Non-Steroidal Anti-Inflammatory Drugs and Increased Risk of Sudden Cardiac Death

In: Sudden Cardiac Death ISBN: 978-1-62618-786-3 Editor: Pasquale Santangeli © 2013 Nova Science Publishers, Inc.

No part of this digital document may be reproduced, stored in a retrieval system or transmitted commercially in any form or by any means. The publisher has taken reasonable care in the preparation of this digital document, but makes no expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of information contained herein. This digital document is sold with the clear understanding that the publisher is not engaged in rendering legal, medical or any other professional services.

Chapter 7

Non-Steroidal Anti-Inflammatory Drugs and Increased Risk of Sudden Cardiac Death

Soyun M. Hwang,1 Jennifer E. Gilda,1 Ziyou Cui1 and Aldrin V. Gomes1,2, 1Department of Neurobiology, Physiology, and Behavior, and 2Department of Physiology and Membrane Biology, University of California, Davis, CA, US

Introduction

Non-steroidal anti-inflammatory drugs, or NSAIDs, are commonly prescribed, and some are available as over-the-counter (OTC) medications to alleviate pain, inflammation, and fever associated with a diverse array of medical conditions (Table 1, Figure 1) [1]. Each year in the United States alone, 70 million prescriptions are written for patients suffering from painful conditions such as rheumatoid arthritis, osteoarthritis, migraine, ankylosing spondylitis, Reiter’s syndrome, acute gout, menstrual pain, and others [2].

 Corresponding author: Dr. Aldrin V. Gomes, 176 Briggs Hall, One Shields Avenue, Department of Neurobiology, Physiology, and Behavior, University of California, Davis, Davis, CA 95616, US. E-mail: [email protected]. Phone: 530-752-3207; Fax: 530-752-5582. 124 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al.

Including the use of OTC NSAIDs, approximately 30 billion dosages are consumed annually [3].

Table 1. Types of NSAIDs, their common trade names, and typical clinical uses

Trade names, NSAID Clinical Uses Reference Other names Non-prescription The prototypical analgesic for mild to Anacin®, moderate pain, inflammation, fever, Ascriptin®, and minor aches. Also inhibits platelet Aspirin Bayer®, [164] aggregation and is used in the Bufferin®, prevention of arterial and venous Excedrin® thrombosis. Most commonly used for rheumatism Motrin®, and arthritis. Also used for headaches, Advil®, Ibuprofen muscle aches, arthritis, toothaches, [165] Nuprin®, backaches, common cold, and Medipren® menstrual cramps. Used to relieve pain, tenderness, swelling, and stiffness associated with osteoarthritis, rheumatoid arthritis, juvenile arthritis, and ankylosing Naproxen sodium Aleve® [166] spondylitis. Can also be used for more general pain from headaches, backaches, fever, cold, toothache, and menstrual pain. Prescription Prescribed for the treatment of osteoarthritis, rheumatoid arthritis, acute pain, painful menstruation and Celecoxib Celebrex® menstrual symptoms, and to reduce [167] numbers of colon and rectum polyps in patients with familial adenomatous polyposis. For relieving pain, tenderness, swelling, and stiffness caused by Diclofenac Voltaren® osteoarthritis, rheumatoid arthritis, and [167, 168] sodium ankylosing spondylitis. Also used to treat painful menstrual periods. Used for pain, tenderness, swelling, Diflunisal Dolobid® and stiffness caused by osteoarthritis [169, 170] and rheumatoid arthritis. Effective for treating pain associated with osteoarthritis, rheumatoid Etodolac Lodine® [171] arthritis, and ankylosing spondylitis. Also alleviates postoperative pain.

Non-Steroidal Anti-Inflammatory Drugs ... 125

Trade names, NSAID Clinical Uses Reference Other names Effective for osteoarthritis, rheumatoid arthritis, ankylosing spondylitis, and in the alleviation of postoperative pain. It Fenoprofen Nalfon® [172] is pharmacologically similar to aspirin, but causes less gastrointestinal bleeding. Helps reduce bone reabsorption in periodontal disease by inhibiting carbonic anhydrase. Used to relieve Flurbiprofen Ansaid® [173] pain, tenderness, swelling, and stiffness caused by osteoarthritis and rheumatoid arthritis. Relieves moderate to severe pain, tenderness, swelling, and stiffness Indomethacin Indocin® caused by osteoarthritis, rheumatoid [174] arthritis, ankylosing spondylitis, bursitis, and tendinitis. An ibuprofen-type anti-inflammatory Orudis®, analgesic and antipyretic used in the Ketoprofen Oruvail®, [162, 175] treatment of rheumatoid arthritis and Ketoflam® osteoarthritis. Used to relieve pain, tenderness, Meclofenamate Meclomen®, swelling, and stiffness caused by [176] sodium Ponstel® osteoarthritis and rheumatoid arthritis. Used to relieve the inflammation, swelling, stiffness, and joint pain associated with osteoarthritis and Oxaprozin Daypro® [177] rheumatoid arthritis. Also prescribed for juvenile rheumatoid arthritis in children six years of age and older. A butyl-diphenyl-pyrazolidinedione that has anti-inflammatory, antipyretic, and analgesic activities. It has been Phenylbutazone Butazolidin® [178] used for ankylosing spondylitis, rheumatoid arthritis, and reactive arthritis. Used to treat osteoarthritis, rheumatoid arthritis, ankylosing spondylitis, and Sulindac Clinoril® shoulder pain caused by bursitis and [179] tendinitis. Also used to relieve gouty arthritis. Relieves pain, tenderness, swelling, and stiffness caused by osteoarthritis, Tolmetin Tolectin® rheumatoid arthritis, and juvenile [180] rheumatoid arthritis in children two years of age and older. Commonly used analgesic, antipyretic, Trilisate and anti-inflammatory drug. Salicylate Salicylate [180] Disalcid® is the major drug for the treatment of rheumatic diseases. 126 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al.

While the majority of NSAIDs commonly ingested have minor toxic effects, with increasing general consumption, more attention is being directed toward their risk profile for gastrointestinal, renal, and cardiovascular side effects [4]. Therefore, an important question regarding NSAID usage is which NSAID has the lowest risks relative to its intended benefits [5].

NSAIDs Mechanism of Action

All NSAIDs work through a similar mechanism involving the inhibition of cyclooxygenases (COX) to exert their anti-inflammatory, analgesic and anti- pyretic effects [6]. Cyclooxygenases are a group of intracellular enzymes that catalyze the conversion of arachidonic acid into prostanoids, which are biologically active lipids that modulate the body’s inflammatory response [6, 7]. Prostanoids are divided into three major groups: prostaglandins (PGs), prostacyclins (PGI2s), and thromboxanes (TXs), among which prostaglandins are thought to be the main promoter of inflammation and clotting factors [6, 8]. Prostaglandins are also responsible for stimulating secretions of mucin, bicarbonate, and phospholipids by gastric epithelial cells to promote proper gastric function [9]. Two COX isoforms have been investigated closely over the past three decades: COX-1 and COX-2 [10, 11]. COX-1, commonly referred to as the “housekeeping” enzyme, is the most abundant form and is involved in many pathophysiological processes, such as hemostatic integrity, gastric cytoprotection, platelet aggregation, and kidney function [12]. Activation of COX-1 produces prostacyclin, which prevents thrombosis by inhibiting platelet activation [13] and is found to be upregulated in patients with polypectomy-induced ulcers for gastric healing and protective effects [14].

a b

Figure 1. (Continued). Non-Steroidal Anti-Inflammatory Drugs ... 127

c Hwang et al.

Figure 1. Structures of NSAIDs. A. Non-prescription NSAIDs. B. COX-2 Specific NSAIDs. C. Prescription NSAIDs.

In contrast, COX-2 is usually undetectable in most tissues unless induced by pro-inflammatory cytokines or growth factors to produce prostaglandins. However, COX-2 is constitutively expressed in the mammalian brain and has diverse functions in the central nervous system [15, 16]. Overexpression of COX-2 has been linked to neuro-inflammation, the inflammatory demyelinating effects of multiple sclerosis, Alzheimer’s disease, dementia, and other brain injuries or trauma [17, 18]. Therefore, COX-2 is thought to be largely responsible for the production of messenger molecules involved in fever, pain, and inflammation [19]. COX-3, a splice variant of the COX-1 gene, is the most recently discovered isoform [20]. It is expressed in the central nervous system and may be inhibited by acetaminophen, but its involvement in the regulation of prostanoids to maintain hemostatic, gastric, and renal homeostasis is still being investigated. However, COX-3 seems to have little relevance in humans as this isoform does not demonstrate COX activity [21]. 128 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al.

Types of NSAIDs

Since each COX isoform functions in different pathophysiological responses, understanding the specific inhibitory effects of NSAIDs on COX may explain the related benefits and toxicity of NSAID usage [22, 23]. There are two main types of NSAIDs: non-selective and COX-2 selective (also known as coxibs) [24]. Non-selective NSAIDs, which inhibit both COX-1 and COX-2 isoforms, are also called the traditional NSAIDs, because the use of non-selective NSAIDs predates the development of coxibs [25]. Examples of commonly used NSAIDs are summarized in Table 1. Common examples of non-selective NSAIDs include aspirin, ibuprofen, naproxen, sulindac, diclofenac, and ketoprofen [25]. NSAIDs may also be classified into two groups based upon their intracellular half-lives after oral administration [26]. NSAIDs with short half-lives (<6hrs, such as ibuprofen, diclofenac and ketoprofen) need to be taken more often than NSAIDs with long half-lives (>6hrs, such as naproxen and celecoxib). Naproxen is usually taken 1-2 times a day while ibuprofen is taken every 6-8 hours to maintain its effect. Because COX-1 is a constitutive enzyme for gastric cytoprotection and COX-2 is a prostaglandin synthase for pro-inflammatory responses, traditional NSAIDs can successfully reduce inflammation by inhibiting COX-2, but they can also cause gastric or renal irritation, toxicity, and other complications by disrupting COX-1 action in gastrointestinal (GI) homeostasis [27]. For instance, upper GI complications like dyspepsia occur in 15-60% of persons taking non-selective NSAIDs, which is twice the rate seen in nonusers [28]. Also, risks for upper gastrointestinal bleeding or perforation are 3.8 times higher for patients taking NSAIDs [29]. Such severe gastric toxicity of NSAIDs was reflected by a case study of peptic ulcer surgical therapy, which revealed that 54% of its patients were NSAID users [30]. Furthermore, prostaglandins in the kidneys dilate the vasculature to facilitate renal vascular flow and proper organ perfusion and to mediate the effects of diuretics on sodium and water retention [25]. Inhibition of prostaglandin production by traditional NSAIDs may cause vasoconstriction, thereby significantly reducing renal blood flow and glomerular filtration rate and increasing the risk of renal failure [31]. However, while fluid retention occurs in almost all NSAID users, severe renal conditions such as hyperkalemia and nephritic syndrome occur in only about 1-5% of patients taking NSAIDs [32]. To reduce toxic gastrointestinal and renal side effects while maintaining the benefits of NSAIDs, selective NSAIDs, or COX-2 inhibitors (coxibs), were designed to inhibit only COX-2, the main pro-inflammatory regulator [33]. Non-Steroidal Anti-Inflammatory Drugs ... 129

In various clinical trials and meta-analyses, coxibs significantly reduced gastrointestinal risks in comparison to traditional NSAIDs and had 50% fewer incidences of perforations, ulcers, and gastrointestinal bleeds [34, 35]. Celecoxib, a selective NSAID commonly used for osteoarthritis and rheumatoid arthritis, was also shown to decrease the risk of complicated ulcers by 20-30% [36]. However, the cardiovascular risks of coxibs seem to outweigh the gastrointestinal benefits; the number of hospitalizations over a 3- year period for acute myocardial infarction in patients taking rofecoxib (a coxib) was increased by 30% [37]. Rofecoxib (Vioxx) was approved by the Food and Drug Administration (FDA) in 1999 and was voluntarily withdrawn by Merck in 2004 after more than 88,000 cases of serious heart disease were reported in patients taking this drug. Rofecoxib was one of the most widely used drugs to be withdrawn from the pharmaceutical market [38]. The deleterious cardiovascular effects of rofecoxib were evident in a study in which it was used for Alzheimer’s disease treatment, but these cardiovascular events were under-reported at the time. Recent analyses of the three studies conducted with rofecoxib as a potential treatment of Alzheimer’s disease showed that rofecoxib increased the risk of cardiovascular related death by >3.5-fold [38]. Interestingly, this recent analysis suggest that the increased risk of cardiovascular-related death was statistically significant in June 2001, which meant that rofecoxib was on the market 40 months longer than it should have been [38]. The severe side effects of rofecoxib may be due to the fact that aside from promoting inflammation, COX-2 has other functions in mediating vasodilation and platelet activation for cardiovascular homeostasis [33]. In a knockout mouse model, selective inhibition of COX-2 causes an imbalance in hemostatic regulators, such as thromboxane A2 and prostacyclin I2, and increases the risks of thrombosis and platelet aggregation [39]. Also, while the NSAID-associated side effects of edema and sodium retention in the kidneys may have negligible effects on healthy individuals, they may adversely influence the hemostatic balance for patients with a prior history of heart failure [2]. Such disruption of systemic regulatory functions may explain why NSAIDs have a high cardiovascular risk profile, especially for patients with pre-existing heart conditions [2]. However, due to a general lack of wide-scale clinical trials investigating the cardiotoxicity of NSAIDs, there are still many molecular mechanisms to be investigated in order to explain the wide array of cardiovascular risks associated with NSAIDs [19]. 130 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al.

