Cardiovascular Adverse Effects of Anti-Inflammatory Drugs

Transcription

1 Send Orders of Reprints at Anti-Inflammatory & Anti-Allergy Agents in Medicinal Chemistry, 2013, 12, Cardiovascular Adverse Effects of Anti-Inflammatory Drugs Camille Roubille 1,*, Johanne Martel-Pelletier 1, Jean-Marc Davy 2, Boulos Haraoui 1 and Jean-Pierre Pelletier 1 1 Osteoarthritis Research Unit, University of Montreal Hospital Research Center (CRCHUM), Notre-Dame Hospital, 1560 Sherbrooke Street East, Montreal, Quebec, Canada; 2 Cardiology Department, Arnaud de Villeneuve University Hospital, Montpellier, France Abstract: Anti-inflammatory drugs consist of non-steroidal anti-inflammatory drugs (NSAIDs) including non-selective nsnsaids, aspirin, and cyclooxygenase-2 (COX-2)-selective inhibitors also referred to as coxibs, and glucocorticoids (GCs). They are worldwide prescribed drugs for many musculoskeletal conditions, such as osteoarthritis and inflammatory rheumatic diseases. Anti-inflammatory drugs can exert deleterious effects on the cardiovascular system, excluding aspirin. NSAIDs, especially coxibs, have been demonstrated to increase cardiovascular risk and have generated many concerns leading to the reassessment of their benefit/risk ratio. GCs may also induce cardiovascular events, but evidence seems to be less clear. Before prescribing these drugs, an assessment of cardiovascular risk may be judicious. In this review, anti-inflammatory drugs, coxibs, nsnsaids and GCs, and the risk of cardiovascular events will be discussed. Keywords: Cardiovascular system, cardiovascular diseases, anti-inflammatory agents, NSAIDs, cyclooxygenase 2 inhibitors, glucocorticoids. INTRODUCTION Anti-inflammatory drugs consist of non-steroidal antiinflammatory drugs (NSAIDs) including non-selective nsnsaids, aspirin, and cyclooxygenase-2 (COX-2)-selective inhibitors also referred to as coxibs, and glucocorticoids (GCs). nsnsaids are worldwide prescribed drugs mainly used to reduce pain and improve function in many musculoskeletal conditions, such as osteoarthritis and inflammatory rheumatic diseases. Whereas nsnsaids inhibit the two recognized forms of enzymes responsible for the formation of prostanoids, COX-1 and COX-2, coxibs are selective inhibitors of the COX-2 isoenzyme [1] (Fig. (1)). Coxibs vary in their selectivity for the COX-2 versus the COX-1 enzyme; for instance, rofecoxib > valdecoxib > parecoxib > celecoxib. Coxibs were developed to reduce the gastrointestinal adverse effects caused by nsnsaids. Indeed, as the anti-inflammatory effect was believed to be mediated by the inhibition of COX-2 and the adverse gastrointestinal effects by the inhibition of COX-1, selective COX-2 inhibitors were thought to be safer. Nevertheless, like traditional nsnsaids, coxibs have also been shown to induce cardiovascular adverse effects, limiting their use. Coxibs were shown to be as effective as conventional nsnsaids for pain relief with fewer gastrointestinal complications, but with a potential cardiovascular risk. Cardiovascular risk seems to be shared by all the NSAIDs, selective or non-selective of COX-2, *Address correspondence to this author at the Osteoarthritis Research Unit, University of Montreal Hospital Research Center (CRCHUM), Notre-Dame Hospital, 1560 Sherbrooke Street East, Montreal, Quebec, Canada; Tel: , Ext ; Fax: ; camille.roubille@gmail.com excluding aspirin. Only aspirin offers primary and secondary cardiovascular prophylaxis and will not be described here, nor will the non-cardiovascular side effects of NSAIDs and of GCs, as they are not the thrust of the present review. A recent meta-analysis demonstrated that all NSAIDs compared to placebo provide, at different levels, cardiovascular and cerebrovascular risks: rofecoxib was associated with the highest risk of myocardial infarction (rate ratio 2.12, 95% credibility interval1.26 to 3.56), followed by lumiracoxib (2.00, 0.71 to 6.21). Ibuprofen was associated with the highest risk of stroke (3.36, 1.00 to 11.6), followed by diclofenac (2.86, 1.09 to 8.36). Etoricoxib (4.07, 1.23 to 15.7) and diclofenac (3.98, 1.48 to 12.7) were associated with the highest risk of cardiovascular death. was the least harmful [2]. Before prescribing these drugs, an assessment of cardiovascular risk and calculating the patient s baseline risk of cardiovascular diseases with, for instance, the Framingham risk calculators, may be judicious. The American Heart Association (AHA) guidelines discourage NSAID treatment (except aspirin) in patients with established cardiovascular disease, such as those with myocardial infarction (MI) or heart failure. However, if NSAID treatment is unavoidable, the guidelines propose that the duration of the treatment should be as short as possible [3]. In regard to osteoarthritis, NSAIDs are recommended for patients who are unresponsive to appropriate dosages of acetaminophen, and should be prescribed at the lowest possible dose, for the shortest duration, and preferentially for inflammatory flares [4]. Moreover, for patients older than 75 years, oral NSAIDs are not recommended by the guidelines, which prefer topical NSAIDs and other medications such as duloxetine, tramadol, or intra-articular hyaluronan injections [4]. COX-2 inhibitors are associated with lower X/13 $ Bentham Science Publishers

2 56 Anti-Inflammatory & Anti-Allergy Agents in Medicinal Chemistry, 2013, Vol. 12, No. 1 Roubille et al. Proatherogene ROS Epoxides Leukotrienes Arachidonic Acid Eicosanoids COX-1 Platelets Thromboxane A2 COX-2 PGE 2 PGI 2 PGE 1 Platelet aggregation Vasoconstriction Blood pressure Inflammation Reperfusion injury Infarct size Myocardial stunning Blood pressure Collateral circulation Vasoconstriction Blood pressure Platelet aggregation Infarct size Blood pressure vasodilatation Fig. (1). Schematic pathophysiology linking COX-1, COX-2 and cardiovascular risk. gastrointestinal toxicity than nsnsaids [5]. For patients who have suffered from upper gastrointestinal ulcers without any gastrointestinal bleeding in the prior year, the osteoarthritis guidelines recommend the use of either a COX-2 inhibitor rather than an nsnsaid, or an nsnsaid combined with a proton-pump inhibitor. For patients with reports of gastrointestinal bleeding within the past year, the use of a COX-2 inhibitor in association with a proton-pump inhibitor is recommended [4]. With regard to the GCs, their potential harmful effects on the cardiovascular system are well-known but not very clearly evidenced in the literature. In this review, anti-inflammatory drugs, coxibs, nsnsaids and GCs, and the risk of cardiovascular events will be discussed (see Table 1 and 2). Taking into account the wide spectrum of the field, this review is aimed at a comprehensive synthesis rather than an exhaustive listing. COXIBS Although many trials were not powered to detect a difference in the rates of cardiovascular events, more such events were documented in the coxib group than in the other groups. Over the past twelve years, a substantial body of literature highlighted the cardiovascular risk of coxibs (Table 1). Indeed, many randomized controlled trials (s) studying the gastrointestinal benefits of coxibs in comparison to nsnsaids generated some concerns about the cardiovascular safety profile of coxibs and led to reconsidering their benefit/risk ratio. Rofecoxib The VIGOR (Vioxx gastrointestinal outcomes research) trial [6] assessed 8,076 patients suffering from rheumatoid arthritis (RA) who received either naproxen 500 mg twice daily, a traditional nsnsaid, or rofecoxib 50 mg/d, a COX- 2 inhibitor that was expected to be associated with less gastrointestinal adverse events than naproxen. This study showed that rofecoxib indeed was associated with less gastrointestinal side effects but also a higher risk of MI than naproxen (MI was less common in the naproxen group (0.1%) than in the rofecoxib group (0.4%), 95% confidence interval (CI) 0.1 to 0.6%, relative risk (RR) 0.2, 95% CI 0.1 to 0.7). This result was interpreted by the authors as a beneficial cardioprotective effect of naproxen and not as a harmful effect of rofecoxib. Other studies have provided conflicting data [7, 8] demonstrating that rofecoxib was not associated with an increased cardiovascular risk compared to placebo (RR 0.84, 95% CI 0.51 to 1.30) or non-naproxen nsnsaids (RR 0.79, 95% CI 0.40 to 1.55) [8]. Notwithstanding, Komstam et al. also found that rofecoxib was associated with an increased cardiovascular risk compared to naproxen (RR 1.69, 95% CI 1.07 to 2.69) and also concluded a cardioprotective antiplatelet effect of naproxen [8]. However, several studies and meta-analyses have demonstrated a consistent increased cardiovascular risk of rofecoxib [9-12] and refuted the cardioprotective effect of naproxen [12, 13]. The use of rofecoxib >25 mg/day was associated with a higher adjusted RR of myocardial infarction than rofecoxib 25 mg/d [9, 11, 12].

