Cardiovascular Safety of Nonsteroidal Anti-inflammatory Drugs

Islam Y. Elgendy, MD; Anthony A. Bavry, MD, MPH


January 19, 2016

Nonsteroidal anti-inflammatory drugs (NSAIDs) are commonly used analgesics worldwide. In 2013, over 100 million prescriptions were written for NSAIDs in the United States,[1] and many are available over the counter.

NSAIDs exert their effect by nonselective inhibition of both cyclooxygenase (COX)-1 and COX-2 isoenzymes. COX-1 is produced constantly in the tissues, whereas COX-2 is induced mainly during inflammation. Selective COX-2 inhibitors (coxibs) were developed to reduce the gastrointestinal ulceration and bleeding related to COX-1 inhibition. COX-2 inhibitors have analgesic and anti-inflammatory efficacy similar to that of nonselective COX inhibitors.

Although aspirin has been shown to be beneficial for secondary prevention of cardiovascular events,[2] the cardiovascular safety of other NSAIDs is a concern and is the focus of the present review.

Cardiovascular Risk of Selective COX-2 Inhibitors

In the early 2000s, concerns about cardiovascular side effects from COX-2 inhibitors were noted, namely with rofecoxib and valdecoxib.[3,4,5] In a meta-analysis of 138 randomized trials, COX-2 inhibitors were associated with a significant increase in the risk for myocardial infarction (MI) (risk ratio [RR], 1.86; 95% confidence interval [CI], 1.33-2.59; P =.0003) and vascular events (RR, 1.42; 95% CI, 1.13-1.78; P = .003) compared with placebo.[6]

In September 2004, the manufacturer voluntarily withdrew rofecoxib from the market. In late 2004, the US Food and Drug Administration (FDA) declared a "black box" warning for valdecoxib, stating that it is contraindicated in patients undergoing coronary artery bypass surgery. At the same time, the FDA also issued a public health advisory regarding the use of COX-2 inhibitors that advised physicians to weigh the benefits vs risks of using valdecoxib and celecoxib. This statement considered patients who are at high risk for gastrointestinal bleeding, are intolerant of nonselective NSAIDs, or are not doing well on nonselective NSAIDs to be appropriate candidates for COX-2 inhibitors.[7]

Animal studies suggest that the adverse cardiovascular effects observed with COX-2 inhibitors are due to enhanced endothelial thrombosis (as a result of reduction in prostacyclin synthesis), sodium and water retention, and loss of the protective effects of COX-2 upregulation in the setting of MI, resulting in larger infarction size and thinning of the left ventricular wall in the infarct zone.[8]

Cardiovascular Risk of Nonselective NSAIDs

Concerns about the cardiovascular side effects with COX-2 inhibitors has led to an increase in the use of nonselective NSAIDs during the past few years.[9] Examples of these agents include ibuprofen and naproxen.

Earlier meta-analyses of observational studies and randomized trials considered naproxen and ibuprofen as possible "safer agents."[6,10] Multiple studies had consistently shown an increased risk for cardiovascular events with diclofenac, but mixed findings for naproxen and ibuprofen.[11,12] Subsequently, data accrued linking naproxen and ibuprofen with an increased risk for cardiovascular events. In a network meta-analysis,[13] ibuprofen was associated with the highest risk for stroke among the different types of NSAIDs assessed.

The Alzheimer's Disease Anti-inflammatory Prevention Trial (ADAPT) was a randomized trial designed to evaluate the effect of naproxen vs celecoxib and vs placebo on cognitive function in elderly persons. It was prematurely terminated after the Adenoma Prevention with Celecoxib trial reported increased cardiovascular risks with celecoxib.[14] As it happened, ADAPT did not show an increased risk for cardiovascular events for celecoxib, but there was possible evidence of cardiovascular and cerebrovascular hazard with naproxen compared with placebo.[15] A 2007 American Heart Association Scientific Statement on NSAIDs noted limitations of the ADAPT study and suggested that naproxen remained the preferred choice.[16]

In a systematic review of meta-analyses conducted on this topic, the Safety of Nonsteroidal Anti-Inflammatory Drugs Project found only seven studies on traditional NSAIDs and concluded that there are important knowledge gaps regarding the cardiovascular safety of traditional NSAIDs.[17] In a large meta-analysis by the Coxib and traditional NSAID Trialists' Collaboration, the authors concluded that the cardiovascular risks of high-dose diclofenac, and possibly ibuprofen, were similar to those of coxibs, whereas high-dose naproxen was associated with less cardiovascular risk than other NSAIDs.[18]

NSAIDs in Patients with Coronary Artery Disease

The use of both COX-2 inhibitors and nonselective NSAIDs in the post-MI period has been linked to increased mortality. Gislason and colleagues[19] analyzed over 58,000 patients with MI from the Danish National Patient Registry, more than 20,000 of whom had filled at least one NSAID prescription. They demonstrated a hazard ratio (HR) for mortality of 2.8 for any use of rofecoxib (95% CI, 2.41-3.25), 2.57 for celecoxib (95% CI, 2.15-3.08), 1.5 for ibuprofen (95% CI, 1.36-1.67), 2.4 for diclofenac (95% CI, 2.09-2.80), and 1.29 for other NSAIDs (95% CI, 1.16-1.43). A UK study[20] found that patients who took ibuprofen along with aspirin 30 days after being hospitalized for cardiovascular disease had a higher risk for all-cause mortality (HR, 1.93; 95% CI, 1.30-2.87; P =.0011) and cardiovascular mortality (HR, 1.73; 95% CI, 1.05-2.84; P = .0305) than those taking aspirin alone.