NSAIDs and Cardiotoxicity

Over the last decade, COX-2 selective NSAIDs have become increasingly associated with increased risk of cardiovascular diseases (Table 2). The first research trial that attracted the media and researchers’ attention to the cardiovascular risk profile of NSAID usage was the Vioxx Gastrointestinal Outcomes Research, or VIGOR, in 2001 [34].

Table 2. Effects of Different NSAIDs on Cardiovascular Safety

NSAID Subjects and Conditions Cardiovascular Safety Profile Reference 1,031 patients with cardiac Reduced risk of cardiac events risk treated with 100mg (myocardial infarction, stroke, and [181] aspirin vs. 4,495 nonusers cardiac death) by 10%. Reduced risk of cardiovascular death Aspirin Six randomized studies of by 21%, risk of nonfatal myocardial 9,853 patients, 50% treated infarction by 26%, risk of stroke by [182] with 50-325mg of aspirin vs. 25%. Increased risk of severe 50% with placebo bleeding by 2.2-fold. 1,228 patients treated with Increased risk of serious coronary <1600mg ibuprofen vs. 1,188 heart disease by a factor of 1.34 and [49] with 1000mg naproxen cardiovascular death by 1.25. 389 new users of 200mg Increased risk of atrial fibrillation by [183] Ibuprofen ibuprofen vs. 2,660 nonusers 40%. Meta-analysis of 31 trials Increased risk of myocardial involving 116,429 patients infarction by a factor of 1.61, stroke [50] with 117,218 years of follow- by 3.36, and cardiac death by 2.39. up 719 Alzheimer’s patients Increased risk of cardiac events treated with 220mg naproxen (myocardial infarction, stroke, cardiac [51] vs. 1,083 treated with placebo death) by a factor of 1.63. Meta-analysis of 31 trials Increased the risk of stroke by a factor involving 116,429 patients of 1.76, but did not affect the risks of Naproxen [50] with 117,218 years of follow- cardiac death or myocardial up infarction. Meta-analysis of 24 case- Naproxen was risk-neutral at all doses control studies involving for cardiovascular conditions [48] 24,468 patients compared to other NSAIDs. 804 patients treated with Increased risk of serious coronary Diclofenac 150mg diclofenac vs. 1,188 heart disease by a factor of 1.44 and [49] with 1000mg naproxen cardiovascular death by 1.52. 172,362 patients treated with 100mg or higher dosage Two-fold increased risk of [184] diclofenac vs. 568,525 cardiovascular death. Diclofenac nonusers 292 new users of 200mg Increased the risk of atrial fibrillation [183] diclofenac vs. 1,647 nonusers by 70%. Non-Steroidal Anti-Inflammatory Drugs ... 131

Subjects and Conditions Cardiovascular Safety Profile Reference 11,459 patients treated with No significant increase in Etoricoxib 60 or 90mg etoricoxib vs. cardiovascular or thrombotic events [185] 11,212 with 150mg diclofenac (risk factor range of 0.9-1.08). 3,987 patients treated with 400mg celecoxib vs. 3,981 No serious cardiovascular event in [52] with 2,400 mg ibuprofen or comparison to non-selective NSAIDs. 75mg diclofenac Increased the risk of cardiovascular 726 Alzheimer’s patients death, myocardial infarction, stroke, Celecoxib treated with 220mg naproxen [51] and other cardiac events by a factor of vs. 1,083 treated with placebo 1.10. 685 patients treated with Increased the risk of cardiovascular 400mg celecoxib vs. 671 with death by a factor of 2.3 at 400mg and [53] 800mg celecoxib vs. 679 with by 3.4 at 800mg administration. placebo Reduced gastrointestinal events by 4,047 patients treated with 50% compared to naproxen, but 50mg rofecoxib (Vioxx) vs. [34, 40] increased thrombotic cardiovascular 4,029 with 1,000mg naproxen events by a factor of 2.37. 1% of patients taking either rofecoxib 122 patients treated with or naproxen developed stroke or Rofecoxib 25mg rofecoxib vs. 118 with transient ischemic attack. 1% of [186] 220mg naproxen vs. 111 with rofecoxib users developed myocardial placebo infarction. 1,287 patients treated with Increased the risk of cardiac events 25mg rofecoxib vs. 1,299 with (myocardial infarction, sudden death) [41] placebo by a factor of 2.80 and stroke by 2.32. Risk of cardiac events (myocardial CABG patients: 555 treated infarction, cardiac arrest, and stroke) with parecoxib and valdecoxib increased by a factor of 3.7 for Valdecoxib (Bextra) vs. 556 with [43] parecoxib and valdecoxib group and valdecoxib only vs. 560 with by 2.0 for valdecoxib only group placebo compared to placebo group.

In a randomized controlled trial involving more than 8,000 patients, one group of patients received 50mg of rofecoxib (Vioxx) once a day and another group of patients received 500mg of naproxen twice a day [34]. Initially, this trial was intended to investigate the gastrointestinal benefits of a coxib in comparison to a regular NSAID; rofecoxib reduced gastrointestinal events by 50% in comparison to naproxen [40]. However, rofecoxib doubled the risks of cardiac events, including sudden cardiac death, ischemic injury, chest pain, and cardiac thrombosis [40]. However, in a systematic review analyzing the risks and benefits of selective COX-2 inhibitors, Chaiamnuay et al. argued that the seemingly adverse cardiovascular effects of rofecoxib in the VIGOR study could be due to possibility that they were compared to the cardioprotective effects of naproxen [19]. To address 132 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al. this concern, a randomized trial of 2,586 patients who were treated with either 25mg rofecoxib or a dose of placebo each day, was subsequently carried out [41]. This study showed that patients using rofecoxib for 18 months or more were 2.5 times more likely to undergo cardiac events, including myocardial infarction and sudden cardiac death [41]. These clinical results as well as others resulted in the withdrawal of rofecoxib from the worldwide market in 2004 [40]. Valdecoxib (Bextra), was another selective COX-2 inhibitor to be withdrawn from the US market in 2005 due to its cardiovascular toxicity [42]. In a study involving 1,600 patients who had undergone a coronary artery bypass graft surgery (CABG), three groups received either valdecoxib and its intravenous prodrug paracoxib, valdecoxib only, or placebo only [43]. The risks of cardiac arrest, stroke, and myocardial infarction were doubled for the valdecoxib only group and almost quadrupled for the valdecoxib and paracoxib group [43, 44]. However, an integrated analysis of the use of valdecoxib and paracoxib in non-cardiac disease patients later showed that neither coxib increased the risk of cardiovascular events [45]. In CABG patients, valdecoxib could have aggravated the stress-induced platelet aggregation already present from the pump-assisted extracorporeal circulation, thus leading to an increased risk of cardiac death [46]. Therefore, the administration of coxibs for certain patients remains controversial and the proper use of coxibs to balance their intended benefits and cardiovascular risks must be investigated further [47]. While it is unclear if many or all NSAIDs behave similarly to rofecoxib or valdecoxib, their withdrawal from the market urged the investigation of the widespread use of NSAIDs [5]. There have been several meta-analyses of NSAID usage that compiled years of clinical trials, and they seem to agree that naproxen has the least harmful effects on cardiovascular health when compared to other NSAIDs [48- 50]. In a report by Trelle et al., with the exception of naproxen, diclofenac, and etoricoxib, it was shown that all other NSAIDs increased the risk of myocardial infarction by a factor of 1.3-2.8 (Table 2) [50]. Naproxen did show a 1.76-fold increase in the risk of stroke, but in comparison to ibuprofen, celecoxib, and other non-selective and selective NSAIDs, naproxen exhibited an insignificant or much lower probability of risk for cardiovascular events [50]. Non-Steroidal Anti-Inflammatory Drugs ... 133

Ray et al. also reported that naproxen had insignificant, if not cardioprotective, effects in comparison to other NSAIDs [49]. However, in the Alzheimer's Disease Anti-Inflammatory Prevention Trial (ADAPT), naproxen treatment for early onset of dementia was terminated early because of an increase in cardiovascular and cerebrovascular events [51]. When compared to a placebo treatment, naproxen increased the risk of cardiac events such as myocardial infarction, stroke, and cardiac death by a factor of 1.63 [51]. Most randomized trials used different placebos or control groups to compare against the administration of NSAIDs, which makes it difficult to determine their validity and compare the cardiotoxicity ratios [50]. In the Celecoxib Long-term Arthritis Safety Study (CLASS), celecoxib’s cardiovascular effects were deemed insignificant in comparison to other NSAIDs such as ibuprofen and diclofenac [52]. The CLASS trial also reported that celecoxib had less upper GI irritation, hepatotoxicity, and renal toxicity. On the other hand, a randomized trial for Adenoma Prevention with Celecoxib (APC) treated the patients with a dose of placebo or 400mg of celecoxib, both administered twice daily for a 3-year period, revealed a significant increase in cardiovascular risk with celecoxib [53]. Celecoxib was associated with increased risk of sudden cardiac death, cardiovascular diseases, and nonfatal myocardial infarction. Other NSAIDs also demonstrated increased cardiotoxic effects in patients who had prior heart conditions [53]. However, at least two studies with celecoxib suggest that the cardiovascular risk of this NSAID is similar to other non-selective NSAIDs or placebo [54, 55]. When administered to patients with prior myocardial infarction, within weeks of treatment, diclofenac increased the risk of cardiac death by an average factor of 2.56, ibuprofen by 1.47, and celecoxib by 1.65 [56]. Overall, all NSAIDs significantly increased the risks of recurrent myocardial infarction and cardiac death for these patients by a risk factor range of 1.34 to 1.86 [56]. Aside from using various controls to compare NSAID usage against, choosing different dosages of NSAIDs also complicated the cross-analyses of many randomized trials [57]. In 2011, an American Heart Association study showed that ibuprofen, even at an average dosage of 1600mg per day, increased the risk of a recurrent heart attack by 50% in patients with previous record of myocardial infarction [56]. Around the same time, a systematic review of observational studies by McGettigan et al. reported that a low-dosage use of ibuprofen at 1200mg per 134 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al. day or less had no significant effect on cardiovascular risk, even in the high risk patient population [48]. In this systematic review, it was found that ibuprofen at a low dosage was comparable to naproxen in its low risk profile for cardiotoxicity. Furthermore, in a handful of other randomized trials previously mentioned, such as APC and CLASS, NSAIDs were used at higher dosages than normal since they were used to treat specific clinical symptoms [48, 58]. Consequently, the cardiotoxicity profile of NSAIDs used in specialized trials may not be applicable to the general public using NSAIDs purchased over the counter [48]. Such disparity in systematic reports indicates the need for more clinical trials to investigate the effects of NSAIDs at varying dosages, comparison to different controls, patient population, and other variables to develop a more suitable risk profile for the general use of NSAIDs.

Molecular Mechanisms Involved in NSAID-Induced Cardiotoxicity

As previously described, both selective and non-selective NSAIDs have adverse cardiovascular effects. The investigations of the mechanisms through which NSAIDs cause adverse cardiovascular events demonstrate that NSAIDs have multiple targets and mechanisms by which they cause adverse cardiovascular events, including COX-dependent and COX-independent mechanisms (Figure 2) [59, 60]. The main mechanisms known to be involved in NSAID-induced cardiotoxicity are discussed below.

Shift Towards a Prothrombotic State

Metabolism of arachidonic acid produces multiple metabolites which have both pro- and antithrombotic effects. COX inhibition by NSAIDs creates an imbalance of arachidonic acid metabolites, shifting the balance toward a prothrombotic state in which the blood has a higher tendency to coagulate, increasing the risk of adverse cardiovascular events [61].

Non-Steroidal Anti-Inflammatory Drugs ... 135

Decreased Prostacyclin Production It was originally believed that the main mechanism by which COX-2 inhibitors caused adverse cardiovascular events was the selective inhibition of COX-2 [60].

Hwang et al.

Figure 2. Schematic representation of the molecular mechanisms involved in NSAID- induced cardiotoxicity. Abbreviations: NSAIDs, non-steroidal anti-inflammatory drugs; 20-HETE, 20-hydroxyeicosatetraenoic acid; COX, cyclooxygenase; EGR-1, early growth response gene-1; NAG-1, NSAID-activated gene-1; PTEN, phosphatase and tensin homolog; p27, protein 27; IκB, inhibitor of NF-κB; NF-κB, nuclear factor kappa-light-chain-enhancer of activated B cells; Bax, Bcl-2-associated X protein; Bcl- 2, B-cell lymphoma 2; ROS, reactive oxygen species; MPT pores, mitochondrial permeability transition pores; Kv channels, voltage-gated potassium channels; ATP, adenosine triphosphate.