3 Cardiovascular Adverse Effects of Anti-Inflammatory Drugs Anti-Inflammatory & Anti-Allergy Agents in Medicinal Chemistry, 2013, Vol. 12, No Table 1. The Relationship between coxibs and NSAID therapies and cardiovascular events. Drug Reference Design Treatment Outcome VIGOR trial Bombardier et al [6] No increase in CV risk Komstam et al [8] Non-naproxen RR 0.84 ( ) RR 0.79 ( ) RR 1.69 ( ) Solomon et al [9] Matched case-control study Celecoxib No NSAID use OR 1.24 ( ) OR1.14 ( ) Kimmel et al [10] Case-control study Celecoxib OR 2.72 ( ) OR 3.39 ( ) Levesque et al [11] Case-control study No NSAID use Rate ratio 1.24 ( ) Rofecoxib Graham et al [12] Case-control study Celecoxib risk of coronary disease OR 1.59 ( ) Juni et al [13] and NSAID RR 2.30 ( ) APPROVE study Baron et al [14] CV events (MI+strokes) RR 1.92 ( ) CV risk Garner et al [17] Cochrane RR 2.36 ( ) RR 4.48 ( ) VICTOR trial CV events Kerr et al [18] RR 2.66 ( ) Zhang et al [19] risk of arrhythmias RR 2.90 ( ) Trelle et al [34] Rate ratio 2.12 ( ) CV events Celecoxib APC trial Bertagnolli et al [20] Risk ratio 2.6 ( ) for 200mgX2/d Risk ratio 3.4 ( ) for 400mgX2/d CV risk Cross Trial Safety Analysis - Solomon et al [25] Pooled analysis of s HR 1.1 ( ) for 400 mg/d HR 1.8 ( ) for 200mgX2/d HR 3.1 ( ) for 400mgX2/d

4 58 Anti-Inflammatory & Anti-Allergy Agents in Medicinal Chemistry, 2013, Vol. 12, No. 1 Roubille et al. (Table 1) continued Drug Reference Design Treatment Outcome CLASS study Silverstein et al [26] Ibuprofen Diclofenac No CV excess Trelle et al [34] Rate ratio 2.00 ( ) Lumiracoxib TARGET trial Farkouh et al [27] Ibuprofen Non-significant increase in CV events HR 1.14 ( ) Ott et al [28] CV events Valdecoxib/ Parecoxib Nussmeier et al [29] CV events Risk ratio 3.7 ( ) for combination Risk ratio 2.0 ( ) for valdecoxib alone MEDAL program Cannon et al [30] Diclofenac No excess CV events HR 0.95 ( ) hypertension events Etoricoxib Trelle et al [34] Reginster et al [35] s risk of CV death Rate ratio 4.07 ( ) CV events Curtis et al [36] Non-naproxen NSAIDs (ibuprofen+diclofenac) No excess CV events RR 0.83 ( ) ADAPT study [42] deaths, strokes, MI, HF HR 1.63 ( ) CV death nsnsaids Gislason et al [24] Case-crossover analysis Ibuprofen: HR 1.50 ( ) Diclofenac: HR 2.40 ( ) Trelle et al [34] Ibuprofen: strokes Rate ratio 3.36 ( ) Diclofenac: CV death Rate ratio 3.98 ( ) = randomized controlled trial, = meta-analysis, MI = myocardial infarction, CV = cardiovascular, HF = heart failure, OR = odds ratio, RR = relative risk, HR = hazard ratio. The APPROVe (Adenomatous Polyp PRevention On Vioxx) study was a randomized, placebo-controlled, doubleblind trial assessing whether long-term use (three years) of rofecoxib 25 mg/d could reduce the risk of recurrence of adenomatous polyps in 2,586 patients with history of colorectal adenomas [14]. This study was terminated prematurely because preliminary data demonstrated that rofecoxib was associated with an increased risk of cardiovascular events beginning after 18 months of treatment, due to increased MI and strokes compared to placebo (1.50 events per 100 patient-years in the rofecoxib group vs 0.78 events per 100 patient-years in the placebo

5 Cardiovascular Adverse Effects of Anti-Inflammatory Drugs Anti-Inflammatory & Anti-Allergy Agents in Medicinal Chemistry, 2013, Vol. 12, No Table 2. The relationship between GC therapy and cardiovascular events. Patient Group Outcome Reference GC users cohort studies risk of CV events Wei et al [48] risk of HF Souverein et al [50] incidence of MI Wolfe et al [53] carotid artery plaque and arterial stiffness Del Rincon et al [52] risk of CV events in RF-positive patients Davis et al [67] RA cohort studies risk of mortality >1 year and >10 years Sihvonen et al [54] No excess CV events of low dose GC therapy Da Silva et al [46] No excess CV events Bernatsky et al [60] No excess CV events Avina-Zubieta et al [59] No increase in atherosclerosis or endothelial dysfunction Hafstrom et al [73] carotid abnormalities and atherosclerosis Doria et al. [55] SLE cohort studies risk of CV events Svenungsson et al [56] No increase in CV diseases Roman et al [63] No increase in coronary-artery atherosclerosis Asanuma et al [62] Polymyalgia rheumatica cohort study No increase in CV diseases Maradit Kremers et al. 2007[61] GC = glucocorticoids, CV = cardiovascular, HF = heart failure, MI = myocardial infarction, RA = rheumatoid arthritis, SLE = systemic lupus erythematosus, RF = rheumatoid factor. group, RR 1.92, 95% CI 1.19 to 3.11, p=0.008) [15]. In the long-term follow-up of participants of that trial, a persistent increased cardiovascular risk one year after rofecoxib discontinuation was reported [16]. The Cochrane analysis of rofecoxib efficacy and toxicity in RA patients also reported the same increase in cardiovascular risk for rofecoxib compared to naproxen (RR 2.36, 95% CI 1.38 to 4.02) and of having non-fatal MI (RR 4.48, 95% CI 1.52 to 13.23) [17]. These results corroborate those of the VICTOR (Vioxx In Colorectal cancer Therapy: definition of Optimal Regimen) trial, which was a placebo-controlled study to assess whether rofecoxib 25 mg/d could reduce the recurrence of tumor after surgery and chemotherapy or radiotherapy. Data showed that patients in the rofecoxib group had an increased frequency of cardiovascular events (RR 2.66, 95% CI 1.03 to 6.86, p=0.04) [18]. Rofecoxib has been shown to be associated with more acute MI than celecoxib in patients 65 years of age and older (odds ratio [OR] 1.24, 95% CI 1.05 to 1.46, p=0.011) [9]. Other studies have confirmed these results [10, 12]. In 2006, Zhang et al. reported from the analysis of 114 randomized double-blind clinical trials that compared to controls, rofecoxib was associated with increased risk of arrhythmia (RR 2.90, 95% CI 1.07 to 7.88) and composite renal events (RR 1.53, 95% CI 1.33 to 1.76) [19]. Rofecoxib was voluntarily withdrawn from the worldwide market in Celecoxib In the APC (Adenomatous Prevention with Celecoxib) trial, celecoxib was found to be effective for the prevention of colorectal adenomas, but also to increase the risk of cardiovascular events compared to placebo (RR 2.6, 95% CI 1.1 to 6.1 for the low dose of 200 mg twice daily, and RR 3.4, 95% CI 1.5 to 7.9 for the high dose of 400 mg twice daily), although the trial was not designed for cardiovascular purposes [20, 21], leading to the early termination of the study [20]. This cardiotoxicity appeared to be dose-related. Of the 679 patients randomized to placebo, 4 (0.6%) either died or sustained a non-fatal MI, whereas of the 1,356 patients assigned to the celecoxib arms, 27 (2.0%) died or had a non-fatal MI, which means a 3.4-fold increased risk of cardiovascular events for celecoxib compared to placebo. The number of patients who need to be treated by celecoxib to observe one MI or death, named as the number needed to harm (NNH) has been calculated at 71 patients [22]. In data from this study combined with the Pre-SAP (Prevention of Spontaneous Adenomatous Polyps) trial, celecoxib showed a nearly 2-fold-increase in cardiovascular risk [23]. Moreover, in a Danish study which reported the risk of recurrent MI and death related to the use of selective COX-2 inhibitors and nsnsaids in patients after first-time acute MI, the death rate of patients exposed to celecoxib was 165 per 1,000 patientyear, (95% CI 137 to 198), and the NNH was 14 [24]. The NNH was found lower than in the APC trial, which suggests that harmful effects of coxibs such as celecoxib may be more significant in patients with a prior MI.

