Concomitant use of nonsteroidal anti-inflammatory drugs (NSAIDs) and oral anticoagulants in patients with atrial fibrillation (AF) increases the risk for major bleeding and stroke, according to a new analysis of the RE-LY trial. "Avoid NSAIDs in anticoagulated patients with AF," senior author Michael Ezekowitz, MD, PhD, Thomas Jefferson University, Philadelphia, Pennsylvania, told theheart.org | Medscape Cardiology. "In this post hoc analysis of the RE-LY study, we found that concomitant NSAID use with any anticoagulant increases extracranial risk particularly, as well as stroke risk in patients with AF," he said of the study, published July 9 in the Journal of the American College of Cardiology. In an accompanying editorial, Sam Schulman, MD, PhD, McMaster University, Hamilton, Ontario, Canada, and James Aisenberg, MD, Icahn School of Medicine at Mount Sinai, New York, New York, write that NSAIDs along with anticoagulants represent "double trouble." "This study shows that both with warfarin and the newer anticoagulant dabigatran, you get more bleeding, but you can also get more thrombotic complications when you add an NSAID," Schulman told theheart.org | Medscape Cardiology. "This is a very common scenario, because anticoagulants are taken by patients who have AF and those patients are typically elderly, and often have osteoarthritis and need to take something for their osteoarthritis," he said. Nonselective NSAIDs, including ibuprofen, naproxen, meloxicam, diclofenac, and ketorolac, all have the ability to increase the risk for bleeding and were permitted for use in RE-LY. "Thus, we decided to do this post hoc analysis to test whether these drugs, when used with warfarin and dabigatran, carried an additional bleeding or clotting risk," said Ezekowitz, also co-principal investigator of the RE-LY trial. Results of the parent study in 18,113 patients with AF showed that dabigatran (Pradaxa, Boehringer Ingelheim) at the 110-mg or 150-mg dose twice daily was noninferior to warfarin for stroke or systemic embolism, although myocardial infarction (MI) risk was significantly increased with the 150-mg dose. The new analysis, led by Anthony P. Kent, MD, Yale New Haven Health, Bridgeport, Connecticut, involved 2279 of the participants, who used NSAIDs at least once during the study period. Patients taking NSAIDs with an oral anticoagulant (OAC) had significantly higher rates of major bleeding than those who did not use NSAIDs (5.4% vs 3.2%; hazard ratio [HR], 1.68; 95% confidence interval [CI], 1.40 - 2.02). The finding was present across all OAC treatment groups. NSAID use was also associated with significantly more major gastrointestinal bleeding (HR, 1.81; 95% CI, 1.35 - 2.43) and more frequent hospitalizations (HR, 1.64; 95% CI, 1.51 - 1.77). Rates of MI and all-cause mortality were similar with and without NSAID use. In a time-varying covariate analysis, however, the risk for stroke or systemic embolism was significantly elevated with NSAID use (HR, 1.50; 95% CI, 1.12 - 2.01). While ischemic stroke was significantly elevated with NSAID use (HR, 1.55; P = .0102), hemorrhagic stroke rates were similar (HR, 1.08; P = .8631). In an interaction analysis, NSAID use did not alter the ability of the dabigatran 150-mg or 110-mg dose to prevent stroke/systemic embolism relative to warfarin (P for interaction = .59 and .54, respectively). The authors cite as study limitations the patients' ability to periodically start and stop NSAIDs; a lack of data regarding the specific type of nonselective NSAID, dose, and reason for NSAID use; and failure to capture baseline prevalence of osteoarthritis (OA), rheumatoid arthritis, or other inflammatory conditions that might partially explain NSAID use. The authors also point out that future research is still required to investigate the effects of NSAIDs when used with direct OACs in patients with AF. "Doctors should tell their patients with AF who are receiving anticoagulant therapy not to use over-the-counter pain killers," Ezekowitz said. He suggested that Tylenol (acetaminophen) could be an alternative but cautioned, "We have to be careful regarding the overuse of opioids." Double Trouble, No Single Way Out Telling patients to take Tylenol is not a solution, contends Schulman. "Most patients say Tylenol does nothing for my osteoarthritis, and that's because it's not an anti-inflammatory agent. There is an inflammatory component in arthritis and that is what needs to be treated," he said. NSAIDs are known to increase bleeding risk, especially in the elderly, but the bleeding was largely thought to occur in the stomach. Now, there is increasing evidence that bleeding occurs in the intestines, Schulman said. What was surprising in the study was the fact that very few patients who were taking NSAIDs were receiving a proton-pump inhibitor, he noted. That said, "This only protects the stomach and does not afford protection against intestinal bleeding, so it's not a panacea." Until safe and effective non-NSAID, nonopioid analgesics are available, "We are stuck with the high-stakes, common question of how to manage patients with AF and OA," Schulman said. Doctors should consider using a cyclooxygenase-2 inhibitor such as celecoxib (Celebrex, Pfizer), which is associated with less bleeding and does not decrease platelet function but still has anti-inflammatory properties, he suggested. Source