Selective inhibition of COX-2 leads to a decrease in the production of antithrombotic prostacyclin without a reduction in prothrombotic thromboxane levels, favoring a prothrombotic state which can result in clot formation and increased risk of adverse cardiovascular effects [60, 61].

Increased 20-HETE Levels There are multiple enzymes that process arachidonic acid to produce a variety of metabolites, called eicosanoids [62]. 20-hydroxyeicosatetraenoic 136 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al. acid (20-HETE) is synthesized from arachidonic acid by the enzyme cytochrome P450 [63]. In a study aimed at evaluating the mechanisms behind the adverse cardiovascular effects of the selective NSAID rofecoxib, it was found that levels of 20-HETE were increased greater than 120-fold in mice treated with rofecoxib, which correlated with decreased tail bleeding and blood clotting time and increased platelet aggregability [61]. These parameters serve as a possible readout for cardiac risk, since bleeding time correlates to platelet aggregability, which is related to myocardial infarction and stroke [61]. Additionally, increased platelet aggregability and decreased bleeding time has been observed in patients with myocardial infarction [61]. The increase in 20-HETE levels was also seen for the selective COX-1 inhibitor SC-560 and the non-selective COX inhibitor indomethacin, which both differ structurally from rofecoxib [61]. One possible explanation of this effect is that inhibition of the cyclooxygenases leaves more arachidonic acid available for conversion into other metabolites. However, if this was the case, one would expect an increase in other arachidonic acid metabolites, a phenomenon which is not observed [61]. An alternative explanation is that rofecoxib induces expression of other enzymes involved in 20-HETE synthesis; however, increased expression of these enzymes was not observed [61]. The most likely explanation for the increase in 20-HETE levels is that inhibition of COX blocks the metabolism of 20-HETE, giving rise to increased 20-HETE levels, and therefore increased platelet aggregability and adverse cardiovascular outcomes [64]. Because COX inhibition is a general property of NSAIDs, and increased 20-HETE levels were seen for both selective and non-selective COX inhibitors, an increase in 20-HETE levels is likely to be a general effect of NSAIDs.

Induction of Apoptosis

Increased cardiomyocyte apoptosis is associated with adverse cardiovascular outcomes [65, 66]. Multiple groups have reported elevated levels of apoptosis in patients suffering from dilated cardiomyopathy, a condition which precedes heart failure and is characterized by enlargement of the heart chambers and contractile dysfunction [67-69]. Others have reported increased rates of apoptosis in hearts that have undergone myocardial Non-Steroidal Anti-Inflammatory Drugs ... 137 infarction [65, 66, 70]. Additionally, induction of apoptosis in transgenic mice is sufficient to cause lethal cardiomyopathy, while inhibition of cardiomyocyte death prevents heart failure [71]. Together, these reports suggest that the increases in the rate of cardiomyocyte apoptosis have a significant effect on heart function and identify cardiomyocyte apoptosis as an essential step in the progression to heart failure. Several reports have demonstrated that various NSAIDs induce apoptosis [72-79].

Inhibition of Prostaglandin Signaling Prostaglandins, the metabolites of arachidonic acid, bind to prostaglandin receptors called EP receptors, which apart from regulating processes involved in pain and inflammation, also control apoptosis and cell survival [80]. Blocking prostaglandin signaling by preventing prostaglandin formation may play a role in NSAID-induced apoptosis. Binding of prostaglandin E2 (PGE2) to the PGE2 receptors EP2 and EP4 activates cAMP-dependent protein kinase (PKA), which leads to phosphorylation of glycogen synthase kinase-3 (GSK- 3) [81]. This alters APC/β-catenin/TCF signaling, which under normal conditions negatively regulates apoptosis and promotes cell proliferation [80].

PGE2 can also activate the epidermal growth factor (EGF) receptor, inhibiting phosphoinositide 3-kinase (PI3K)/protein kinase B (PKB)/Akt signaling that would normally activate peroxisome proliferator-activated receptor γ (PPARγ). PPARγ is a ligand-binding transcription factor that regulates cell proliferation, differentiation, and apoptosis, and alterations in its activity could play a role in NSAID-induced apoptosis [82-85]. Interestingly, NSAIDs have also been shown to act as direct PPARγ ligands at high concentrations [86]. The effects of NSAIDs on PPARγ activity suggest another mechanism by which NSAIDs induce apoptosis.

Inhibition of Akt Signaling Akt signaling is one of the most important pathways in promoting cell survival and suppressing apoptosis [87]. As discussed in the previous section,

Akt signaling is controlled by PGE2 and EGF receptor signaling, which is activated by PGE2. Consequently, inhibition of Akt signaling via inhibition of COX-mediated prostaglandin formation may play a role in inducing apoptosis [80]. In addition to this COX-dependent mechanism, a COX-2-independent mechanism of altering Akt signaling has been demonstrated [79]. Kulp et al. used DMC, a celecoxib analogue lacking COX-2 inhibitory activity, to investigate non-COX-2 targets of celecoxib in a human prostate cancer cell line. Celecoxib and DMC blocked Akt activation by inhibiting 138 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al. phosphoinositide-dependent kinase-1 (PDK-1), a kinase which phosphorylates Akt, resulting in its activation. Inhibition of Akt signaling by celecoxib and DMC caused G1 cell cycle arrest and apoptosis. These data indicate that Akt inhibition can occur independently of COX-2 inhibitory action [79]. Additionally, the NSAIDs NS-398 and indomethacin prevent Akt activation by activating phosphatase and tensin homolog (PTEN) [87].

Increased Expression of EGR-1 Early growth response gene 1 (EGR-1) is a zinc finger transcription factor which is induced by stress, injury, and growth factors [88]. NSAIDs increase expression of EGR-1, which has pro-apoptotic functions through its control of p53 [89], NSAID-activated gene-1 (NAG-1) [90], and PTEN [91], whose roles in NSAID-induced apoptosis are further discussed in subsequent sections [80].

Activation of PTEN The main functions of the tumor suppressor phosphatase and tensin homolog (PTEN) involve its lipid phosphatase activity but it could also function in a phosphatase-independent manner [92]. PTEN regulates several intracellular pathways including pathways involving the extracellular signal- regulated kinase (ERK)1/2, phosphoinositide 3-kinase (PI3K)/protein kinase B (PKB/AKT), janus kinase (JAK)/signal transducers and activators of transcription (STAT) [92]. The NSAIDs NS-398 and indomethacin were shown to activate PTEN in human colon cancer cells [87]. Virolle et al. showed that exposing cells to ultraviolet light induced the transcription factor EGR-1, which caused an increase in the levels of PTEN messenger RNA and protein and induced apoptosis. In cells lacking EGR-1, PTEN was not induced, and cells were resistant to apoptosis, demonstrating that EGR-1 directly activates PTEN, which has pro-apoptotic effects [91]. PTEN dephosphorylates phosphatidylinositol (PtdIns), which prevents activation of PDK-1, and thus prevents activation of Akt, which may contribute to induction of apoptosis [87].

Upregulation of NAG-1/GDF 15 NSAIDs have been shown to induce the expression of NSAID-activated gene-1 (NAG-1)/growth differentiation factor 15 (GDF 15), resulting in induction of apoptosis [93, 94]. The traditional NSAIDs sulindac sulfide and indomethacin, and the selective COX-1 inhibitor SC-560 induce the expression of the EGR-1 transcription factor, which leads to up-regulation of Non-Steroidal Anti-Inflammatory Drugs ... 139

NAG-1. NAG-1 is a target of EGR-1, and EGR-1 may be required for NSAID- induced up-regulation in NAG-1. It is likely that NSAIDs induce NAG-1 expression through upregulation of EGR-1 [90, 95].

Altered p53 Signaling NSAIDs have been shown to increase levels of the protein p53, a transcription factor that regulates anti-proliferative processes and apoptosis [96]. NSAID-induced apoptosis correlates to the p53 gene status: in a gastric cancer cell line, cells expressing p53 are more susceptible to aspirin- and indomethacin-induced apoptosis [77, 97, 98]. This is associated with upregulation of Bax [77, 97] and Bak [77] and downregulation of Bcl-2 [97]. NSAIDs also appear to regulate p53-dependent apoptosis through increased expression 15-lipoxygenase-1 (15-LOX-1), an enzyme that metabolizes arachidonic acid, producing mainly the metabolite 15-HETE [62]. Shureiqi et al. demonstrated that in colon cancer cells, the NSAIDs sulindac sulfone and NS-398 induce expression of 15-LOX-1, which precedes apoptosis, and overexpression of 15-LOX-1 caused phosphorylation of p53, which increased expression of p53 target genes and resulted in growth inhibitory effects. This was observed in the absence of COX-2, and therefore is likely independent of COX inhibition [80, 99]. Additionally, a p53 binding site has been identified in the promoter of NAG-1, a protein with pro-apoptotic effects [90].

Altered NF-κB Signaling The transcription factor nuclear factor kappa B (NF-κB) regulates genes which promote cell survival, including FLIP, TRAFs and c-IAPs, which prevent the activation of caspases involved in apoptosis [100, 101]. Activation of NF-κB promotes cell proliferation and protects against apoptosis, while inhibition of NF-κB decreases anti-apoptotic signaling in the cell, causing a shift in favor of apoptosis. Reports on the effects of NSAIDs on NF-κB signaling are conflicting. A study conducted by Smartt et al. reported that the COX-2 inhibitor NS- 398, which induced apoptosis in colorectal carcinoma cells, increased binding of NF-κB to DNA, but did not increase transcriptional activity. Inhibition of NF-κB binding to DNA did not further enhance apoptosis induction, indicating that NF-κB was not transcriptionally active [102]. However, others have reported that NSAIDs inhibit NF-κB activation. Aspirin has been shown to inhibit NF-κB activation by inhibiting the proteasome [96]. NF-κB activation is regulated by IκB, the inhibitor of NF-κB. Degradation of IκB by the proteasome causes translocation of NF-κB into the nucleus where it initiates 140 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al. transcription of target genes [96]. Proteasome inhibition results in decreased degradation of IκB, and therefore decreased NF-κB activity. It is likely that some NSAIDs induce apoptosis partially through altered NF-κB signaling, though these mechanisms remain to be clarified and may differ between NSAIDs.

Increased p27Kip1 Expression The protein encoded by p27Kip1 controls cell cycle progression by preventing activation of CDK-cyclin complexes and is involved in regulating apoptosis. Huang et al. demonstrated that in human lung cancer cells, the NSAIDs NS-398 and indomethacin increase levels of p27 by downregulating expression of proteasome subunits, leading to decreased degradation of p27, which has an antiproliferative effect [103].

Upregulation of Bax and Bak and Downregulation of Bcl-2 B-cell lymphoma 2 (Bcl-2) and Bcl-2-associated X protein (Bax) play opposing roles in signaling apoptosis; Bax promotes cell death, while Bcl-2 promotes cell survival. Bcl-2 can heterodimerize with multiple proteins, which can have pro- or anti-apoptotic effects, and evidence indicates that the Bax and Bcl-2 proteins form a dimer which is important in controlling apoptosis. The primary function of Bcl-2 and Bax is to control the integrity of the mitochondrial membrane to influence the release of proteins from the mitochondria which can have pro- or anti-apoptotic effects [65]. Bcl-2 homologous antagonist killer (Bak) is another member of the Bcl-2 gene family that is involved in initiating apoptosis [104, 105]. NSAIDs have been shown to upregulate Bax and Bak and downregulate Bcl-2. The protein Bax is thought to be controlled by p53, presenting a possible explanation for its upregulation; however NSAIDs have been shown to induce expression of Bax in p53 mutants, indicating that increased expression induced by NSAIDs is independent of p53 [97]. Zhou et al. reported that aspirin and indomethacin increased levels of Bax and Bak, but did not affect Bcl-2 levels [77]. Dikshit et al. demonstrated that aspirin induces apoptosis in mouse Neuro 2a cells through inhibition of the proteasome. Apoptosis was induced due to various proteasome substrates failing to be degraded, including Bax and p53 [96]. Proteasome inhibition could partially explain the increased levels of Bax which contribute to NSAID-induced apoptosis.