6 60 Anti-Inflammatory & Anti-Allergy Agents in Medicinal Chemistry, 2013, Vol. 12, No. 1 Roubille et al. In the Cross Trial Safety Analysis, Solomon et al. [25] demonstrated from pooled data of 6 placebo-controlled trials, a dose regimen dependent increase in cardiovascular risk, with the 400 mg once daily dose of celecoxib associated with the lowest risk (hazard ratio [HR] 1.1, 95% CI 0.6 to 2.0), the 200 mg twice daily dose of celecoxib associated with an intermediate risk (HR 1.8, 95% CI 1.1 to 3.1), and the 400 mg twice daily dose of celecoxib associated with the highest risk (HR 3.1, 95% CI 1.5 to 6.1). Moreover, they observed that patients with the lowest baseline risk were at the lowest relative risk for celecoxib-related events, given that the risk of cardiovascular events increases with baseline cardiovascular risk. The CLASS (Celecoxib Long-term Arthritis Safety Study) study [26] aimed at determining whether celecoxib was associated with a lower incidence of upper gastrointestinal toxic effects and other adverse effects during a 6- month treatment period and compared celecoxib 400 mg twice daily, ibuprofen 800 mg three times daily, and diclofenac 75 mg twice daily. Aspirin use for cardiovascular prophylaxis ( 325 mg/d) was permitted and 20.9% of the patients were taking this low-dosage of aspirin. No difference was noted in the incidence of cardiovascular events between celecoxib, ibuprofen and diclofenac irrespective of aspirin use. Lumiracoxib In a meta-analysis of 31 s, Trelle et al. assessed the cardiovascular safety of non-steroidal anti-inflammatory drugs and found an increased risk of MI for lumiracoxib compared to placebo (rate ratio 2.00, 95% credibility interval 0.71 to 6.21) [2]. In the TARGET (Therapeutic Arthritis Research and Gastrointestinal Event Trial) trial [27], cardiovascular safety of the most selective coxib, lumiracoxib, was compared with that of naproxen and ibuprofen in 18,000 patients with osteoarthritis. Data showed a non-significant increase in cardiovascular side effects with lumiracoxib 400 mg/d than with naproxen 500 mg twice daily or ibuprofen 800 mg three times daily [27]. Indeed, at 1-year follow-up, there were 109 cardiovascular or cerebrovascular events reported, of which 59 (0.65%) occurred in the lumiracoxib group and 50 (0.55%) in the nsnsaid groups (HR 1.14, 95% CI 0.78 to 1.66; p=0.51). These data could suggest that lumiracoxib may be as safe as naproxen and ibuprofen. On the other hand, as there was no placebo group, these findings could also raise serious concerns as these three drugs could thus pose increased cardiovascular risks at similar levels [22]. Another concern is whether aspirin use might reduce the cardiovascular risk of coxibs, as the relative risk of MI was reduced from 2.37 in nonusers to 1.36 in those patients taking aspirin when lumiracoxib was compared with naproxen. Valdecoxib and parecoxib A randomized controlled trial [28] in patients undergoing coronary artery bypass graft (CABG) surgery and receiving either placebo or parecoxib, the intravenously administered prodrug of valdecoxib, followed by oral valdecoxib for 14 days, revealed an excess of cardiovascular events, which were statistically non-significant. Indeed, the incidences of cerebrovascular complications, MI, and renal dysfunction, were proportionately greater in the parecoxib/valdecoxib group although not significantly different between the groups [28]. In another study that randomly assigned 1,671 postcoronary artery bypass patients to valdecoxib plus parecoxib, valdecoxib alone, or placebo, the relative risk for cardiovascular events over 30 days was 3.7 (95% CI 1.0 to 13.5; p=0.03) for the combination and 2.0 (95% CI 0.5 to 8.1; p=0.31) for valdecoxib alone [29]. Valdecoxib was withdrawn from the market in Etoricoxib The MEDAL (Multinational Etoricoxib and Diclofenac Arthritis Long-term) program consisted of three trials: the EDGE (Etoricoxib versus Diclofenac sodium Gastrointestinal tolerability and Effectiveness) trial, EDGE- II and the MEDAL study. It was the first cardiovascular outcomes program prospectively designed to evaluate the relative thrombotic cardiovascular risk of a coxib and a traditional NSAID. The results from the entire pooled MEDAL program comparing etoricoxib to diclofenac in osteoarthritis and RA patients reported similar rates of thrombotic cardiovascular events. The event rates were 1.24 per 100 patient-years for etoricoxib compared to 1.30 per 100 patient-years for diclofenac (HR 0.95, 95% CI 0.81 to 1.11) [30]. These data concluded that the thrombotic cardiovascular risk for etoricoxib was similar to that of diclofenac. Discontinuations because of hypertension were more frequent with both doses of etoricoxib (60 mg/d and 90 mg/d) than with diclofenac. In the MEDAL study, a higher rate of congestive heart failure was seen with etoricoxib 90 m/d than with diclofenac, although non-significant, but there was no difference between diclofenac and etoricoxib 60 mg/d [30]. More particularly, the MEDAL study alone reported the non-inferiority of etoricoxib and diclofenac with respect to cardiovascular risk, with an HR of 0.96 (95% CI 0.81 to 1.15) [31]. These results are consistent with those from the pooled MEDAL program. The EDGE trial demonstrated that rates of thrombotic cardiovascular events were similar with etoricoxib and diclofenac (1.25 vs 1.15 events per 100 patient-years, respectively; HR 1.07, 95% CI 0.65 to 1.74) [32]. However, etoricoxib 90 mg/d was associated with more hypertension compared to diclofenac 50 mg three times daily and resulted in discontinuation due to hypertension-related adverse events (2.3% vs 0.7%; p<0.001) [32]. These results were confirmed in the EDGE-II trial, with an incidence of discontinuations for hypertension-related adverse events significantly higher with etoricoxib 90 mg/d (2.5%) compared with diclofenac 75 mg twice daily (1.5%; p<0.001) [33]. However, this did not translate into an increase in thrombotic cardiovascular events in the etoricoxib group. It should be noted that in one meta-analysis, Trelle et al. reported that etoricoxib (rate ratio 4.07, 95% credibility interval 1.23 to 15.7) and diclofenac (rate ratio 3.98, 95% credibility interval 1.48 to 12.7) were associated with the highest risk of cardiovascular death [34]. It would seem, therefore, that these medications would not be among the

