Meta-analysis Confirms Connection Between High-Dose NSAIDs, Vascular Risk

Download this article's Factoid (PDF & PPT for Gold Subscribers)

High-dose non-steroidal anti-inflammatory drugs (NSAIDs) increase the relative risk of major vascular events by about one-third, though naproxen appears to be an exception. The absolute increase is greatest in patients at high baseline vascular risk, according to a meta-analysis of randomized trials published online May 30, 2013, ahead of print in the Lancet.

Colin Baigent, MD, and members of the Coxib and traditional NSAID Trialists’ (CNT) Collaboration (Oxford, United Kingdom) analyzed 639 randomized trials comparing 1 NSAID vs. another (n = 165,456 person-years) or an NSAID vs. placebo (n = 68,342 person-years). Most trials reported individual patient-level data.

Primary outcomes were major vascular events (nonfatal MI, nonfatal stroke, or death from a vascular cause) and upper gastrointestinal (GI) complications (upper GI perforation, obstruction, or bleed).

Diclofenac, Ibuprofen as Risky as Coxibs

Major vascular events were elevated by more than a third in conjunction with coxibs or diclofenac, while ibuprofen showed a trend in the same direction but the difference did not reach significance. In contrast, naproxen was not associated with excess vascular risk. Hospitalization due to heart failure and upper GI complications increased across the board. Only coxibs imparted a significant rise in all-cause mortality (table 1).

Table 1. Effect of Various NSAIDs: Adjusted RR (95% CI)






Major Vascular Events

1.37 (1.14-1.66)

1.41 (1.12-1.78)

1.44 (0.89-2.33)

0.93 (0.69-1.27)

Hospitalization Due to Heart Failure

2.28 (1.62-3.20)

1.85 (1.17-2.94)

2.49 (1.19-5.20)

1.87 (1.10-3.16)


1.22 (1.04-1.44)

1.20 (0.94-1.54)

1.61 (0.90-2.88)

1.03 (0.71-1.49)

Upper GI Complications

1.81 (1.17-2.81)

1.89 (1.16-3.09)

3.97 (2.22-7.10)

4.22 (2.71-6.56)


Roughly 99% of all major vascular events occurred in trials involving a coxib or high-dose traditional NSAID (diclofenac 150 mg, ibuprofen 2,400 mg, or naproxen 1,000 mg daily). The rise in major vascular events was driven mainly by major coronary events, and the degree of increase in risk was independent of baseline characteristics.

Overall, the absolute increase was “small but serious,” the paper notes. Compared with placebo, a coxib or diclofenac added 3 major vascular events per 1,000 person-years, 1 of which was fatal.

According to predictive models, the absolute effects of NSAIDs were particularly evident in patients at high baseline risk.

Among patients with 2% annual risk of major vascular events, coxibs would likely result in 7 additional events per 1,000 person-years vs. placebo, 2 of which would be fatal. Diclofenac would add 8 events, 2 of which would be fatal, and ibuprofen would add 9 additional events, 3 of which would be fatal.

Among patients with 0.5% annual risk of upper GI complications, coxibs and diclofenac would likely result in 4 more events per 1,000 person-years than would placebo, ibuprofen 15 more events, and naproxen 16 more events.

Naproxen Appears Unique

“Our meta-analysis, which is unaffected by selection and other biases inherent in observational studies, showed clearly that the vascular risks of diclofenac, and ibuprofen, are similar to coxibs, but that naproxen is not associated with an increased risk of major vascular events,” Dr. Baigent and colleagues write. “However, it also showed that the excess risk of both vascular and gastrointestinal events can be predicted once the baseline risks of such hazards are known, which could help clinical decision-making.”

Naproxen’s lack of harm may stem from the fact that the high-dose “regimen is capable of producing COX-1 inhibition that is sufficiently prolonged and intense to result in platelet inhibition in some individuals, which would attenuate any adverse vascular events of COX-2 inhibition,” the researchers note.

They caution that the results for naproxen should be interpreted cautiously for several reasons, including uncertainty over the effects of concomitant aspirin and lower doses and longer durations of the drug. Moreover, they emphasize that naproxen still comes with a risk of upper GI complications, though “such bleeds are less likely than vascular events to result in disability and such hazards could be mitigated with proton-pump inhibitors [PPIs].”