Non-Steroidal Anti-Inflammatory Drugs ... 141

Downregulation of Specifity Proteins NSAIDs have been shown to affect the activity and degradation of Specifity proteins (Sps), which are transcription factors that control expression of housekeeping genes and other genes, including manganese superoxide dismutase (MnSOD). MnSOD is an antioxidant which is important for decreasing oxidative stress caused by reactive oxygen species. Significantly, overexpression of MnSOD has been shown to have myocardial protective effects; endogenous overexpression of MnSOD in rats led to an increased myocardial tolerance to ischemia-reperfusion [106]. Abdelrahim et al., demonstrated that the COX-2 inhibitors celecoxib, nimesulfide, and NS-398 enhanced the degradation of Sp1 and Sp4. [107]. This increased degradation was accompanied by increased levels of ubiquitination and was blocked by the proteasome inhibitor gliotoxin, suggesting that these COX-2 inhibitors increased degradation in a proteasome-dependent manner. In addition, celecoxib has been shown to decrease phosphorylation of Sp1. Decreased phosphorylation and increased degradation of Sp proteins are important for the NSAID-induced downregulation of vascular endothelial growth factor (VEGF) [104] [80], a protein that has anti-apoptotic effects and promotes angiogenesis [108-110]. The NSAIDs mentioned here likely induce apoptosis partially through their effects on the levels and activity of Specificity proteins.

Generation of Reactive Oxygen Species Induction of apoptosis by various drugs is accompanied by the generation of reactive oxygen species (ROS), indicating that oxidative stress caused by ROS generation may play a role in inducing apoptosis [111, 112]. Many different NSAIDs have been shown to generate ROS. Kusuhara et al. examined the role of reactive oxygen species in NSAID-induced apoptosis in cultured rat gastric cells and found that multiple NSAIDs, including indomethacin and diclofenac, generate ROS and induce apoptosis [113]. Treatment with antioxidants suppressed apoptotic DNA fragmentation induced by indomethacin and flubiprofen, demonstrating an important link between NSAID-induced apoptosis and the generation of ROS [113]. It has also been shown that diclofenac affects mitochondrial permeability transition, which results in the generation of ROS [114]. Generation of ROS increases levels of oxidative stress, which causes mitochondrial permeability transition (MPT) pores to open. The opening of the MPT pores allows the release of cytochrome c from mitochondria, activating caspase-9 and caspase-3, inducing apoptosis via the intrinsic/mitochondrial pathway [65].

142 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al.

Proteasome Inhibition Proteasome inhibition has many implications in the heart. Dikshit et al. demonstrated that aspirin inhibited the proteasome in mouse Neuro 2A cells, and induced apoptosis [96]. As previously mentioned, proteasome inhibition led to decreased degradation of several proteasome substrates, including IκB, p53, Bax, and p27, and induced apoptosis. Huang et al. showed that NSAIDs decreased expression of the proteasome catalytic subunits β5, β1i (LMP2), and β5i (LMP7), which led to increased levels of p27, and had an anti-proliferative effect [103]. Additionally, ubiquitin-proteasome system (UPS) dysfunction plays a role in cardiac proteinopathy, a cardiac disease in which proteins form harmful aggregates in the cell, disrupting cellular homeostasis and playing a role in progression to heart failure [115]. The proteasome is largely responsible for degrading damaged proteins such as oxidized proteins and preventing harmful protein aggregates from forming in the cell. As mentioned in the previous section, NSAIDs have been shown to generate reactive oxygen species (ROS), which can cause oxidative stress and damage proteins. NSAIDs may have multiple effects on the proteasome including indirect inhibition of the proteasome (possibly by oxidizing some of the proteasomal subunits) as well as increased levels of oxidized proteins, which may overload the proteasome. Although the cause of NSAID-induced proteasome dysfunction is not known, inhibition of the proteasome contributes to proteasome functional insufficiency and consequently cardiac proteinopathy [115]. High levels of proteasome inhibition result in cell death [116] and proteasome inhibitors are currently used in cancer treatments [117]. Interestingly, an important side effect of these anti-cancer proteasome inhibitors is increased cardiotoxicity [118]. This may be due to differences in the susceptibility of proteasomes from different tissues to proteasome inhibitors. Purified proteasomes from murine cardiac tissue were more susceptible to proteasome inhibitors than proteasomes from murine liver tissue [119]. It is possible that cardiac proteasomes are more susceptible to inhibition by NSAIDs than proteasomes in other tissues.

Inadequate Supply of ATP

As previously described, NSAID-induced mitochondrial dysfunction plays a role in apoptosis. Additionally, mitochondrial dysfunction caused by NSAIDs affects ATP production. The heart requires a constant supply of ATP, Non-Steroidal Anti-Inflammatory Drugs ... 143 and mitochondrial dysfunction leading to an inadequate supply of ATP causes cardiac contractile dysfunction which can lead to heart failure if severe [120, 121]. Mitochondria account for more than a third of the myocardial mass of the adult heart, highlighting the importance of mitochondria in supplying the heart with ATP [122].

Mitochondrial Dysfunction Leading to Defective Oxidative Phosphorylation Many NSAIDs have been shown to induce opening of mitochondrial permeability transition (MPT) pores in vitro, which induces membrane depolarization. In addition to the role MPT pores play in apoptosis, prolonged opening of MPT pores leads to mitochondrial dysfunction and defective oxidative phosphorylation, resulting in insufficient levels of ATP [122]. Additionally, various NSAIDs have also been shown to inhibit mitochondrial complex I, which may play a role in NSAID-induced mitochondrial dysfunction [120].

Altered Cardiac Electrical Activity

Cardiac electrical activity depends on the generation of action potentials in cardiomyocytes, which is based on the opening and closing of ion channels. Ion channel dysfunction has been implicated in inherited and acquired heart diseases, indicating the importance of ion channels in normal heart rhythm [123]. Mutations in genes encoding cardiac ion channels are also associated with cardiac arrhythmias and sudden cardiac death [124]. NSAIDs have been shown to inhibit multiple different ion channels via direct and indirect mechanisms, offering another explanation for the mechanism of NSAIDs in causing heart failure.

Inhibition of Voltage-Gated Potassium Channels Potassium channels play an important role in heart function. Heart failure is associated with altered potassium currents, in particular Ca2+-independent + + transient outward K (Ito) and inward rectifier K current (IKI). + Ito currents are due to voltage-gated K (Kv) channels [125]. The selective NSAID celecoxib has been shown to inhibit rat, Drosophila, and human voltage-activated delayed rectifier K+ (Kv2) channels and lead to arrhythmia in Drosophila hearts and rat heart cells. 144 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al.

The fact that Drosophila lack cyclooxygenase enzymes suggests that the mechanism of Kv2 inhibition is independent of the COX inhibitory action of celecoxib [59]. Additionally, treatment with aspirin failed to reproduce these results, suggesting that Kv2 inhibition is not a general property of NSAIDs. In another study, celecoxib was shown to inhibit Kv1.5, Kv4.3, and Kv7.1 channels while rofecoxib were shown to inhibit Kv4.3 channels. Celecoxib prolonged the duration of the action potential in mouse cardiomyocytes and had the opposite effect in guinea pig cardiomyocytes, again demonstrating that Kv channel inhibition has significant effects on the heart [126]. It is likely that inhibition of Kv2 channels contributes to the cardiotoxicity of some NSAIDs.

Inhibition of Voltage-Gated Calcium and Sodium Channels Celecoxib, but not rofecoxib or diclofenac, was found to dramatically suppress L-type voltage-sensitive calcium currents in A7r5 rat aortic smooth muscle cells and freshly isolated rat mesenteric artery myocytes [127]. Activation of L-type Ca2+ currents is correlated with heart contractions, and data suggest that differences in L-type voltage-dependent Ca2+ channel expression and activity may play a role in heart failure [128]. A celecoxib analog, 2,5-dimethyl-celecoxib, which is unable to inhibit COX-2, mimicked celecoxib in its enhancement of vascular KCNQ5 currents and suppression of L-type calcium currents, suggesting that these effects are independent of its COX-2 inhibitory actions [127]. These effects of celecoxib, rofecoxib, and diclofenac may partially explain the differential risk of cardiovascular events in patients taking different drugs of this class [127, 129]. Additionally, sodium channels are highly important for normal cardiac electrical activity, and changes in sodium channel function have been associated with myocardial ischemia, ventricular arrhythmias, and heart failure [130]. At low micromolar concentrations celecoxib suppressed both tetrodotoxin-sensitive and tetrodotoxin-resistant Na+ channel currents in a dose- and frequency-dependent manner in rat dorsal root ganglion neurons [131]. Celecoxib seems to bind to both inactivated and resting Na+ channels.

Molecular Mechanisms Involved in NSAID-Induced Sudden Cardiac Death

Sudden cardiac death (SCD) as the name suggests, is a sudden, unexpected natural death due to loss of cardiac function, which occurs within Non-Steroidal Anti-Inflammatory Drugs ... 145 one hour of the onset of symptoms [132]. SCD is the leading cause of natural deaths in adults in the US, accounting for over 450,000 adult deaths (>35 years old) each year [132]. Men are at a greater risk of SCD in comparison to women [133]. A significant proportion of all cardiac deaths are classified as SCD. Sudden cardiac death occurs when the electrical system of the heart becomes irregular and causes arrhythmia. While many patients are likely to have anatomical and/or functional cardiac abnormalities that predispose them to acquire ventricular arrhythmias, only a small percentage actually develop SCD [134]. As shown in Figures 2 and 3, several pathways are likely involved in the increased risk of NSAID-induced SCD. The most direct method by which NSAIDs are suggested to cause cardiac death is by altering cardiac electrical activity and inducing arrhythmias [135-137]. Arrhythmia has been associated with increased mortality due to increased risk of cardiac arrest, stroke, heart failure, sudden arrhythmic death syndrome, and sudden cardiac death (Figure 3). The most common life-threatening arrhythmia is ventricular fibrillation, the uncoordinated contraction of the ventricles [135]. Ventricular fibrillation prevents the heart chambers from pumping blood throughout the body and if left untreated can result in loss of consciousness, as blood flow to the brain is decreased, and subsequent death can occur within minutes. The importance of ventricular arrhythmia in SCD prompted the United States Food and Drug Administration (FDA) to approve Microvolt T-wave alternans (MTWA) measurements as a non-invasive predictor of the risk of potentially fatal ventricular arrhythmia. MTWA refers to an alternating pattern (beat-to-beat variability) in the amplitude or shape of the T-wave portion of an electrocardiogram. MTWA is able to detect small irregular waveforms which routine electrocardiogram (ECG) does not detect. However, the predictive value of MTWA varies, depending on the type of patients investigated [138, 139]. Other likely possibilities for inducing sudden cardiac death include NSAID-induced contractile dysfunction, increased blood clotting, and significant increases in cardiac myocyte apoptosis. Studies suggest that defects in mitochondrial oxidative pathways can also contribute to SCD. Tissues such as kidney, cardiac, skeletal, or smooth muscle, which are highly dependent on oxygen, are more sensitive to defective mitochondrial oxidative phosphorylation. Several NSAIDs have been shown to inhibit or uncouple oxidative phosphorylation, resulting in reduced cell viability [54]. This reduction in oxidative phosphorylation results in significantly decreased ATP production. Some research suggests that the decreased oxidative 146 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al. phosphorylation caused by NSAIDs may be due to inadequate mitochondrial inner membrane potential for ATP synthesis by the F0F1-ATPase.

Hwang et al.

Figure 3. Possible alterations in cardiac molecular pathways that lead to sudden cardiac death.

Evidence also suggests that NSAIDs (such as diclofenac sodium and mefenamic acid) induce mitochondrial permeability transition (MPT) pores to open which leads to both reduction in oxidative phosphorylation (due to membrane depolarization) as well as initiation of apoptosis [55]. The consequences of an inadequate supply of ATP are numerous, but the most likely adverse effect related to increased risk of SCD is contractile dysfunction. Muscle contraction requires a significant amount of ATP for myosin to bring about interaction and movement relative to the actin thin filament [140, 141]. Patients with severe contractile dysfunction have been shown to be at high risk for SCD [142, 143]. Contractile dysfunction is likely to affect cardiomyocyte Ca2+ homeostasis, as observed in hearts containing mutant myofilament proteins associated with cardiomyopathies [144]. This Non-Steroidal Anti-Inflammatory Drugs ... 147 abnormal Ca2+ homeostasis can lead to ventricular arrhythmias. Impairment of mitochondrial function affecting mitochondrial membrane potential or the electron transport chain has been shown to affect cardiac Ca2+ cycling, leading to proarrhythmic Ca2+ alternans [145]. Most NSAIDs are highly lipophilic and are able to be readily transported through the cell membrane and become enriched within the mitochondrion [146]. Although most NSAIDs are acidic, even non-acidic NSAIDs such as nabumetone have been shown to affect the mitochondrial inner membrane potential and reduce ATP synthesis [147]. Interestingly, the COX-2 selective inhibitor celecoxib, which does not seem to affect oxidative phosphorylation, is still associated with an increased risk of SCD. However, coxibs seem to be associated with a lower risk than some other non-selective NSAIDs. Neither nabumetone or celecoxib cause in vitro uncoupling [148]. Even though some NSAIDs themselves do not affect oxidative phosphorylation, their metabolites sometimes do. Salicylate, the main metabolite of aspirin, is able to uncouple oxidative phosphorylation, unlike its parent compound aspirin [146]. The increased risk of SCD associated with celecoxib may be related to this compound’s ability to affect ion channels, ultimately affecting the electrical activity of the cardiomyocytes. The intrinsic apoptotic pathway initiated by the MPT pore opening is also likely to be detrimental for the cardiomyocyte. Caspases are proteolytic enzymes that carry out the ordered destruction of the cell without induction of inflammation or leakage of cellular components [149]. Adult cardiomyocytes are post-mitotic cells, and therefore show limited capability to respond to acute damage. In the normal myocardium, apoptosis occurs at an extremely low rate (0.001-0.002%), but under pathological conditions such as in the hearts of patients with end-stage dilated cardiomyopathy, apoptosis increases to 0.08– 0.25% [67, 68, 149]. This 40-250 fold increase in apoptosis is substantial compared to the absolute amount of cardiomyocyte apoptosis in diseased hearts (<0.3%), and is pathologically important as over-expression of caspase 8 in mice increased apoptosis only 11.5 fold (from approx. 0.002 to 0.023%) but resulted in dilated cardiomyopathy after two months and death within six months [71]. NSAIDs have also been shown to induce a prothrombotic state by inhibition of COX-2, which results in decreased levels of the antithrombotic prostacyclin without reducing prothrombotic thromboxane levels [60, 61]. Persons in a prothrombotic state are at an increased risk for blood clotting and ischemia. 148 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al.