7 Cardiovascular Adverse Effects of Anti-Inflammatory Drugs Anti-Inflammatory & Anti-Allergy Agents in Medicinal Chemistry, 2013, Vol. 12, No first choices for pain relief with regard to safety, especially in individuals with, or at risk for, cardiovascular disease [3]. Another study in osteoarthritis patients assessing the efficacy of etoricoxib 60 mg/d compared to naproxen 500 mg twice daily found a greater proportion of patients experiencing a thrombotic cardiovascular event in the etoricoxib group than in the naproxen group [35]. One pooled analysis of randomized trials reported no difference in the incidence of cardiovascular thrombotic events between patients treated with etoricoxib and patients receiving non-naproxen NSAIDs (RR 0.83, 95% CI 0.26 to 2.64 for etoricoxib 60 mg/d versus the combined non-naproxen traditional NSAID group, ibuprofen 2400 mg/d and diclofenac 150 mg/d). A trend toward more events with etoricoxib versus naproxen 1000 mg/d was observed (RR 1.70, 95% CI 0.91 to 3.18) [36]. Given all these data concerning coxibs, it is likely that trials of newer drugs in the category will be extended in order to provide additional evidence of cardiovascular safety. Why such a Cardiovascular Risk for Coxibs? Several mechanisms have been proposed to explain the cardiovascular risk attributed to coxibs. The absence of COX-2 in platelets may explain that thromboxane A2 (TXA2) generation is unaffected as it is mediated by COX-1, whereas prostacyclin production is inhibited. Thus, a coxibinduced imbalance between prostacyclin and TXA2 production leading to an inhibition of COX-2 generated prostacyclin without an opposing reduction in TXA2 has been one hypothesis to explain the coxib-related cardiovascular risk. Prostacyclin inhibits platelet aggregation, prevents the proliferation of vascular smooth-muscle cells in vitro, and causes vasodilation. On the contrary, TXA2 causes platelet aggregation, vasoconstriction, and vascular proliferation. This imbalance could create a continued TXA2 production and increased risk of thrombosis (Fig. (1)). Moreover, coxibs and other nsnsaids can increase blood pressure by a variety of mechanisms. COX-1 and COX-2 are co-localized in the macula densa. Coxibs in the elderly cause sodium retention [37] and may result in edema formation, reduction in glomerular filtration rate [37], and exacerbation of hypertension [38]. In the MEDAL program [30], heart failure risk alone, and not thrombotic cardiovascular endpoints, was significantly associated with change in blood pressure. On the other hand, baseline systolic blood pressure was associated with significantly higher risk of cardiovascular events (p<0.001). Thus, baseline blood pressure, but not change in blood pressure, was significantly associated with risk of thrombotic cardiovascular events. The cardiovascular risk of etoricoxib and diclofenac did not appear in this study to be related to the blood pressure-elevating effects of these agents [39]. Other factors including differences in dose, timing, duration, interactions with background drugs, different patient baseline cardiovascular risks, and the targeted disease may explain the discrepancies of s as regards cardiovascular events related to coxibs. For instance, the time from the beginning of exposure to the cardiovascular events differs notably between trials: they may be short-time events such as in the study of Gislason et al. [24], or long-time events: in the APPROVe trial, rofecoxib was associated with an increased risk of cardiovascular events beginning after 18 months of treatment [16], and in the APC trial, excess cardiovascular events occurred after three years of treatment [21]. Of interest, Schjerning et al. recently demonstrated in patients with prior MI that the duration of coxib or nsnsaid treatment was not of importance in the cardiovascular risk of these drugs [40]. Indeed, they found that the risks of death and recurrent MI were independent of the duration of NSAID treatment, and that the risk with some NSAIDs became apparent immediately (diclofenac) or soon after (rofecoxib and ibuprofen) treatment onset. These results challenge the current recommendations by the AHA of low-dose and short-term use of NSAIDs as being safe as we have seen above in patients with established cardiovascular disease [3], because this study demonstrates that even short-term NSAID treatment may be associated with increased cardiovascular risk in patients with prior MI [41]. Moreover, this study highlights the cardiovascular risk of nsnsaids, demonstrating that coxibs are not the only NSAID of concern as regards cardiovascular issues. nsnsaids Since cardiovascular safety of coxibs has been challenged, similar concerns have emerged regarding traditional nsnsaids. Many results of s regarding coxibs also involve nsnsaids and are discussed above (Table 1). The risk posed by nsnsaids in patients with high cardiovascular risk, such as elderly patients with prior MI, is rarely studied, as these patients are frequently excluded from trials. Nevertheless, the prematurely terminated Alzheimer s Disease Anti-inflammatory Prevention (ADAPT) study assessed celecoxib 200 mg twice daily and naproxen sodium 220 mg twice daily versus placebo in decreasing the risk of developing Alzheimer s disease [42]. Outcome measures were death, non-fatal MI, stroke, congestive heart failure, transient ischemic attack and anti-hypertensive treatment. Results were 5.54% in the celecoxib arm, 8.25% in the naproxen arm and 5.68% in the placebo arm, which conferred an HR of 1.10 (95% CI 0.67 to 1.79) for celecoxib and 1.63 (95% CI 1.04 to 2.55) for naproxen. These data suggested that increased risk of thrombotic events was observed not only with coxibs but also with nsnsaids in general, as the ADAPT trial found an increased risk for cardiovascular events in patients given naproxen but not in those given celecoxib. However, this trial may have had several limitations [3]. The AHA guidelines recommend naproxen as the preferred NSAID choice for patients who do not tolerate or are not sufficiently pain-relieved by nonpharmacological approaches, and/or by acetaminophen or opioids [3]. With regard to MI, flurbiprofen has demonstrated a potent anti-platelet activity, and has been indicated in secondary prevention after coronary reperfusion [43], especially in patients who had transient contraindications to

