulmonary Rehabilitation and the Society for Academic Emergency Medicine J. Am. Coll. Cardiol., August 14, 2007; 50(7): 652 - 726. [Full Text] [PDF]
|
|
|
|
 |
|
 J. L Sullivan The Big Idea: the coxib crisis Iron, aspirin and heart disease risk revisited J R Soc Med, July 1, 2007; 100(7): 346 - 349. [Full Text] [PDF]
|
|
|
|
 |
|
 B. T. Ivandic, E. Giannitsis, P. Schlick, P. Staritz, H. A. Katus, and T. Hohlfeld Determination of Aspirin Responsiveness by Use of Whole Blood Platelet Aggregometry Clin. Chem., April 1, 2007; 53(4): 614 - 619. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. M. Antman, J. S. Bennett, A. Daugherty, C. Furberg, H. Roberts, and K. A. Taubert Use of Nonsteroidal Antiinflammatory Drugs: An Update for Clinicians: A Scientific Statement From the American Heart Association Circulation, March 27, 2007; 115(12): 1634 - 1642. [Full Text] [PDF]
|
|
|
|
|
|
E. Camenzind, P. G. Steg, and W. Wijns A Cause for Concern Circulation, March 20, 2007; 115(11): 1440 - 1455. [Full Text] [PDF]
|
|
|
|
 |
|
 W. B. White Cardiovascular Effects of the Cyclooxygenase Inhibitors Hypertension, March 1, 2007; 49(3): 408 - 418. [Full Text] [PDF]
|
|
|
|
|
|
B. I. Jugdutt Cyclooxygenase Inhibition and Adverse Remodeling During Healing After Myocardial Infarction Circulation, January 23, 2007; 115(3): 288 - 291. [Full Text] [PDF]
|
|
|
|
 |
|
 T. Szerafin, N. Erdei, T. Fulop, E. T. Pasztor, I. Edes, A. Koller, and Z. Bagi Increased Cyclooxygenase-2 Expression and Prostaglandin-Mediated Dilation in Coronary Arterioles of Patients With Diabetes Mellitus Circ. Res., September 1, 2006; 99(5): e12 - 317. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. D. Levy Myocardial 15-Epi-lipoxin A4 Generation Provides a New Mechanism for the Immunomodulatory Effects of Statins and Thiazolidinediones Circulation, August 29, 2006; 114(9): 873 - 875. [Full Text] [PDF]
|
|
|
|
 |
|
 D. L. Bhatt NSAIDS and the risk of myocardial infarction: do they help or harm? Eur. Heart J., July 2, 2006; 27(14): 1635 - 1636. [Full Text] [PDF]
|
|
|
|
|
|
G. H. Gislason, S. Jacobsen, J. N. Rasmussen, S. Rasmussen, P. Buch, J. Friberg, T. K. Schramm, S. Z. Abildstrom, L. Kober, M. Madsen, et al. Risk of Death or Reinfarction Associated With the Use of Selective Cyclooxygenase-2 Inhibitors and Nonselective Nonsteroidal Antiinflammatory Drugs After Acute Myocardial Infarction Circulation, June 27, 2006; 113(25): 2906 - 2913. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. S. Hochman and N. R. Shah What Price Pain Relief? Circulation, June 27, 2006; 113(25): 2868 - 2870. [Full Text] [PDF]
|
|
|
|
 |
|
 E. J Sanchez-Delgado Life without COX 2 inhibitors: Risks and benefits are determined by dose and potency. BMJ, June 17, 2006; 332(7555): 1451 - 1452. [Full Text]
|
|
|
|
|
|
C. Buccellati, A. Sala, G. Rossoni, V. Capra, G. E. Rovati, A. Di Gennaro, G. Folco, S. Colli, and C. Casagrande Pharmacological Characterization of 2NTX-99 [4-Methoxy-N1-(4-trans-nitrooxycyclohexyl)-N3-(3-pyridinylmethyl)-1,3-benzenedicarboxamide], a Potential Antiatherothrombotic Agent with Antithromboxane and Nitric Oxide Donor Activity in Platelet and Vascular Preparations J. Pharmacol. Exp. Ther., May 1, 2006; 317(2): 830 - 837. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. E Vonkeman, J. R B J Brouwers, and M. A F J van de Laar Understanding the NSAID related risk of vascular events. BMJ, April 15, 2006; 332(7546): 895 - 898. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||