A frequent cause of SCD for persons at risk is pulmonary embolism and acute thrombosis, which leads to the fatal ventricular arrhythmia [150]. Although severe dysfunction of any of these pathways (contractile dysfunction, etc.) is likely to be a main contributor to SCD, combinations of different types of cardiovascular dysfunction would also likely lead to SCD. Another potential contributor to NSAID-induced SCD is ROS. Several NSAIDs, including indomethacin and diclofenac, are known to generate ROS [111-113]. In a canine post-myocardial infarction (MI) model of SCD Belevych et al. found that redox modification of ryanodine receptors promotes generation of Ca2+ alternans by increasing the steepness of the Ca2+ release– load relationship [151]. Hence, it is possible that increases in ROS may affect the redox status of several important cardiovascular proteins which leads to arrhythmias and then SCD. The oldest and best known NSAID, aspirin, is atypical when compared to other NSAIDs since it has been shown to lower the risk of heart attacks and strokes due to its anticoagulant effect. All currently available evidence suggests that daily use of aspirin is beneficial in men ages 45 to 79 who are at risk of having a heart attack within the next ten years [152]. The benefits of aspirin in preventing heart attacks are only seen in older females (>55 years old) [152, 153]. Unlike other NSAIDs, aspirin does not seem to be associated with SCD. While many of the cases of SCD associated with NSAIDs are likely due to NSAIDs themselves, it is important to recognize that some of the incidents of SCD in patients taking NSAIDs may be related to undiagnosed genetic mutations in these patients. Mutations associated with several myofibrillar genes, including troponin T, troponin I, and myosin heavy chain, are associated with increased risk of SCD [141, 154]. Some of the SCDs related to NSAIDs were in persons who used NSAIDs with other drugs, making it difficult to discern the NSAID-independent effects of other drug(s). Another important consideration is the interaction of NSAIDs with other drugs. The use of aspirin with other NSAIDs can be a dangerous combination and has been shown to increase the risk of GI bleeding. Taking aspirin in combination with ibuprofen has been shown by one research group to interfere with aspirin’s beneficial effects while another group found that taking ibuprofen in combination with aspirin reduced the beneficial heart-protective effects of aspirin [155, 156]. Both investigations suggest that this combination of NSAIDs should be avoided. Use of NSAIDs together with quinolones has been shown to increase the risk of quinolones’ adverse central nervous system effects, such as inducing seizures [157].

Non-Steroidal Anti-Inflammatory Drugs ... 149

Conclusion

The evidence undoubtedly suggests that NSAID use is associated with an increased risk of cardiovascular problems and in some cases the risk of adverse cardiovascular events, including SCD, increases with long term exposure to NSAIDs. Taking NSAIDs for long periods should therefore be avoided if possible. Persons who have high blood pressure or kidney disease or who are at risk of heart attack or stroke should be careful when considering taking NSAIDs even for short periods. In such patients, the use of NSAIDs (with the exception of low-dose aspirin) can be associated with as much as a 10-fold or greater increase in heart failure [158], while the risk to the general population is a 2-fold increase. In heart failure patients, NSAIDs increase the mortality risk by approximately 1.2-1.3 fold for naproxen and ibuprofen, 1.7 for celecoxib and rofecoxib, and 2.1 for diclofenac [159]. Current clinical data suggests that patients at risk of cardiovascular problems that require NSAID treatment should use naproxen [142, 143, 160]. SCD is the leading cause of death in young athletes, and screening of athletes by ECG was found to increase the ability to identify athletes at risk [161]. It may be possible to use ECG to screen patients taking NSAIDs long-term for increased risk of SCD. More extensive and accurate information needs to be provided to doctors, and the number of prescriptions for NSAIDs should be reduced, as greater than 40% of prescriptions were found to be unnecessary [162]. Doctors should be encouraged to advise patients to take the lowest dose of NSAID necessary to reduce pain. Doctors should also encourage patients to take NSAIDs for as short a period of time as necessary. Significant clinical trials are needed to determine the effects of NSAIDs at varying dosage levels and lengths of treatment time on cardiovascular diseases. By doing so, we can determine the dosage and period of time at which NSAIDs become unsafe and the benefits of the NSAIDs are outweighed by the risk. Additionally, it is unclear if a significant proportion of SCDs associated with NSAIDs occur in patients who are already at risk of SCD due to genetic or physiological factors. Myocarditis, various valvular diseases (such as aortic stenosis), inherited arrhythmia syndromes, and hypertrophic and dilated cardiomyopathy are associated with an increased risk of SCD. The current guidelines to reduce NSAID-associated complications are mainly limited to prevention of cardiovascular and GI related problems and are underutilized [160]. The possible availability of generic versions of celecoxib within the next few years (the patent for celecoxib will expire in 2014) may result in an increased use of this COX-2 selective NSAID [163]. More clinical studies and 150 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al. evidence-based analyses which take into account gastrointestinal and cardiovascular risks at an individual level are needed so that informed decisions can be made regarding the use of NSAIDs.

Acknowledgments

The authors' work was supported by UC Davis Startup funds.

References

[1] Peterson, K., McDonagh, M., Thakurta, S., Dana, T., Roberts, C., Chou, R., et al. Drug Class Review: Nonsteroidal Antiinflammatory Drugs (NSAIDs): Final Update 4 Report. Portland (OR)2010. [2] Bleumink, G. S., Feenstra, J., Sturkenboom, M. C., Stricker, B. H. Nonsteroidal anti-inflammatory drugs and heart failure. Drugs. 2003;63: 525-34. [3] Green, G. A. Understanding NSAIDs: from aspirin to COX-2. Clinical cornerstone. 2001;3:50-60. [4] Tamblyn, R., Berkson, L., Dauphinee, W. D., Gayton, D., Grad, R., Huang, A., et al. Unnecessary prescribing of NSAIDs and the management of NSAID-related gastropathy in medical practice. Annals of internal medicine. 1997;127:429-38. [5] Psaty, B. M., Weiss, N. S. NSAID trials and the choice of comparators-- questions of public health importance. The New England journal of medicine. 2007;356:328-30. [6] Vane, J. R. Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs. Nature: New biology. 1971;231:232-5. [7] Mitchell, J. A., Akarasereenont, P., Thiemermann, C., Flower, R. J., Vane, J. R. Selectivity of nonsteroidal antiinflammatory drugs as inhibitors of constitutive and inducible cyclooxygenase. Proceedings of the National Academy of Sciences of the United States of America. 1993; 90:11693-7. [8] Weissmann, G. Prostaglandins as modulators rather than mediators of inflammation. Journal of lipid mediators. 1993;6:275-86. Non-Steroidal Anti-Inflammatory Drugs ... 151

[9] Redfern, J. S., Feldman, M. Role of endogenous prostaglandins in preventing gastrointestinal ulceration: induction of ulcers by antibodies to prostaglandins. Gastroenterology. 1989;96:596-605. [10] Copeland, R. A., Williams, J. M., Giannaras, J., Nurnberg, S., Covington, M., Pinto, D., et al. Mechanism of selective inhibition of the inducible isoform of prostaglandin G/H synthase. Proceedings of the National Academy of Sciences of the United States of America. 1994;91: 11202-6. [11] DeWitt, D. L., Meade, E. A., Smith, W. L. PGH synthase isoenzyme selectivity: the potential for safer nonsteroidal antiinflammatory drugs. The American journal of medicine. 1993;95:40S-4S. [12] Wang, D., DuBois, R. N. Pro-inflammatory prostaglandins and progression of colorectal cancer. Cancer letters. 2008;267:197-203. [13] Higgs, E. A., Higgs, G. A., Moncada, S., Vane, J. R. Prostacyclin (PGI2) inhibits the formation of platelet thrombi in arterioles and venules of the hamster cheek pouch. British journal of pharmacology. 1978;63:535-9. [14] Kobayashi, K., Arakawa, T., Nakamura, H., Chono, S., Yamada, H., Satoh, H., et al. Protective action of endogenous prostacyclin (PGI2) and prostaglandin E2 (PGE2) in endoscopic polypectomy-induced human ulcers. Gastroenterologia Japonica. 1982;17:430-3. [15] Strauss, K. I., Marini, A. M. Cyclooxygenase-2 inhibition protects cultured cerebellar granule neurons from glutamate-mediated cell death. Journal of neurotrauma. 2002;19:627-38. [16] Cao, C., Matsumura, K., Yamagata, K., Watanabe, Y. Endothelial cells of the rat brain vasculature express cyclooxygenase-2 mRNA in response to systemic interleukin-1 beta: a possible site of prostaglandin synthesis responsible for fever. Brain research. 1996;733:263-72. [17] Minghetti, L. Cyclooxygenase-2 (COX-2) in inflammatory and degenerative brain diseases. Journal of neuropathology and experimental neurology. 2004;63:901-10. [18] Strauss, K. I. Antiinflammatory and neuroprotective actions of COX2 inhibitors in the injured brain. Brain, behavior, and immunity. 2008;22: 285-98. [19] Chaiamnuay, S., Allison, J. J., Curtis, J. R. Risks versus benefits of cyclooxygenase-2-selective nonsteroidal antiinflammatory drugs. American journal of health-system pharmacy : AJHP : official journal of the American Society of Health-System Pharmacists. 2006;63:1837-51. [20] Schwab, J. M., Schluesener, H. J., Laufer, S. COX-3: just another COX or the solitary elusive target of paracetamol? Lancet. 2003;361:981-2. 152 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al.

[21] Chandrasekharan, N. V., Dai, H., Roos, K. L., Evanson, N. K., Tomsik, J., Elton, T. S., et al. COX-3, a cyclooxygenase-1 variant inhibited by acetaminophen and other analgesic/antipyretic drugs: cloning, structure, and expression. Proceedings of the National Academy of Sciences of the United States of America. 2002;99:13926-31. [22] Kim, S. Y., Solomon, D. H. Pharmacotherapy: comparative safety of nonsteroidal anti-inflammatory drugs. Nature reviews Cardiology. 2011; 8:193-5. [23] Solomon, D. H. Strategies for making analgesia safer: the role of comparative effectiveness research. Arthritis and rheumatism. 2010;62: 1568-70. [24] Cryer, B., Feldman, M. Cyclooxygenase-1 and cyclooxygenase-2 selectivity of widely used nonsteroidal anti-inflammatory drugs. The American journal of medicine. 1998;104:413-21. [25] Batlouni, M. [Nonsteroidal anti-inflammatory drugs: cardiovascular, cerebrovascular and renal effects]. Arquivos brasileiros de cardiologia. 2010;94:556-63. [26] Conaghan, P. G. A turbulent decade for NSAIDs: update on current concepts of classification, epidemiology, comparative efficacy, and toxicity. Rheumatol. Int. 2012;32:1491-502. [27] Vane, J. R. Introduction: mechanism of action of NSAIDs. British journal of rheumatology. 1996;35 Suppl. 1:1-3. [28] Gabriel, S. E., Jaakkimainen, L., Bombardier, C. Risk for serious gastrointestinal complications related to use of nonsteroidal anti- inflammatory drugs. A meta-analysis. Annals of internal medicine. 1991; 115:787-96. [29] Hernandez-Diaz, S., Rodriguez, L. A. Association between nonsteroidal anti-inflammatory drugs and upper gastrointestinal tract bleeding/ perforation: an overview of epidemiologic studies published in the 1990s. Archives of internal medicine. 2000;160:2093-9. [30] Sarosi, G. A., Jr., Jaiswal, K. R., Nwariaku, F. E., Asolati, M., Fleming, J. B., Anthony, T. Surgical therapy of peptic ulcers in the 21st century: more common than you think. American journal of surgery. 2005;190: 775-9. [31] Bennett, W. M., Henrich, W. L., Stoff, J. S. The renal effects of nonsteroidal anti-inflammatory drugs: summary and recommendations. American journal of kidney diseases : the official journal of the National Kidney Foundation. 1996;28:S56-62. Non-Steroidal Anti-Inflammatory Drugs ... 153

