Concomitant NSAID, Antithrombotic Therapy Increases Bleeding, CV Events After MI
Among patients receiving antithrombotic therapy after MI, concomitant use of nonsteroidal anti-inflammatory drugs (NSAIDs) is linked to increased risk of bleeding and thrombotic events even with treatment as short as a few days, according to a study published in the February 24, 2015, issue of the Journal of the American Medical Association.
“More research is needed to confirm these findings; however, physicians should exercise appropriate caution when prescribing NSAIDs for patients who have recently experienced MI,” the study authors write.
Researchers led by Anne-Marie Schjerning Olsen, MD, PhD, of Copenhagen University Hospital Gentofte (Hellerup, Denmark), used Danish nationwide registries to study 61,971 patients 30 years and older (median age 67.7 years; 63.2% male) who were treated for a first-time MI and survived to 30 days postdischarge between 2002 and 2011. About 1 in 3 filled at least 1 NSAID prescription within 30 days; the only NSAID available over-the-counter was low-dose ibuprofen.
Over a median follow-up period of 3.5 years, 29.2% of patients died and 8.5% reported bleeding events, including 799 fatal bleeds and 2,157 GI bleeding events. Roughly one-third (30%) of patients experienced the combined cardiovascular endpoint (cardiovascular death, nonfatal recurrent MI, and ischemic stroke, TIA, or systemic arterial emboli).
Crude bleeding rates per 100 person-years were higher in those on concomitant NSAID treatment than in those who were not (4.2 vs 2.2). The same pattern was seen with the combined cardiovascular endpoint (11.2 vs 8.3). In adjusted analysis, NSAID use was associated with a doubling in the overall risk of bleeding and an increased risk of the cardiovascular endpoint as well. Event risks were heightened among all individual antithrombotic regimens (table 1) and NSAID groups—selective COX-2 inhibitors (rofecoxib and celecoxib) or nonselective COX inhibitors (ibuprofen and diclofenac).
Moreover, use of any NSAID compared with none was associated with substantial bleeding risk with both short-term (0-3 days; adjusted HR 3.37; 95% CI 2.57-4.41) and long-term (31-90 days; adjusted HR 1.67; 95% CI 1.32-2.11) treatment. These patterns remained for individual antithrombotic groups. GI bleeding, specifically, was more likely after NSAID treatment than without (HR 2.65; 95% CI 2.28-3.09).
Sensitivity analyses excluding patients who received NSAIDs before study inclusion and those with rheumatoid arthritis revealed bleeding risks comparable to the main analysis.
No Safe Therapeutic Window
The results show “no safe therapeutic window for concomitant NSAID use, because even short-term (0-3 days) treatment was associated with increased risk of bleeding compared with no NSAID use,” Dr. Schjerning Olsen and colleagues write. Also, they say, the findings confirm previous studies showing that NSAIDs are not associated with decreased cardiovascular risk.
“This is of considerable public health relevance because NSAIDs are among the most commonly used medications worldwide and any antithrombotic treatment invariably increases bleeding risk,” the authors write, adding that the fact ibuprofen is available without prescription and with no purchase amount limits in many parts of the world is “particularly worrying.”
While NSAID use is hard to avoid even in high-risk patients, they write, “these results highlight the importance of considering the balance of benefits and risks before initiating any NSAID treatment.”
According to the authors, future studies should focus on the GI effects of concomitant NSAID use with new oral anticoagulants.
Talk to Patients About NSAID Use
In an editorial accompanying the study, Charles L. Campbell, MD, of the University of Tennessee-Chattanooga (Chattanooga, TN), and David J. Moliterno, MD, of the University of Kentucky (Lexington, KY), express surprise at the adverse effects of short-term NSAID therapy. “This finding is important because it is established that bleeding events are strongly related to subsequent adverse cardiovascular events among patients with acute coronary syndromes,” they write.
They cite a similar study showing that patients with early bleeding complications have a 1.5 times higher mortality risk within 6 months, and although the mechanisms behind this are unclear, “it has been suggested that bleeding results in a heightened inflammatory state as well as interruption of antithrombotic therapies in some cases.”
Despite the limitations that accompany observational studies such as this, Drs. Campbell and Moliterno say, the “cumulative evidence available is an important reminder that while NSAIDs can be helpful and at times necessary medications for satisfactory quality of life, use of these medications among patients with a history of a recent MI is likely to be associated with clinically meaningful bleeding and ischemic risks.”
Moreover, “an even greater health care effect might occur in many countries, such as the United States, where… physicians may be unaware whether their patients are taking NSAIDs,” they observe.
Findings from the ongoing PRECISION trial, which has randomized about 24,000 patients with symptomatic rheumatoid arthritis or osteoarthritis and a high risk for or established cardiovascular disease to 1 of 3 NSAIDs, are expected next year, according to the editorial. Until then, Drs. Campbell and Moliterno conclude, “practitioners would do well to advise patients with cardiovascular disease against all NSAID use (except low-dose aspirin), especially patients with a recent acute coronary syndrome.”
1. Schjerning Olsen A-M, Gislason GH, McGettigan P, et al. Association of NSAID use with risk of bleeding and cardiovascular events in patients receiving antithrombotic therapy after myocardial infarction. JAMA. 2015;313:805-814.
2. Campbell CL, Moliterno DJ. Potential hazards of adding nonsteroidal anti-inflammatory drugs to antithrombotic therapy after myocardial infarction: time for more than a gut check [editorial]. JAMA. 2015;313:801-802.
- The study was funded by the Danish Council for Independent Research.
- Drs. Schjerning Olsen, Campbell, and Moliterno report no relevant conflicts of interest.