But in an editorial accompanying the paper, Marie R. Griffin, MD, MPH, of Vanderbilt University Medical Center (Nashville, TN), stresses that “clinical trials are not the whole story.

“Indeed observational studies often identify and quantify important safety concerns missed by clinical trials of short duration, done in select and small populations,” she continues, adding, “Baigent and colleagues’ meta-analysis offers considerable certainty about relative and absolute major vascular risks of high doses of the most commonly prescribed NSAID doses, but leaves large gaps about risks associated with lower NSAID doses, longer durations of use, and residual effects after stopping treatment.”

Getting the Word Out

Kishore J. Harjai, MD, of Columbia University Medical Center (New York, NY), told TCTMD in an e-mail communication and telephone interview that the meta-analysis confirms the risks that have been shown for NSAIDs in the past. Notably, it also dispels the notion that naproxen might actually be protective against vascular events.

“Yet, it is important to realize that the study population appears to be more representative of patients we would see in a generalist, rather than cardiologist, practice,” Dr. Harjai said, noting that only a fifth of patients were on aspirin while diabetes and atherosclerosis were rare at 9% each. “So, the relevance of these findings to a cardiologist’s patients or to a post-PCI or post-MI patient is somewhat limited.”

Deepak L. Bhatt, MD, MPH, of Brigham and Women’s Hospital (Boston, MA), said the meta-analysis is noteworthy for several reasons. “It’s published in a prestigious journal. It’s an enormous study, . . . and it provides some precision as far as the risk associated with this class of drugs. Those are all valuable contributions,” he commented.

While most cardiologists are already aware of the vascular risks associated with high-dose NSAIDs, other clinicians and their patients may not be, Dr. Bhatt said. For example, most NSAIDs are prescribed by primary care physicians or bought over-the-counter. Not only does this meta-analysis help get out the word to a broader audience, but “it solidifies prior signals,” he noted.

Proceed with Caution

Healthy people taking an occasional NSAID should not worry, Dr. Bhatt added. “But for an 80-year-old patient with prior heart attacks and strokes who is taking daily NSAIDs the risk is substantially higher.” Moreover, there are other risks to high-dose NSAIDS such as GI bleeding, hypertension, fluid retention, and exacerbation of heart failure, he said. “So, overall the message is to think twice. If a patient doesn’t have to be on an NSAID, then don’t put them on one.”

Dr. Harjai suggested choosing naproxen for patients at high vascular risk who need long-term NSAIDs and adding a PPI if they are also at high risk of GI complications. “On the other hand, in patients at low vascular risk and high GI risk, I would consider using either celecoxib or diclofenac in conjunction with a PPI agent. And judicious use of acetaminophen, narcotics, and nonpharmacologic measures may be helpful,” he said.

In the editorial, Dr. Griffin advises that given the known risks of high-dose NSAIDs, clinicians should employ alternatives like evidence-based nonpharmacological treatments, topical treatments, and NSAID regimens that minimize risk. “Identification of safe and effective strategies for chronic pain is sorely needed,” she concludes. “In the meantime, long-term use of high-dose NSAIDs should be reserved for those who receive considerable symptomatic benefit from the treatment and understand the risks.” 

Study Details

Mean age at randomization was 61 years. Approximately two-thirds of patients were female, and 79% were white. The indication for NSAID treatment was rheumatoid arthritis or osteoarthritis in about 80% of patients.



  1. Coxib and traditional NSAID Trialists’ (CNT) Collaboration. Vascular and upper gastrointestinal effects of non-steroidal anti-inflammatory drugs: Meta-analyses of individual participant data from randomised trials. Lancet. 2013;Epub ahead of print.
  2. Griffin MR. High-dose non-steroidal anti-inflammatories: Painful choices. Lancet. 2013;Epub ahead of print.

Related Stories:

  • Drs. Baigent, Griffin, and Harjai report no relevant conflicts of interest.
  • Dr. Bhatt reports receiving research grants from Amarin, AstraZeneca, Bristol-Myers Squibb, Eisai, Ethicon, Medtronic, Sanofi-Aventis, and The Medicines Company.