[32] Whelton, A., Hamilton, C. W. Nonsteroidal anti-inflammatory drugs: effects on kidney function. Journal of clinical pharmacology. 1991;31: 588-98. [33] FitzGerald, G. A., Patrono, C. The coxibs, selective inhibitors of cyclooxygenase-2. The New England journal of medicine. 2001;345: 433-42. [34] Bombardier, C., Laine, L., Reicin, A., Shapiro, D., Burgos-Vargas, R., Davis, B., et al. Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. VIGOR Study Group. The New England journal of medicine. 2000;343:1520-8, 2 p following 8. [35] Langman, M. J., Jensen, D. M., Watson, D. J., Harper, S. E., Zhao, P. L., Quan, H., et al. Adverse upper gastrointestinal effects of rofecoxib compared with NSAIDs. JAMA : the journal of the American Medical Association. 1999;282:1929-33. [36] Laine, L. The gastrointestinal effects of nonselective NSAIDs and COX- 2-selective inhibitors. Seminars in arthritis and rheumatism. 2002;32: 25-32. [37] Rahme, E., Nedjar, H. Risks and benefits of COX-2 inhibitors vs non- selective NSAIDs: does their cardiovascular risk exceed their gastrointestinal benefit? A retrospective cohort study. Rheumatology (Oxford). 2007;46:435-8. [38] Madigan, D., Sigelman, D. W., Mayer, J. W., Furberg, C. D., Avorn, J. Under-reporting of cardiovascular events in the rofecoxib Alzheimer disease studies. American heart journal. 2012;164:186-93. [39] Cheng, Y., Austin, S. C., Rocca, B., Koller, B. H., Coffman, T. M., Grosser, T., et al. Role of prostacyclin in the cardiovascular response to thromboxane A2. Science. 2002;296:539-41. [40] Juni, P., Nartey, L., Reichenbach, S., Sterchi, R., Dieppe, P. A., Egger, M. Risk of cardiovascular events and rofecoxib: cumulative meta- analysis. Lancet. 2004;364:2021-9. [41] Bresalier, R. S., Sandler, R. S., Quan, H., Bolognese, J. A., Oxenius, B., Horgan, K., et al. Cardiovascular events associated with rofecoxib in a colorectal adenoma chemoprevention trial. The New England journal of medicine. 2005;352:1092-102. [42] Sun, S. X., Lee, K. Y., Bertram, C. T., Goldstein, J. L. Withdrawal of COX-2 selective inhibitors rofecoxib and valdecoxib: impact on NSAID and gastroprotective drug prescribing and utilization. Current medical research and opinion. 2007;23:1859-66. 154 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al.

[43] Nussmeier, N. A., Whelton, A. A., Brown, M. T., Langford, R. M., Hoeft, A., Parlow, J. L., et al. Complications of the COX-2 inhibitors parecoxib and valdecoxib after cardiac surgery. The New England journal of medicine. 2005;352:1081-91. [44] Furberg, C. D., Psaty, B. M., FitzGerald, G. A. Parecoxib, valdecoxib, and cardiovascular risk. Circulation. 2005;111:249. [45] Schug, S. A., Joshi, G. P., Camu, F., Pan, S., Cheung, R. Cardiovascular safety of the cyclooxygenase-2 selective inhibitors parecoxib and valdecoxib in the postoperative setting: an analysis of integrated data. Anesthesia and analgesia. 2009;108:299-307. [46] Borgdorff, P., Tangelder, G. J., Paulus, W. J. Cyclooxygenase-2 inhibitors enhance shear stress-induced platelet aggregation. Journal of the American College of Cardiology. 2006;48:817-23. [47] Scheiman, J. M. Balancing risks and benefits of cyclooxygenase-2 selective nonsteroidal anti-inflammatory drugs. Gastroenterology clinics of North America. 2009;38:305-14. [48] McGettigan, P., Henry, D. Cardiovascular risk and inhibition of cyclooxygenase: a systematic review of the observational studies of selective and nonselective inhibitors of cyclooxygenase 2. JAMA : the journal of the American Medical Association. 2006;296:1633-44. [49] Ray, W. A., Varas-Lorenzo, C., Chung, C. P., Castellsague, J., Murray, K. T., Stein, C. M., et al. Cardiovascular risks of nonsteroidal antiinflammatory drugs in patients after hospitalization for serious coronary heart disease. Circulation Cardiovascular quality and outcomes. 2009;2:155-63. [50] Trelle, S., Reichenbach, S., Wandel, S., Hildebrand, P., Tschannen, B., Villiger, P. M., et al. Cardiovascular safety of non-steroidal anti- inflammatory drugs: network meta-analysis. BMJ. 2011;342:c7086. [51] ADAPT Research Group. Cardiovascular and cerebrovascular events in the randomized, controlled Alzheimer's Disease Anti-Inflammatory Prevention Trial (ADAPT). PLoS clinical trials. 2006;1:e33. [52] Silverstein, F. E., Faich, G., Goldstein, J. L., Simon, L. S., Pincus, T., Whelton, A., et al. Gastrointestinal toxicity with celecoxib vs nonsteroidal anti-inflammatory drugs for osteoarthritis and rheumatoid arthritis: the CLASS study: A randomized controlled trial. Celecoxib Long-term Arthritis Safety Study. JAMA : the journal of the American Medical Association. 2000;284:1247-55. Non-Steroidal Anti-Inflammatory Drugs ... 155

[53] Bertagnolli, M. M., Eagle, C. J., Zauber, A. G., Redston, M., Solomon, S. D., Kim, K., et al. Celecoxib for the prevention of sporadic colorectal adenomas. The New England journal of medicine. 2006;355:873-84. [54] Famaey, J. P., Whitehouse, M. W. About some common biochemical and pharmacological properties of bile salts and non steroidal anti- inflammatory drugs. Archives internationales de physiologie et de biochimie. 1978;86:577-91. [55] Uyemura, S. A., Santos, A. C., Mingatto, F. E., Jordani, M. C., Curti, C. Diclofenac sodium and mefenamic acid: potent inducers of the membrane permeability transition in renal cortex mitochondria. Archives of biochemistry and biophysics. 1997;342:231-5. [56] Schjerning Olsen, A. M., Fosbol, E. L., Lindhardsen, J., Folke, F., Charlot, M., Selmer, C., et al. Duration of treatment with nonsteroidal anti-inflammatory drugs and impact on risk of death and recurrent myocardial infarction in patients with prior myocardial infarction: a nationwide cohort study. Circulation. 2011;123:2226-35. [57] Jana, N. R. NSAIDs and apoptosis. Cellular and molecular life sciences : CMLS. 2008;65:1295-301. [58] Farkouh, M. E., Greenberg, B. P. An evidence-based review of the cardiovascular risks of nonsteroidal anti-inflammatory drugs. The American journal of cardiology. 2009;103:1227-37. [59] Frolov, R. V., Berim, I. G., Singh, S. Inhibition of delayed rectifier potassium channels and induction of arrhythmia: a novel effect of celecoxib and the mechanism underlying it. The Journal of biological chemistry. 2008;283:1518-24. [60] Warner, T. D., Mitchell, J. A. COX-2 selectivity alone does not define the cardiovascular risks associated with non-steroidal anti-inflammatory drugs. Lancet. 2008;371:270-3. [61] Liu, J. Y., Li, N., Yang, J., Qiu, H., Ai, D., Chiamvimonvat, N., et al. Metabolic profiling of murine plasma reveals an unexpected biomarker in rofecoxib-mediated cardiovascular events. Proceedings of the National Academy of Sciences of the United States of America. 2010; 107:17017-22. [62] Harizi, H., Corcuff, J. B., Gualde, N. Arachidonic-acid-derived eicosanoids: roles in biology and immunopathology. Trends in molecular medicine. 2008;14:461-9. [63] Kaduce, T. L., Fang, X., Harmon, S. D., Oltman, C. L., Dellsperger, K. C., Teesch, L. M., et al. 20-hydroxyeicosatetraenoic acid (20-HETE) 156 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al.

metabolism in coronary endothelial cells. The Journal of biological chemistry. 2004;279:2648-56. [64] Escalante, B., Sessa, W. C., Falck, J. R., Yadagiri, P., Schwartzman, M. L. Vasoactivity of 20-hydroxyeicosatetraenoic acid is dependent on metabolism by cyclooxygenase. The Journal of pharmacology and experimental therapeutics. 1989;248:229-32. [65] Van Empel, V. P., Bertrand, A. T., Hofstra, L., Crijns, H. J., Doevendans, P. A., De Windt, L. J. Myocyte apoptosis in heart failure. Cardiovascular research. 2005;67:21-9. [66] Olivetti, G., Quaini, F., Sala, R., Lagrasta, C., Corradi, D., Bonacina, E., et al. Acute myocardial infarction in humans is associated with activation of programmed myocyte cell death in the surviving portion of the heart. Journal of molecular and cellular cardiology. 1996;28:2005- 16. [67] Olivetti, G., Abbi, R., Quaini, F., Kajstura, J., Cheng, W., Nitahara, J. A., et al. Apoptosis in the failing human heart. The New England journal of medicine. 1997;336:1131-41. [68] Saraste, A., Pulkki, K., Kallajoki, M., Heikkila, P., Laine, P., Mattila, S., et al. Cardiomyocyte apoptosis and progression of heart failure to transplantation. European journal of clinical investigation. 1999;29:380- 6. [69] Guerra, S., Leri, A., Wang, X., Finato, N., Di Loreto, C., Beltrami, C. A., et al. Myocyte death in the failing human heart is gender dependent. Circulation research. 1999;85:856-66. [70] Ottaviani, G., Lavezzi, A. M., Rossi, L., Matturri, L. Proliferating cell nuclear antigen (PCNA) and apoptosis in hyperacute and acute myocardial infarction. European journal of histochemistry : EJH. 1999; 43:7-14. [71] Wencker, D., Chandra, M., Nguyen, K., Miao, W., Garantziotis, S., Factor, S. M., et al. A mechanistic role for cardiac myocyte apoptosis in heart failure. The Journal of clinical investigation. 2003;111:1497-504. [72] Braun, F. K., Al-Yacoub, N., Plotz, M., Mobs, M., Sterry, W., Eberle, J. Nonsteroidal anti-inflammatory drugs induce apoptosis in cutaneous T- cell lymphoma cells and enhance their sensitivity for TNF-related apoptosis-inducing ligand. The Journal of investigative dermatology. 2012;132:429-39. [73] Hawcroft, G., D'Amico, M., Albanese, C., Markham, A. F., Pestell, R. G., Hull, M. A. Indomethacin induces differential expression of beta- Non-Steroidal Anti-Inflammatory Drugs ... 157

catenin, gamma-catenin and T-cell factor target genes in human colorectal cancer cells. Carcinogenesis. 2002;23:107-14. [74] Piazza, G. A., Rahm, A. K., Finn, T. S., Fryer, B. H., Li, H., Stoumen, A. L., et al. Apoptosis primarily accounts for the growth-inhibitory properties of sulindac metabolites and involves a mechanism that is independent of cyclooxygenase inhibition, cell cycle arrest, and p53 induction. Cancer research. 1997;57:2452-9. [75] Shiff, S. J., Koutsos, M. I., Qiao, L., Rigas, B. Nonsteroidal antiinflammatory drugs inhibit the proliferation of colon adenocarcinoma cells: effects on cell cycle and apoptosis. Experimental cell research. 1996;222:179-88. [76] Shiff, S. J., Qiao, L., Tsai, L. L., Rigas, B. Sulindac sulfide, an aspirin- like compound, inhibits proliferation, causes cell cycle quiescence, and induces apoptosis in HT-29 colon adenocarcinoma cells. The Journal of clinical investigation. 1995;96:491-503. [77] Zhou, X. M., Wong, B. C., Fan, X. M., Zhang, H. B., Lin, M. C., Kung, H. F., et al. Non-steroidal anti-inflammatory drugs induce apoptosis in gastric cancer cells through up-regulation of bax and bak. Carcinogenesis. 2001;22:1393-7. [78] Zimmermann, K. C., Waterhouse, N. J., Goldstein, J. C., Schuler, M., Green, D. R. Aspirin induces apoptosis through release of cytochrome c from mitochondria. Neoplasia. 2000;2:505-13. [79] Kulp, S. K., Yang, Y. T., Hung, C. C., Chen, K. F., Lai, J. P., Tseng, P. H., et al. 3-phosphoinositide-dependent protein kinase-1/Akt signaling represents a major cyclooxygenase-2-independent target for celecoxib in prostate cancer cells. Cancer research. 2004;64:1444-51. [80] Wang, X., Baek, S. J., Eling, T. COX inhibitors directly alter gene expression: role in cancer prevention? Cancer metastasis reviews. 2011; 30:641-57. [81] Chen, Y., Hughes-Fulford, M. Prostaglandin E2 and the protein kinase A pathway mediate arachidonic acid induction of c-fos in human prostate cancer cells. Br. J. Cancer. 2000;82:2000-6. [82] Banu, S. K., Lee, J., Speights, V. O., Jr., Starzinski-Powitz, A., Arosh, J. A. Selective inhibition of prostaglandin E2 receptors EP2 and EP4 induces apoptosis of human endometriotic cells through suppression of ERK1/2, AKT, NFkappaB, and beta-catenin pathways and activation of intrinsic apoptotic mechanisms. Mol. Endocrinol. 2009;23:1291-305. [83] Elrod, H. A., Sun, S. Y. PPARgamma and Apoptosis in Cancer. PPAR research. 2008;2008:704165. 158 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al.

