Bruce L Davidson, MD, writes… Our study published in JAMA Internal Medicine [reference below] used patient information (with no identifying information, so confidentiality was preserved) from the two EINSTEIN clinical studies of acute DVT and acute PE treatment to ask the questions, “What effect does taking aspirin, or non-steroidal anti-inflammatory drugs (NSAIDs) other than aspirin, have on bleeding risk if the patient is also taking a blood thinner (anticoagulant)?”
In the studies, the blood thinners tested were rivaroxaban (Xarelto) and enoxaparin (Lovenox) followed by long-term warfarin (Coumadin, Jantoven).
The results for NSAIDs were quite surprising to me. First, I didn’t realize that nearly a quarter (23%) of people were out there taking NSAIDs. This was the same proportion of people taking them in a large study of patients taking warfarin for irregular heart beat (atrial fibrillation). It turns out that people take NSAIDs for headaches, sore throat, fever, and muscle aches and pains. I thought people mostly took aspirin or (generic) Tylenol (acetaminophen) for that, but plenty of people around the world take ibuprofen (Motrin, Advil) or related drugs. Second, although I knew aspirin increased bleeding risk, I didn’t know that NSAIDs gave an even higher risk. For example, we found that the risk of a major bleed (a bleed in the brain, or another vital organ, or requiring transfusion, or a bleed that was fatal) was 2 ½ times higher if a patient on an anticoagulant was also taking an NSAID! And one-quarter of these major bleeds happened in the first 8 days of taking an NSAID—you don’t have to be taking them a long time. Most were not gastrointestinal bleeds, so taking antacids or other stomach-protecting drugs won’t reliably provide protection. The risk of a major bleed with aspirin while taking an anticoagulant was 50% higher (1.5 times higher) than while taking an anticoagulant without aspirin.
In absolute terms, the risk for NSAIDs was 5 major bleeds among 100 patients taking NSAID for a year while anticoagulated with rivaroxaban, compared to 8 major bleeds for warfarin. For aspirin, it was 3 among 100 patients taking the rivaroxaban/aspirin combination compared to 7 for the warfarin/aspirin combination. But neither of these differences, though they appear quite different for rivaroxaban and warfarin, was statistically significant. The bleeding risk for both anticoagulant drugs when NSAID or aspirin were added was similar.
Some people must take aspirin for arterial atherosclerosis-related disease, for example if they have had a heart attack or stroke. If you are among them, don’t stop taking it, but take care and don’t take NSAID. It is rare that an anticoagulated person MUST take an NSAID. Instead, take generic Tylenol (such as 1 ½ extra-strength caplets, 750 mg, every 6 hours) for fever, aches and pains, headaches, etc. Don’t exceed 3, or at most 4 grams per day (i.e. 6 to 8 tablets).
The message for patients: Discuss with your doctor and don’t take aspirin or NSAID while taking an anticoagulant blood thinner, unless you must.
Davidson, BL et al. Bleeding risk of patients with acute venous thromboembolism taking nonsteroidal anti-inflammatory drugs or aspirin. JAM Internal Medicine; April 14, 2014 (ePub ahead of print).
Bruce L Davidson, MD, MPH; Division of Pulmonary and Critical Care Medicine, University of Washington School of Medicine, Seattle.