8 62 Anti-Inflammatory & Anti-Allergy Agents in Medicinal Chemistry, 2013, Vol. 12, No. 1 Roubille et al. aspirin such as planned surgery. Nevertheless, at present, given the current advances in coronary reperfusion after MI, flurbiprofen is considered to provide an insufficient medical benefit and is no longer used. A Danish study observed the risk of recurrent MI and death related to the use of selective COX-2 inhibitors and nsnsaids in patients after first-time acute MI between 1995 and 2002 and demonstrated a dose-related excess mortality associated with the use of nsnsaids in patients with prior MI [24]. Of the 58,462 patients included, 21,093 (36.1%) claimed at least one prescription of one of the NSAIDs. 9,773 re-hospitalizations (18.6%) for MI and 16,561 deaths (28.3%) were reported during the observation period. There was significantly increased dose-related risk of death associated with any use of either of the nsnsaids (ibuprofen: HR 1.50, 95% CI 1.36 to 1.67, and diclofenac: HR 2.40, 95% CI 2.09 to 2.80) or the selective COX-2 inhibitors (rofecoxib: HR 2.80, 95% CI 2.41 to 3.25, and celecoxib: HR 2.57, 95% CI 2.15 to 3.08) (p<0.001) in these post-mi patients. Higher death rates were associated with exposure to any of the NSAIDs compared with nonexposure. There was a trend toward increased risk of readmission for MI related to the use of both the selective COX-2 inhibitors and nsnsaids, but without a clear doserelated effect. Trelle et al. demonstrated that ibuprofen was associated with the highest risk of stroke (rate ratio 3.36, 95% credibility interval 1.00 to 11.6), followed by diclofenac (2.86, 1.09 to 8.36), which was also associated with the highest risk of cardiovascular death (3.98, 1.48 to 12.7) [2]. Kearney et al. also reported that high doses of ibuprofen (800 mg three times daily) and high doses of diclofenac (75 mg twice daily) were both associated with an increased risk of vascular events but that the risks of a high dose of naproxen (500 mg twice daily) were smaller. Nevertheless, it is not clear whether the cardiovascular effects of lower and typical daily doses of these drugs would differ from those identified in this meta-analysis [44]. GLUCOCORTICOIDS Glucocorticoids play a pivotal role in the management of many inflammatory and rheumatic conditions such as RA. It has yet to be established whether the use of exogenous GCs is associated with an increased risk of cardiovascular morbidity. At present, few of the common beliefs about the incidence, the prevalence and the impact of adverse effects of GCs are supported by clear scientific evidence (Table 2). GCs are known to induce hypertension [45], diabetes mellitus, weight gain and dyslipidemia [46]. Therefore, it would seem logical to consider GCs a deleterious therapy promoting cardiovascular events, yet the evidence is not clear. The effects of GCs on the cardiovascular system, especially among patients suffering from inflammatory rheumatic conditions, have never been well documented, mainly because no large s have been conducted with sufficient statistical power and duration of exposure to detect differences in cardiovascular morbidity and mortality [47]. Randomization to GC treatment versus placebo in a longterm clinical trial powered to detect cardiovascular outcomes would be the most valid study design for this question. Moreover, although the literature supports the development of these potentially harmful effects with high dose GC therapy [48], it has been recognized that there is no firm evidence linking low dose GC therapy and cardiovascular diseases in the RA population [46]. Guidelines recommend assessing and treating where indicated the factors that could be enhanced by GCs before initiating GC therapy, although there is no evidence that this is effective (level of evidence IV, expert opinions) [49]. It is also recommended that during GC treatment, patients should be monitored for body weight, blood pressure, peripheral edema, cardiac insufficiency, serum lipids, and blood and/or urine glucose depending on the individual patient s risk, GC dose and duration (level of evidence IV, expert opinions). Limited data on GC exposure and cardiovascular diseases based on large administrative databases have been published. One case-control study compared more than 50,000 patients with cardiovascular diseases and the same number of matched controls without cardiovascular diseases, all receiving GC therapy. There was a significant association between GC treatment and cardiovascular or cerebrovascular events (adjusted odds ratio [OR] 1.25, 95% CI 1.21 to 1.29) [47, 50]. Current use of oral GCs was associated with an increased risk for heart failure (adjusted OR 2.66, 95% CI 2.46 to 2.87) [50]. Another cohort study of more than 150,000 people reported a rate of incident cardiovascular diseases of 17/1,000 patient-years in subjects not exposed to GCs and of 24/1,000 patient-years in subjects treated with GCs. After adjustment for known confounders, the excess cardiovascular risk was attributable to the group exposed to supra-physiological high doses of GCs (>7.5 mg per day), which represents 2% of the studied population, in whom the adjusted relative risk for a cardiovascular event was 2.56 (95% CI 2.18 to 2.99) [48]. Moreover, GCs were most strongly associated with occurrence of heart failure and least strongly with cerebrovascular disease events. Intriguingly, adjustment for the presence of components of metabolic syndrome such as hypertension, diabetes mellitus and dyslipidemia often does not eliminate the deleterious effect of GC therapy, suggesting that GC therapy may independently of classic cardiovascular risk factors promote cardiovascular events [48, 50, 51]. Anti-inflammatory GC therapy may induce atherogenesis in patients with inflammatory diseases. For instance, a doseresponse relationship between cumulative exposure to GCs and carotid artery atheroma in RA patients has been reported. This GC effect on atherosclerosis in RA patients may be independent of well-known complications of GC therapy such as diabetes, dyslipidemia or hypertension given that adjustment for cardiovascular risk factors did not suppress the association [52]. Several observational studies have reported harmful effects of GC therapy [48, 50, 51], especially with long-term high doses of GCs [48]. For instance, in RA, GC use has been associated with an increased incidence of MI (OR 1.5, 95% CI 1.2 to 2.0, p=0.002) as well as an expected increased risk of diabetes and hypertension [53]. The risk of mortality