[84] Shimada, T., Kojima, K., Yoshiura, K., Hiraishi, H., Terano, A. Characteristics of the peroxisome proliferator activated receptor gamma (PPARgamma) ligand induced apoptosis in colon cancer cells. Gut. 2002;50:658-64. [85] Nagamine, M., Okumura, T., Tanno, S., Sawamukai, M., Motomura, W., Takahashi, N., et al. PPAR gamma ligand-induced apoptosis through a p53-dependent mechanism in human gastric cancer cells. Cancer science. 2003;94:338-43. [86] Townsend, K. P., Pratico, D. Novel therapeutic opportunities for Alzheimer's disease: focus on nonsteroidal anti-inflammatory drugs. FASEB J. 2005;19:1592-601. [87] Chu, E. C., Chai, J., Tarnawski, A. S. NSAIDs activate PTEN and other phosphatases in human colon cancer cells: novel mechanism for chemopreventive action of NSAIDs. Biochemical and biophysical research communications. 2004;320:875-9. [88] Silverman, E. S., Collins, T. Pathways of Egr-1-mediated gene transcription in vascular biology. The American journal of pathology. 1999; 154:665-70. [89] Nair, P., Muthukkumar, S., Sells, S. F., Han, S. S., Sukhatme, V. P., Rangnekar, V. M. Early growth response-1-dependent apoptosis is mediated by p53. The Journal of biological chemistry. 1997;272:20131- 8. [90] Baek, S. J., Kim, J. S., Moore, S. M., Lee, S. H., Martinez, J., Eling, T. E. Cyclooxygenase inhibitors induce the expression of the tumor suppressor gene EGR-1, which results in the up-regulation of NAG-1, an antitumorigenic protein. Molecular pharmacology. 2005;67:356-64. [91] Virolle, T., Adamson, E. D., Baron, V., Birle, D., Mercola, D., Mustelin, T., et al. The Egr-1 transcription factor directly activates PTEN during irradiation-induced signalling. Nature cell biology. 2001;3:1124-8. [92] Chetram, M. A., Hinton, C. V. PTEN regulation of ERK1/2 signaling in cancer. J. Recept. Signal Transduct. Res. 2012;32:190-5. [93] Kim, J. H., Chang, J. H., Rhee, K. H., Yoon, J. H., Kwon, S. H., Song, K., et al. Cyclooxygenase inhibitors induce apoptosis in sinonasal cancer cells by increased expression of nonsteroidal anti-inflammatory drug- activated gene. International journal of cancer Journal international du cancer. 2008;122:1765-73. [94] Eling, T. E., Baek, S. J., Shim, M., Lee, C. H. NSAID activated gene (NAG-1), a modulator of tumorigenesis. Journal of biochemistry and molecular biology. 2006;39:649-55. Non-Steroidal Anti-Inflammatory Drugs ... 159

[95] Baek, S. J., Kim, K. S., Nixon, J. B., Wilson, L. C., Eling, T. E. Cyclooxygenase inhibitors regulate the expression of a TGF-beta superfamily member that has proapoptotic and antitumorigenic activities. Molecular pharmacology. 2001;59:901-8. [96] Dikshit, P., Chatterjee, M., Goswami, A., Mishra, A., Jana, N. R. Aspirin induces apoptosis through the inhibition of proteasome function. The Journal of biological chemistry. 2006;281:29228-35. [97] Zhang, Y. J., Dai, Q., Wu, S. M., Zhu, H. Y., Shen, G. F., Li, E. L., et al. Susceptibility for NSAIDs-induced apoptosis correlates to p53 gene status in gastric cancer cells. Cancer investigation. 2008;26:868-77. [98] Zhu, G. H., Wong, B. C., Ching, C. K., Lai, K. C., Lam, S. K. Differential apoptosis by indomethacin in gastric epithelial cells through the constitutive expression of wild-type p53 and/or up-regulation of c- myc. Biochemical pharmacology. 1999;58:193-200. [99] Shureiqi, I., Chen, D., Lotan, R., Yang, P., Newman, R. A., Fischer, S. M., et al. 15-Lipoxygenase-1 mediates nonsteroidal anti-inflammatory drug-induced apoptosis independently of cyclooxygenase-2 in colon cancer cells. Cancer research. 2000;60:6846-50. [100] Sheikh, M. S., Huang, Y. Death receptor activation complexes: it takes two to activate TNF receptor 1. Cell Cycle. 2003;2:550-2. [101] Roth, W., Reed, J. C. FLIP protein and TRAIL-induced apoptosis. Vitamins and hormones. 2004;67:189-206. [102] Smartt, H. J., Elder, D. J., Hicks, D. J., Williams, N. A., Paraskeva, C. Increased NF-kappaB DNA binding but not transcriptional activity during apoptosis induced by the COX-2-selective inhibitor NS-398 in colorectal carcinoma cells. Br. J. Cancer. 2003;89:1358-65. [103] Huang, Y. C., Chuang, L. Y., Hung, W. C. Mechanisms underlying nonsteroidal anti-inflammatory drug-induced p27(Kip1) expression. Molecular pharmacology. 2002;62:1515-21. [104] Tong, Q. S., Zheng, L. D., Wang, L., Liu, J., Qian, W. BAK overexpression mediates p53-independent apoptosis inducing effects on human gastric cancer cells. BMC Cancer. 2004;4:33. [105] Zong, W. X., Li, C., Hatzivassiliou, G., Lindsten, T., Yu, Q. C., Yuan, J., et al. Bax and Bak can localize to the endoplasmic reticulum to initiate apoptosis. J. Cell. Biol. 2003;162:59-69. [106] Suzuki, K., Sawa, Y., Ichikawa, H., Kaneda, Y., Matsuda, H. Myocardial protection with endogenous overexpression of manganese superoxide dismutase. Ann. Thorac. Surg. 1999;68:1266-71. 160 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al.

[107] Abdelrahim, M., Safe, S. Cyclooxygenase-2 inhibitors decrease vascular endothelial growth factor expression in colon cancer cells by enhanced degradation of Sp1 and Sp4 proteins. Molecular pharmacology. 2005; 68:317-29. [108] Gupta, K., Kshirsagar, S., Li, W., Gui, L., Ramakrishnan, S., Gupta, P., et al. VEGF prevents apoptosis of human microvascular endothelial cells via opposing effects on MAPK/ERK and SAPK/JNK signaling. Experimental cell research. 1999;247:495-504. [109] Li, Y., Zhang, F., Nagai, N., Tang, Z., Zhang, S., Scotney, P., et al. VEGF-B inhibits apoptosis via VEGFR-1-mediated suppression of the expression of BH3-only protein genes in mice and rats. The Journal of clinical investigation. 2008;118:913-23. [110] Kasahara, Y., Tuder, R. M., Taraseviciene-Stewart, L., Le Cras, T. D., Abman, S., Hirth, P. K., et al. Inhibition of VEGF receptors causes lung cell apoptosis and emphysema. The Journal of clinical investigation. 2000;106:1311-9. [111] Simizu, S., Takada, M., Umezawa, K., Imoto, M. Requirement of caspase-3(-like) protease-mediated hydrogen peroxide production for apoptosis induced by various anticancer drugs. The Journal of biological chemistry. 1998;273:26900-7. [112] Verhaegen, S., McGowan, A. J., Brophy, A. R., Fernandes, R. S., Cotter, T. G. Inhibition of apoptosis by antioxidants in the human HL-60 leukemia cell line. Biochemical pharmacology. 1995;50:1021-9. [113] Kusuhara, H., Komatsu, H., Sumichika, H., Sugahara, K. Reactive oxygen species are involved in the apoptosis induced by nonsteroidal anti-inflammatory drugs in cultured gastric cells. European journal of pharmacology. 1999;383:331-7. [114] O'Connor, N., Dargan, P. I., Jones, A. L. Hepatocellular damage from non-steroidal anti-inflammatory drugs. QJM : monthly journal of the Association of Physicians. 2003;96:787-91. [115] Su, H., Wang, X. The ubiquitin-proteasome system in cardiac proteinopathy: a quality control perspective. Cardiovascular research. 2010;85:253-62. [116] Gomes, A. V., Zong, C., Ping, P. Protein degradation by the 26S proteasome system in the normal and stressed myocardium. Antioxidants and redox signaling. 2006;8:1677-91. [117] Ferrarini, M., Ferrero, E. Proteasome inhibitors and modulators of angiogenesis in multiple myeloma. Current medicinal chemistry. 2011; 18:5185-95. Non-Steroidal Anti-Inflammatory Drugs ... 161

[118] Nowis, D., Maczewski, M., Mackiewicz, U., Kujawa, M., Ratajska, A., Wieckowski, M. R., et al. Cardiotoxicity of the anticancer therapeutic agent bortezomib. The American journal of pathology. 2010;176:2658- 68. [119] Gomes, A. V., Young, G. W., Wang, Y., Zong, C., Eghbali, M., Drews, O., et al. Contrasting proteome biology and functional heterogeneity of the 20 S proteasome complexes in mammalian tissues. Molecular and cellular proteomics : MCP. 2009;8:302-15. [120] Sandoval-Acuna, C., Lopez-Alarcon, C., Aliaga, M. E., Speisky, H. Inhibition of mitochondrial complex I by various non-steroidal anti- inflammatory drugs and its protection by quercetin via a coenzyme Q- like action. Chemico-biological interactions. 2012;199:18-28. [121] Schoultz, I., McKay, C. M., Graepel, R., Phan, V. C., Wang, A., Soderholm, J. D., et al. Indomethacin-induced translocation of bacteria across enteric epithelia is reactive oxygen species-dependent and reduced by vitamin-C. American journal of physiology Gastrointestinal and liver physiology. 2012. [122] Fosslien, E. Review: Mitochondrial medicine--cardiomyopathy caused by defective oxidative phosphorylation. Annals of clinical and laboratory science. 2003;33:371-95. [123] Amin, A. S., Tan, H. L., Wilde, A. A. Cardiac ion channels in health and disease. Heart rhythm : the official journal of the Heart Rhythm Society. 2010;7:117-26. [124] Priori, S. G., Napolitano, C. Genetics of cardiac arrhythmias and sudden cardiac death. Annals of the New York Academy of Sciences. 2004;1015: 96-110. [125] Kaab, S., Dixon, J., Duc, J., Ashen, D., Nabauer, M., Beuckelmann, D. J., et al. Molecular basis of transient outward potassium current downregulation in human heart failure: a decrease in Kv4.3 mRNA correlates with a reduction in current density. Circulation. 1998; 98: 1383-93. [126] Macias, A., Moreno, C., Moral-Sanz, J., Cogolludo, A., David, M., Alemanni, M., et al. Celecoxib blocks cardiac Kv1.5, Kv4.3 and Kv7.1 (KCNQ1) channels: effects on cardiac action potentials. Journal of molecular and cellular cardiology. 2010;49:984-92. [127] Brueggemann, L. I., Mackie, A. R., Mani, B. K., Cribbs, L. L., Byron, K. L. Differential effects of selective cyclooxygenase-2 inhibitors on vascular smooth muscle ion channels may account for differences in cardiovascular risk profiles. Molecular pharmacology. 2009;76:1053-61. 162 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al.