Another Danish analysis using the National Patient Registry showed that even short-term treatment (around 7-14 days) with most NSAIDs was associated with an increased risk for death or recurrent MI in patients with prior MI.[21] These agents might interfere with aspirin, thus blunting the protective effect of aspirin in post-MI patients.

There are fewer data regarding NSAIDs use in patients with stable coronary artery disease. In a post hoc analysis of the from the INternational VErapamilTrandolapril STudy (INVEST),[22] chronic self-reported use of NSAIDs by hypertensive patients with coronary artery disease was associated with an increased risk for the composite of all-cause mortality, nonfatal MI, and nonfatal stroke (HR, 1.47; 95% CI, 1.19-1.82; P = .0003).

NSAIDs and Risk for Stroke

The evidence for a link between NSAIDs and stroke events is unclear.[23] In the network meta-analysis cited earlier, 26 of the 31 studies reported stroke data, yielding only 337 events.[10] Ibuprofen and diclofenac were associated with the highest risk for stroke.

Another analysis of four cohort studies and two nested case/control studies specifically analyzing stroke risk showed that rofecoxib (RR, 1.64; 95% CI, 1.15-2.33) and diclofenac (RR, 1.27; 95% CI, 1.08-1.48) were associated with increased risk for ischemic stroke; data were inadequate to estimate risk for other individual NSAIDs.[24]

NSAIDs in Women

The risk for cardiovascular events with NSAIDs among women was assessed in a study from the Women's Health Initiative.[25] Regular NSAID use was associated with an increased risk for cardiovascular events (HR, 1.10; 95% CI, 1.06-1.15; P < .001). Celecoxib was associated with a modestly increased hazard for cardiovascular events (HR, 1.13; 95% CI, 1.01-1.27; P = .031). Among aspirin users, concomitant use of selective COX-2 inhibitors was no longer associated with an increased hazard for cardiovascular events.

The authors did not identify an increased risk with ibuprofen (HR, 1.00; 95% CI, 0.93-1.07; P = .996), but did show an increased risk for naproxen (HR, 1.22; 95% CI 1.12-1.34; P < .001). This harmful association remained among concomitant aspirin users.

Is There a Preferred Nonselective NSAID?

At present, data are insufficient to recommend a particular nonselective NSAID as being a safer agent. Ibuprofen has been associated with a relatively high risk for stroke, with some studies documenting no association with adverse cardiovascular events.[10,13] Naproxen is sometimes regarded as a safer agent; however, an observational study in women and an underpowered randomized trial in elderly persons cautioned about an increased risk for adverse cardiovascular events with this agent.[15,25]

Many researchers hoped that the Standard Care versus celecoxib Outcome Trial (SCOT) in patients with arthritis and no known cardiovascular disease would answer the question about the safety of nonselective NSAIDs in this population. However, the trial was prematurely terminated owing to loss of support from the sponsor because of probably futility. At a mean of 3.2 years, the risk for hospitalization for nonfatal MI, nonfatal stroke, or cardiovascular death was noninferior with celecoxib compared with nonselective NSAIDs.[26] Given the low rate of cardiovascular events, the investigators concluded that older, nonselective NSAIDs appeared to be acceptably safe in this population. The early termination, open-label design, and low-risk population left others cautious about overgeneralizing the results.

Since 2005, all available NSAIDs contain a black-box warning that they may increase the risk for serious cardiovascular events, MI, and stroke that might be fatal. This risk might increase with the duration of use.[27]


In patients with recent or remote history of MI, the data shows an increased risk for cardiovascular events with both COX-2 inhibitors and nonselective NSAIDs. The use of COX-2 inhibitors or nonselective NSAIDs is contraindicated in patients with acute MI. In patients with stable ischemic heart disease, caution with these agents is warranted.

The ongoing PRECISION (Prospective Randomized Evaluation of Celecoxib Integrated Safety Versus Ibuprofen Or Naproxen) trial, the first study of patients with high cardiovascular risk but without acute coronary syndrome who are receiving long-term treatment with a selective COX-2 inhibitor or traditional NSAIDs (ibuprofen or naproxen), will provide more information in the future.[28]

Meanwhile, the FDA recommendation for the lowest dosage for the shortest duration prevails. For patients with acute MI, coronary artery disease, or risk factors for cardiovascular disease, the American College of Cardiology/American Heart Association recommends a stepped approach as outlined in the Figure below, with COX-2 selective agents reserved as last-line management of intolerable pain.[29]

Figure. Stepped approach to pain management in patients with cardiovascular disease or cardiovascular risk factors. In patients at low risk for thrombotic events, use the lowest dose possible. Aspirin (81 mg) in all patients; add a proton pump inhibitor if they are taking nonsteroidal anti-inflammatory drugs. Monitor for hypertension, edema, worsening renal function, and gastrointestinal bleeding. If these occur, reduce dose or discontinue the drug. Adapted from Antman E, et al. Circulation. 2007;115:1634-1642.[16]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.