9 Cardiovascular Adverse Effects of Anti-Inflammatory Drugs Anti-Inflammatory & Anti-Allergy Agents in Medicinal Chemistry, 2013, Vol. 12, No in an RA cohort (predominantly from cardiovascular diseases and infections) increases by 14% after 1 year, rising to 69% after 10 years, in patients treated with low dose oral GCs [54]. Similarly, other studies have reported that high cumulative doses of GCs are associated with accelerated vascular disease in patients with systemic lupus erythematosus (SLE) [55, 56]. In contrast, other studies have demonstrated no association between GC therapy and increased cardiovascular events, for example in children antenatally or neonatally treated with GCs and who had no cardiovascular problems or hypertension at school age compared to controls [57, 58] and in RA patients [59, 60]. Among 364 patients with polymyalgia rheumatica, Maradit Kremers et al. reported that patients who received GCs did not have a significantly higher risk for myocardial infarction, heart failure, peripheral vascular disease, or cerebrovascular disease (HR [95% CI] 0.58 [0.29 to 1.18], 0.85 [0.45 to 1.54], 0.58 [0.24 to 1.40], and 0.65 [0.33 to 1.26], respectively) compared with those who did not receive GCs. There was also no significant association between cumulative GC dose and any cardiovascular, peripheral vascular, or cerebrovascular event (p=0.39) [61]. Additionally, some studies have shown no association between long-term GC use and vascular disease in SLE [62, 63]. However, it is difficult to separate the effects of GCs themselves from those of the inflammatory disease which may itself be pro-atherogenic. Indeed, patients with chronic inflammatory diseases such as RA have a higher risk of cardiovascular diseases and related mortality compared to the general population [64]. This is primarily due to classic cardiovascular risk factors but also to systemic inflammation which is independently involved, causing accelerated atherosclerosis, myocardial infarction, cerebrovascular disease and heart failure (see review in this issue by Roubille et al. in Biologics and cardiovascular system). Theoretically, GCs could modulate the risk of cardiovascular diseases in inflammatory patients in two competing ways. The risk could increase due to deleterious effects of GCs on lipids, insulin resistance, hyperglycemia, hypertension, or obesity [65]. On the other hand, by inhibiting systemic inflammation, GCs could also decrease the risk of atherosclerosis and cardiovascular diseases. Thus, it seems difficult to clearly demonstrate the link between GCs and cardiovascular diseases given that the inflammatory disease itself may act as an important confounder because it also promotes cardiovascular risk. The most significant limitation of observational study designs for this issue is the potential for confounding by indication, that is, the possibility that the risk may be due to the underlying disease being treated rather than the use of GC therapy. Disease activity and severity in RA for instance may play an important role in evaluating cardiovascular risk in such inflammatory patients. Patients with more active and/or severe RA receive more GCs and are also more likely to develop cardiovascular diseases. When a flare occurs that requires a high-dose of GCs, and is followed by a cardiovascular complication such as MI, it is unknown whether the MI is due to high dose GC therapy or to the inflammatory flare itself, or even to the association. The same concern exists with long term GC therapy for chronic RA patients: is the long term GC exposure responsible for cardiovascular events or could the many flares of the inflammatory disease needing GC prescription be to blame [66]? Moreover, patient factors such as personal history of coronary heart disease or RA characteristics may influence the significance and the type of the effect, harmful or beneficial. For instance, rheumatoid factor (RF)-positive but not RF-negative patients were reported at increased risk for cardiovascular events following exposure to GCs [67]. There may also be different effects of GCs depending on the cardiovascular outcome considered, for example, MI versus heart failure [66]. Does low dose GC therapy expose the patient to the same cardiovascular risks as higher doses? Da Silva et al. [46] reported safety data of several s of low-dose GC therapy in RA ( 10 mg per day prednisolone equivalent) [68-72]. They reported no excess cardiovascular events, including no increase in blood pressure and no incidence of sudden deaths or arrhythmias, which seem to be more frequent in high dose GC pulses. They argue that the duration of trials may not be sufficient for the development of cardiovascular complications. Nevertheless, they conclude that in patients with RA receiving low dose GCs, the disease itself seems to be a greater cardiovascular risk factor in contrast with higher doses. Hence, the authors suggest that routine toxicity monitoring for patients receiving a low dose of GCs does not seem currently justifiable or cost effective based on existing evidence, except for patients with additional risk factors such as previous osteoporosis, hypertension or diabetes. Furthermore, one compared a low dose of GCs (7.5 mg per day of prednisolone) to placebo in 67 RA patients. After a mean of 5 years, low dose prednisolone did not influence endothelial function or atherosclerosis evaluated by intimamedia thickness and flow-mediated dilation of the brachial artery [73]. With regard to pathophysiology, GCs seem to have complex and often contradictory influences on cardiovascular disease. Cardiovascular effects of GCs may be mediated by GC receptors and mineralocorticoid receptors leading to both systemic and local direct effects on vessels and myocardium (as reviewed in this issue by Merlet et al.). Systemic actions of GCs are associated with increased cardiovascular risk via hypertension, hyperglycemia, dyslipidemia and obesity as described above and are likely to promote cardiovascular disease development. On the other hand, local effects on cells of the cardiovascular system could either promote cardiovascular events via promoting vasoconstriction, profibrotic effects and endothelial dysfunction, or inhibit lesion development via anti-proliferative, anti-inflammatory and anti-migratory effects [74]. The mechanisms of GC mediated hypertension appear to be mainly related to increased peripheral vascular resistance rather than to mineralocorticoid receptor mediated effects of increased sodium retention and plasma volume expansion [75]. This suggests that GCs can, not only contribute to the exacerbation or the development of a cluster of cardiovascular risk factors that are the cornerstone of metabolic syndrome, but also interact directly with the cells of the heart and vascular wall to alter their structure and function.