[128] Bodi, I., Mikala, G., Koch, S. E., Akhter, S. A., Schwartz, A. The L-type calcium channel in the heart: the beat goes on. The Journal of clinical investigation. 2005;115:3306-17. [129] Hasinoff, B. B., Patel, D., Wu, X. The cytotoxicity of celecoxib towards cardiac myocytes is cyclooxygenase-2 independent. Cardiovascular toxicology. 2007;7:19-27. [130] Remme, C. A., Bezzina, C. R. Sodium channel (dys)function and cardiac arrhythmias. Cardiovascular therapeutics. 2010;28:287-94. [131] Park, S. Y., Kim, T. H., Kim, H. I., Shin, Y. K., Lee, C. S., Park, M., et al. Celecoxib inhibits Na+ currents in rat dorsal root ganglion neurons. Brain research. 2007;1148:53-61. [132] Zheng, Z. J., Croft, J. B., Giles, W. H., Mensah, G. A. Sudden cardiac death in the United States, 1989 to 1998. Circulation. 2001;104:2158- 63. [133] Simmons, A., Pimentel, R., Lakkireddy, D. Sudden cardiac death in women. Reviews in cardiovascular medicine. 2012;13:e37-42. [134] Zipes, D. P., Wellens, H. J. Sudden cardiac death. Circulation. 1998;98: 2334-51. [135] Srivathsan, K., Ng, D. W., Mookadam, F. Ventricular tachycardia and ventricular fibrillation. Expert review of cardiovascular therapy. 2009;7: 801-9. [136] Cutler, M. J., Jeyaraj, D., Rosenbaum, D. S. Cardiac electrical remodeling in health and disease. Trends Pharmacol. Sci. 2011;32:174- 80. [137] Thompson, B. S. Sudden cardiac death and heart failure. AACN Adv. Crit. Care. 2009;20:356-65. [138] Cantillon, D. J., Stein, K. M., Markowitz, S. M., Mittal, S., Shah, B. K., Morin, D. P., et al. Predictive value of microvolt T-wave alternans in patients with left ventricular dysfunction. Journal of the American College of Cardiology. 2007;50:166-73. [139] Calo, L., De Santo, T., Nuccio, F., Sciarra, L., De Luca, L., Stefano, L. M., et al. Predictive value of microvolt T-wave alternans for cardiac death or ventricular tachyarrhythmic events in ischemic and nonischemic cardiomyopathy patients: a meta-analysis. Annals of noninvasive electrocardiology : the official journal of the International Society for Holter and Noninvasive Electrocardiology, Inc. 2011;16:388-402. [140] Willott, R. H., Gomes, A. V., Chang, A. N., Parvatiyar, M. S., Pinto, J. R., Potter, J. D. Mutations in Troponin that cause HCM, DCM AND Non-Steroidal Anti-Inflammatory Drugs ... 163

RCM: what can we learn about thin filament function? Journal of molecular and cellular cardiology. 2010;48:882-92. [141] Xu, Q., Dewey, S., Nguyen, S., Gomes, A. V. Malignant and benign mutations in familial cardiomyopathies: insights into mutations linked to complex cardiovascular phenotypes. Journal of molecular and cellular cardiology. 2010;48:899-909. [142] Young, J. B., Abraham, W. T., Smith, A. L., Leon, A. R., Lieberman, R., Wilkoff, B., et al. Combined cardiac resynchronization and implantable cardioversion defibrillation in advanced chronic heart failure: the MIRACLE ICD Trial. JAMA : the journal of the American Medical Association. 2003;289:2685-94. [143] Knollmann, B. C., Roden, D. M. A genetic framework for improving arrhythmia therapy. Nature. 2008;451:929-36. [144] Spirito, P., Seidman, C. E., McKenna, W. J., Maron, B. J. The management of hypertrophic cardiomyopathy. The New England journal of medicine. 1997;336:775-85. [145] Florea, S. M., Blatter, L. A. The role of mitochondria for the regulation of cardiac alternans. Frontiers in physiology. 2010;1:141. [146] Whitehouse, M. W. Biochemical Properties of Anti-Inflammatory Drugs--Iii. Uncoupling of Oxidative Phosphorylation in a Connective Tissue (Cartilage) and Liver Mitochondria by Salicylate Analogues: Relationship of Structure to Activity. Biochemical pharmacology. 1964; 13:319-36. [147] Moreno-Sanchez, R., Bravo, C., Vasquez, C., Ayala, G., Silveira, L. H., Martinez-Lavin, M. Inhibition and uncoupling of oxidative phosphorylation by nonsteroidal anti-inflammatory drugs: study in mitochondria, submitochondrial particles, cells, and whole heart. Biochemical pharmacology. 1999;57:743-52. [148] Mahmud, T., Rafi, S. S., Scott, D. L., Wrigglesworth, J. M., Bjarnason, I. Nonsteroidal antiinflammatory drugs and uncoupling of mitochondrial oxidative phosphorylation. Arthritis and rheumatism. 1996;39:1998- 2003. [149] Whelan, R. S., Kaplinskiy, V., Kitsis, R. N. Cell death in the pathogenesis of heart disease: mechanisms and significance. Annual review of physiology. 2010;72:19-44. [150] Lauribe, P., Benchimol, D., Dartigues, J. F., Dada, S., Benchimol, H., Drouillet, F., et al. Biological risk factors for sudden death in patients with coronary artery disease and without heart failure. International journal of cardiology. 1992;34:307-18. 164 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al.

[151] Belevych, A. E., Terentyev, D., Viatchenko-Karpinski, S., Terentyeva, R., Sridhar, A., Nishijima, Y., et al. Redox modification of ryanodine receptors underlies calcium alternans in a canine model of sudden cardiac death. Cardiovascular research. 2009;84:387-95. [152] Aspirin for the prevention of cardiovascular disease: US. Preventive Services Task Force recommendation statement. Annals of internal medicine. 2009;150:396-404. [153] Verheugt, F. W., Bolte, A. C. The role of aspirin in women's health. Int. J. Womens Health. 2011;3:151-66. [154] Chopra, N., Knollmann, B. C. Genetics of sudden cardiac death syndromes. Current opinion in cardiology. 2011;26:196-203. [155] Yokoyama, H., Ito, N., Soeda, S., Ozaki, M., Suzuki, Y., Watanabe, M., et al. Influence of non-steroidal anti-inflammatory drugs on antiplatelet effect of aspirin. J. Clin. Pharm. Ther. 2012. [156] Meek, I. L., Vonkeman, H. E., Kasemier, J., Movig, K. L., van de Laar, M. A. Interference of NSAIDs with the thrombocyte inhibitory effect of aspirin: A placebo-controlled, ex vivo, serial placebo-controlled serial crossover study. Eur. J. Clin. Pharmacol. 2012. [157] CIPRO® (ciprofloxacin hydrochloride). http://www.accessdata.fda.gov/ drugsatfda_docs/label/2008/019537s68,19847s42,19857s49,20780s26,2 1473s24lbl.pdf: Bayer HealthCare Pharmaceuticals Inc.; 2008. [158] Page, J., Henry, D. Consumption of NSAIDs and the development of congestive heart failure in elderly patients: an underrecognized public health problem. Archives of internal medicine. 2000;160:777-84. [159] Gislason, G. H., Rasmussen, J. N., Abildstrom, S. Z., Schramm, T. K., Hansen, M. L., Fosbol, E. L., et al. Increased mortality and cardiovascular morbidity associated with use of nonsteroidal anti- inflammatory drugs in chronic heart failure. Archives of internal medicine. 2009;169:141-9. [160] Lanza, F. L., Chan, F. K., Quigley, E. M. Guidelines for prevention of NSAID-related ulcer complications. The American journal of gastroenterology. 2009;104:728-38. [161] Asif, I. M., Drezner, J. A. Sudden cardiac death and preparticipation screening: the debate continues-in support of electrocardiogram- inclusive preparticipation screening. Progress in cardiovascular diseases. 2012;54:445-50. [162] Gigante, A., Tagarro, I. Non-steroidal anti-inflammatory drugs and gastroprotection with proton pump inhibitors: a focus on ketoprofen/omeprazole. Clinical drug investigation. 2012;32:221-33. Non-Steroidal Anti-Inflammatory Drugs ... 165

[163] Loyd, M., Rublee, D., Jacobs, P. An economic model of long-term use of celecoxib in patients with osteoarthritis. BMC Gastroenterol. 2007; 7: 25. [164] Ugurlucan, M., Caglar, I. M., Caglar, F. N., Ziyade, S., Karatepe, O., Yildiz, Y., et al. Aspirin: from a historical perspective. Recent patents on cardiovascular drug discovery. 2012;7:71-6. [165] Rainsford, K. D. Ibuprofen: pharmacology, efficacy and safety. Inflammopharmacology. 2009;17:275-342. [166] Suthisisang, C. C., Poolsup, N., Suksomboon, N., Lertpipopmetha, V., Tepwitukgid, B. Meta-analysis of the efficacy and safety of naproxen sodium in the acute treatment of migraine. Headache. 2010;50:808-18. [167] Kellner, H. L., Li, C., Essex, M. N. Efficacy and safety of celecoxib versus diclofenac and omeprazole in elderly arthritis patients: a subgroup analysis of the CONDOR Trial. Current medical research and opinion. 2012. [168] Aygun, D., Kaplan, S., Odaci, E., Onger, M. E., Altunkaynak, M. E. Toxicity of non-steroidal anti-inflammatory drugs: a review of melatonin and diclofenac sodium association. Histology and histopathology. 2012; 27:417-36. [169] Morse, D. R., Esposito, J. V., Furst, M. L. Comparison of prophylactic and on-demand diflunisal for pain management of patients having one- visit endodontic therapy. Oral surgery, oral medicine, and oral pathology. 1990;69:729-36. [170] Castano, A., Helmke, S., Alvarez, J., Delisle, S., Maurer, M. S. Diflunisal for ATTR Cardiac Amyloidosis. Congest. Heart Fail. 2012. [171] Pareek, A., Chandurkar, N., Gupta, A., Desai, Y., Kumar, S. H., Swamy, A., et al. Comparative evaluation of efficacy and safety of etodolac and diclofenac sodium injection in patients with postoperative orthopedic pain. Current medical research and opinion. 2011;27:2107-15. [172] Traa, M. X., Derry, S., Moore, R. A. Single dose oral fenoprofen for acute postoperative pain in adults. Cochrane Database Syst. Rev. 2011: CD007556. [173] Sultan, A., McQuay, H. J., Moore, R. A., Derry, S. Single dose oral flurbiprofen for acute postoperative pain in adults. Cochrane Database Syst. Rev. 2009:CD007358. [174] Kanno, T., Tsuchiya, A., Shimizu, T., Tanaka, A., Nishizaki, T. Novel Indomethacin Action: Selective and Direct Activation of Protein Kinase C-epsilon. Cellular physiology and biochemistry : international journal 166 Soyun M. Hwang, Jennifer E. Gilda, Ziyou Cui et al.

of experimental cellular physiology, biochemistry, and pharmacology. 2012;30:771-7. [175] Sano, T., King, J. N., Seewald, W., Sakakibara, N., Okumura, M. Comparison of oral robenacoxib and ketoprofen for the treatment of acute pain and inflammation associated with musculoskeletal disorders in cats: A randomised clinical trial. Vet. J. 2012. [176] Mongini, F., Bona, G., Garnero, M., Gioria, A. Efficacy of meclofenamate sodium versus placebo in headache and craniofacial pain. Headache. 1993;33:22-8. [177] Kean, W. F. Oxaprozin: kinetic and dynamic profile in the treatment of pain. Current medical research and opinion. 2004;20:1275-7. [178] Sacks, W. The use of an isopyrine-phenylbutazone compound in migraine. South African medical journal = Suid-Afrikaanse tydskrif vir geneeskunde. 1980;58:444-6. [179] Ayyanathan, K., Kesaraju, S., Dawson-Scully, K., Weissbach, H. Combination of Sulindac and Dichloroacetate Kills Cancer Cells via Oxidative Damage. PloS one. 2012;7:e39949. [180] Karademir, S., Oguz, D., Senocak, F., Ocal, B., Karakurt, C., Cabuk, F. Tolmetin and salicylate therapy in acute rheumatic fever: Comparison of clinical efficacy and side-effects. Pediatrics international : official journal of the Japan Pediatric Society. 2003;45:676-9. [181] Sacco, M., Pellegrini, F., Roncaglioni, M. C., Avanzini, F., Tognoni, G., Nicolucci, A. Primary prevention of cardiovascular events with low-dose aspirin and vitamin E in type 2 diabetic patients: results of the Primary Prevention Project (PPP) trial. Diabetes care. 2003;26:3264-72. [182] Berger, J. S., Brown, D. L., Becker, R. C. Low-dose aspirin in patients with stable cardiovascular disease: a meta-analysis. The American journal of medicine. 2008;121:43-9. [183] Schmidt, M., Christiansen, C. F., Mehnert, F., Rothman, K. J., Sorensen, H. T. Non-steroidal anti-inflammatory drug use and risk of atrial fibrillation or flutter: population based case-control study. BMJ. 2011; 343:d3450. [184] Fosbol, E. L., Folke, F., Jacobsen, S., Rasmussen, J. N., Sorensen, R., Schramm, T. K., et al. Cause-specific cardiovascular risk associated with nonsteroidal antiinflammatory drugs among healthy individuals. Circulation Cardiovascular quality and outcomes. 2010;3:395-405. [185] Combe, B., Swergold, G., McLay, J., McCarthy, T., Zerbini, C., Emery, P., et al. Cardiovascular safety and gastrointestinal tolerability of Non-Steroidal Anti-Inflammatory Drugs ... 167

etoricoxib vs diclofenac in a randomized controlled clinical trial (The MEDAL study). Rheumatology (Oxford). 2009;48:425-32. [186] Aisen, P. S., Schafer, K. A., Grundman, M., Pfeiffer, E., Sano, M., Davis, K. L., et al. Effects of rofecoxib or naproxen vs placebo on Alzheimer disease progression: a randomized controlled trial. JAMA : the journal of the American Medical Association. 2003;289:2819-26.

Conflict of interest statement. No conflicts of interest.

Sources of funding. UC Davis Startup funds.