10 64 Anti-Inflammatory & Anti-Allergy Agents in Medicinal Chemistry, 2013, Vol. 12, No. 1 Roubille et al. CONCLUSION Anti-inflammatory drugs can exert deleterious effects on the cardiovascular system. NSAIDs, especially coxibs, have been demonstrated to increase cardiovascular risk and have generated many concerns leading to the reassessment of their benefit/risk ratio. On the other hand, GCs may induce cardiovascular events, but evidence seems to be less clear. Nevertheless, these drugs could counterbalance inflammation, which is per se deleterious. It should be kept in mind that these molecules are not neutral with regard to cardiovascular risk and the choice of drug for a specific patient should be made with care, especially for those at high cardiovascular risk. To some extent, the Holy Grail would remain to treat inflammatory disorders, using drugs without any side effects on the cardiovascular system, to fully obtain the cardiovascular benefit of such a strategy. Biologics could partly address this issue (see review in this issue by Roubille C et al. in Biologics and cardiovascular system) but new anti-inflammatory strategies are mandatory. CONFLICT OF INTEREST Dr Roubille has received research grants from Servier Institute, 53 rue Carnot, Suresnes Cedex, France. ACKNOWLEDGEMENTS The authors thank Virginia Wallis for assistance with the manuscript preparation. ABBREVIATIONS ADAPT = Alzheimer s Disease Anti-inflammatory Prevention Trial AHA = American Heart Association APC = Adenomatous Prevention with Celecoxib APPROVe = Adenomatous Polyp PRevention On Vioxx CABG = Coronary artery bypass graft CV = Cardiovascular CI = Confidence intervalle CLASS = Celecoxib Long-term Arthritis Safety Study COX = Cyclooxygenase EDGE = Etoricoxib versus Diclofenac sodium Gastrointestinal tolerability and Effectiveness GCs = Glucocorticoids HF = Heart failure HR = Hazard ratio = Meta-analysis MEDAL = Multinational Etoricoxib and Diclofenac Arthritis Long-term MI = Myocardial infarction NNH = Number needed to harm NSAID = Non-steroidal anti-inflammatory drugs NYHA = New York Heart Association OR = Odds ratio PGE2 = Prostaglandin E2 PGI2 = Prostacyclin Pre-SAP = Prevention of Spontaneous Adenomatous Polyps RA = Rheumatoid arthritis = Randomized controlled trials ROS = Reactive oxygen species RR = Relative risk RF = Rheumatoid factor SLE = Systemic lupus erythematosus STAT = Signal transducer and activator of transcription TARGET = Therapeutic Arthritis Research and Gastrointestinal Event Trial TXA2 = Thromboxane A2 VICTOR = Vioxx In Colorectal cancer Therapy: definition of Optimal Regimen VIGOR = Vioxx gastrointestinal outcomes research REFERENCES [1] FitzGerald, G.A.; Patrono, C. The coxibs, selective inhibitors of cyclooxygenase-2. N. Engl. J. Med. 2001, 345(6), [2] Trelle, S.; Reichenbach, S.; Wandel, S.; Hildebrand, P.; Tschannen, B.; Villiger, P. M.; Egger, M.; Juni, P. Cardiovascular safety of non-steroidal anti-inflammatory drugs: Network meta-analysis. Bmj, 2011, 342, c7086. [3] Antman, E.M.; Bennett, J.S.; Daugherty, A.; Furberg, C.; Roberts, H.; Taubert, K.A. Use of nonsteroidal antiinflammatory drugs: an update for clinicians: A scientific statement from the American Heart Association. Circulation, 2007, 115(12), [4] Hochberg, M.C.; Altman, R.D.; April, K.T.; Benkhalti, M.; Guyatt, G.; McGowan, J.; Towheed, T.; Welch, V.; Wells, G.; Tugwell, P.; American College of, R. American College of Rheumatology 2012 recommendations for the use of nonpharmacologic and pharmacologic therapies in osteoarthritis of the hand, hip, and knee. Arthritis Care Res., 2012, 64(4), [5] Rostom, A.; Muir, K.; Dube, C.; Jolicoeur, E.; Boucher, M.; Joyce, J.; Tugwell, P.; Wells, G.W. Gastrointestinal safety of cyclooxygenase-2 inhibitors: A Cochrane Collaboration systematic review. Clin. Gastroenterol. Hepatol., 2007, 5(7), , 828 e1-5; quiz 768. [6] Bombardier, C.; Laine, L.; Reicin, A.; Shapiro, D.; Burgos-Vargas, R.; Davis, B.; Day, R.; Ferraz, M.B.; Hawkey, C.J.; Hochberg, M.C.; Kvien, T.K.; Schnitzer, T.J. Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. VIGOR Study Group. N. Engl. J. Med., 2000, 343(21), , 2 p following [7] Weir, M.R.; Sperling, R.S.; Reicin, A.; Gertz, B.J. Selective COX-2 inhibition and cardiovascular effects: A review of the rofecoxib development program. Am. Heart. J., 2003, 146(4), [8] Konstam, M.A.; Weir, M.R.; Reicin, A.; Shapiro, D.; Sperling, R.S.; Barr, E.; Gertz, B.J. Cardiovascular thrombotic events in controlled, clinical trials of rofecoxib. Circulation, 2001, 104(19), [9] Solomon, D.H.; Schneeweiss, S.; Glynn, R.J.; Kiyota, Y.; Levin, R.; Mogun, H.; Avorn, J. Relationship between selective cyclooxygenase-2 inhibitors and acute myocardial infarction in older adults. Circulation, 2004, 109(17), [10] Kimmel, S.E.; Berlin, J.A.; Reilly, M.; Jaskowiak, J.; Kishel, L.; Chittams, J.; Strom, B.L. Patients exposed to rofecoxib and celecoxib have different odds of nonfatal myocardial infarction. Ann. Intern. Med., 2005, 142(3),

11 Cardiovascular Adverse Effects of Anti-Inflammatory Drugs Anti-Inflammatory & Anti-Allergy Agents in Medicinal Chemistry, 2013, Vol. 12, No [11] Levesque, L.E.; Brophy, J.M.; Zhang, B. The risk for myocardial infarction with cyclooxygenase-2 inhibitors: a population study of elderly adults. Ann. Intern. Med., 2005, 142 (7), [12] Graham, D.J.; Campen, D.; Hui, R.; Spence, M.; Cheetham, C.; Levy, G.; Shoor, S.; Ray, W.A. 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(9458), [13] Juni, P.; Nartey, L.; Reichenbach, S.; Sterchi, R.; Dieppe, P.A.; Egger, M. Risk of cardiovascular events and rofecoxib: Cumulative meta-analysis. Lancet, 2004, 364(9450), [14] Baron, J.A.; Sandler, R.S.; Bresalier, R.S.; Quan, H.; Riddell, R.; Lanas, A.; Bolognese, J.A.; Oxenius, B.; Horgan, K.; Loftus, S.; Morton, D.G.A randomized trial of rofecoxib for the chemoprevention of colorectal adenomas. Gastroenterology, 2006, 131(6), [15] Bresalier, R.S.; Sandler, R.S.; Quan, H.; Bolognese, J.A.; Oxenius, B.; Horgan, K.; Lines, C.; Riddell, R.; Morton, D.; Lanas, A.; Konstam, M.A.; Baron, J.A. Cardiovascular events associated with rofecoxib in a colorectal adenoma chemoprevention trial. N. Engl. J. Med., 2005, 352(11), [16] Baron, J.A.; Sandler, R.S.; Bresalier, R.S.; Lanas, A.; Morton, D. G.; Riddell, R.; Iverson, E.R.; Demets, D.L. Cardiovascular events associated with rofecoxib: final analysis of the APPROVe trial. Lancet, 2008, 372(9651), [17] Garner, S.E.; Fidan, D.D.; Frankish, R.R.; Judd, M.G.; Towheed, T.E.; Wells, G.; Tugwell, P. Rofecoxib for rheumatoid arthritis. Cochrane. Database Syst. Rev., 2005, (1), CD [18] Kerr, D.J.; Dunn, J. A.; Langman, M.J.; Smith, J.L.; Midgley, R. S.; Stanley, A.; Stokes, J.C.; Julier, P.; Iveson, C.; Duvvuri, R.; McConkey, C.C. Rofecoxib and cardiovascular adverse events in adjuvant treatment of colorectal cancer. N. Engl. J. Med., 2007, 357(4), [19] Zhang, J.; Ding, E.L.; Song, Y. Adverse effects of cyclooxygenase 2 inhibitors on renal and arrhythmia events: meta-analysis of randomized trials. JA, 2006, 296(13), [20] Bertagnolli, M.M.; Eagle, C.J.; Zauber, A.G.; Redston, M.; Solomon, S.D.; Kim, K.; Tang, J.; Rosenstein, R.B.; Wittes, J.; Corle, D.; Hess, T.M.; Woloj, G.M.; Boisserie, F.; Anderson, W.F.; Viner, J.L.; Bagheri, D.; Burn, J.; Chung, D C.; Dewar, T.; Foley, T.R.; Hoffman, N.; Macrae, F.; Pruitt, R.E.; Saltzman, J.R.; Salzberg, B.; Sylwestrowicz, T.; Gordon, G. B.; Hawk, E. T. Celecoxib for the prevention of sporadic colorectal adenomas. N. Engl. J. Med., 2006, 355 (9), [21] Solomon, S.D.; McMurray, J.J.; Pfeffer, M.A.; Wittes, J.; Fowler, R.; Finn, P.; Anderson, W.F.; Zauber, A.; Hawk, E.; Bertagnolli, M. Cardiovascular risk associated with celecoxib in a clinical trial for colorectal adenoma prevention. N. Engl. J. Med., 2005, 352(11), [22] Antman, E.M.; DeMets, D.; Loscalzo, J. Cyclooxygenase inhibition and cardiovascular risk. Circulation, 2005, 112(5), [23] Solomon, S.D.; Pfeffer, M. A.; McMurray, J.J.; Fowler, R.; Finn, P.; Levin, B.; Eagle, C.; Hawk, E.; Lechuga, M.; Zauber, A.G.; Bertagnolli, M.M.; Arber, N.; Wittes, J. Effect of celecoxib on cardiovascular events and blood pressure in two trials for the prevention of colorectal adenomas. Circulation, 2006, 114(10), [24] Gislason, G.H.; Jacobsen, S.; Rasmussen, J.N.; Rasmussen, S.; Buch, P.; Friberg, J.; Schramm, T.K.; Abildstrom, S.Z.; Kober, L.; Madsen, M.; Torp-Pedersen, C. Risk of death or reinfarction associated with the use of selective cyclooxygenase-2 inhibitors and nonselective nonsteroidal antiinflammatory drugs after acute myocardial infarction. Circulation, 2006, 113 (25), [25] Solomon, S.D.; Wittes, J.; Finn, P.V.; Fowler, R.; Viner, J.; Bertagnolli, M.M.; Arber, N.; Levin, B.; Meinert, C.L.; Martin, B.; Pater, J.L.; Goss, P.E.; Lance, P.; Obara, S.; Chew, E.Y.; Kim, J.; Arndt, G.; Hawk, E. Cardiovascular risk of celecoxib in 6 randomized placebo-controlled trials: The cross trial safety analysis. Circulation, 2008, 117(16), [26] Silverstein, F. E.; Faich, G.; Goldstein, J.L.; Simon, L.S.; Pincus, T.; Whelton, A.; Makuch, R.; Eisen, G.; Agrawal, N. M.; Stenson, W.F.; Burr, A.M.; Zhao, W. W.; Kent, J.D.; Lefkowith, J. B.; Verburg, K. M.; Geis, G. S. Gastrointestinal toxicity with celecoxib vs nonsteroidal antiinflammatory drugs for osteoarthritis and rheumatoid arthritis: the CLASS study: A randomized controlled trial. Celecoxib Long-term Arthritis Safety Study. JA, 2000, 284(10), [27] Farkouh, M.E.; Kirshner, H.; Harrington, R.A.; Ruland, S.; Verheugt, F.W.; Schnitzer, T.J.; Burmester, G.R.; Mysler, E.; Hochberg, M.C.; Doherty, M.; Ehrsam, E.; Gitton, X.; Krammer, G.; Mellein, B.; Gimona, A.; Matchaba, P.; Hawkey, C. J.; Chesebro, J.H. 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 (9435), [28] Ott, E.; Nussmeier, N.A.; Duke, P.C.; Feneck, R.O.; Alston, R.P.; Snabes, M.C.; Hubbard, R.C.; Hsu, P.H.; Saidman, L.J.; Mangano, D.T. Efficacy and safety of the cyclooxygenase 2 inhibitors parecoxib and valdecoxib in patients undergoing coronary artery bypass surgery. J. Thorac. Cardiovasc. Surg., 2003, 125(6), [29] Nussmeier, N.A.; Whelton, A.A.; Brown, M.T.; Langford, R.M.; Hoeft, A.; Parlow, J.L.; Boyce, S.W.; Verburg, K.M. Complications of the COX-2 inhibitors parecoxib and valdecoxib after cardiac surgery. N. Engl. J. Med., 2005, 352(11), [30] Cannon, C.P.; Curtis, S.P.; FitzGerald, G.A.; Krum, H.; Kaur, A.; Bolognese, J. A.; Reicin, A. S.; Bombardier, C.; Weinblatt, M. E.; van der Heijde, D.; Erdmann, E.; Laine, L. Cardiovascular outcomes with etoricoxib and diclofenac in patients with osteoarthritis and rheumatoid arthritis in the Multinational Etoricoxib and Diclofenac Arthritis Longterm (MEDAL) programme: a randomised comparison. Lancet, 2006, 368(9549), [31] Combe, B.; Swergold, G.; McLay, J.; McCarthy, T.; Zerbini, C.; Emery, P.; Connors, L.; Kaur, A.; Curtis, S.; Laine, L.; Cannon, C. P. Cardiovascular safety and gastrointestinal tolerability of etoricoxib vs diclofenac in a randomized controlled clinical trial (The MEDAL study). Rheumatology (Oxford), 2009, 48(4), [32] Baraf, H. S.; Fuentealba, C.; Greenwald, M.; Brzezicki, J.; O'Brien, K.; Soffer, B.; Polis, A.; Bird, S.; Kaur, A.; Curtis, S.P. Gastrointestinal side effects of etoricoxib in patients with osteoarthritis: results of the Etoricoxib versus Diclofenac Sodium Gastrointestinal Tolerability and Effectiveness (EDGE) trial. J. Rheumatol., 2007, 34(2), [33] Krueger, K.; Lino, L.; Dore, R.; Radominski, S.; Zhang, Y.; Kaur, A.; Simpson, R.; Curtis, S. Gastrointestinal tolerability of etoricoxib in rheumatoid arthritis patients: results of the etoricoxib vs diclofenac sodium gastrointestinal tolerability and effectiveness trial (EDGE-II). Ann. Rheum. Dis., 2008, 67(3), [34] Trelle, S.; Reichenbach, S.; Wandel, S.; Hildebrand, P.; Tschannen, B.; Villiger, P.M.; Egger, M.; Juni, P. Cardiovascular safety of nonsteroidal anti-inflammatory drugs: network meta-analysis. BMJ, 342, c7086. [35] Reginster, J.Y.; Malmstrom, K.; Mehta, A.; Bergman, G.; Ko, A. T.; Curtis, S.P.; Reicin, A.S. Evaluation of the efficacy and safety of etoricoxib compared with naproxen in two, 138-week randomised studies of patients with osteoarthritis. Ann. Rheum. Dis., 2007, 66(7), [36] Curtis, S.P.; Ko, A.T.; Bolognese, J.A.; Cavanaugh, P.F.; Reicin, A. S. Pooled analysis of thrombotic cardiovascular events in clinical trials of the COX-2 selective Inhibitor etoricoxib. Curr. Med. Res. Opin., 2006, 22(12), [37] Rossat, J.; Maillard, M.; Nussberger, J.; Brunner, H.R.; Burnier, M. Renal effects of selective cyclooxygenase-2 inhibition in normotensive salt-depleted subjects. Clin. Pharmacol. Ther., 1999, 66(1), [38] Aw, T.J.; Haas, S.J.; Liew, D.; Krum, H. Meta-analysis of cyclooxygenase-2 inhibitors and their effects on blood pressure. Arch. Intern. Med., 2005, 165(5), [39] Krum, H.; Swergold, G.; Gammaitoni, A.; Peloso, P.M.; Smugar, S. S.; Curtis, S. P.; Brater, D.C.; Wang, H.; Kaur, A.; Laine, L.;