Bleeding risk with warfarin high among elderly, especially older than 80

Clinical Question

What is the risk of major hemorrhage among older patients taking warfarin?

Bottom Line

The risk of major hemorrhage among older patients taking warfarin is higher than commonly reported (13.7% during the first year for patients aged 80 and older) and particularly in the first 3 months of treatment. If the decision is made to use anticoagulation, patients should be aware of the risks, the early warning signs of bleeding, and should be followed up closely during the first 3 months in particular to assure that the international normalized ratio (INR) does not exceed 3.0. (LOE = 2b)


Hylek EM, Evans-Molina C, Shea C, Henault LE, Regan S. Major hemorrhage and tolerability of warfarin in the first year of therapy among elderly patients with atrial fibrillation. Circulation 2007;115:2689-2696.  [PMID:17515465]

Study Design

Cohort (prospective)




Outpatient (any)


Previous studies have found an average risk of bleeding of approximately 1% per year for patients taking warfarin for stroke prevention because of atrial fibrillation. However, studies often included patients who were already taking warfarin, thus missing the first few months when bleeding complications may be more common, and have not included many older patients. In this prospective study, the authors identified patients older than 65 years with newly diagnosed atrial fibrillation who were being started on warfarin. Of the entire cohort of 472 patients, 153 were 80 years or older, more than half were 75 years or older, and approximately half were women. Anticoagulation was managed by an on-site coagulation clinic; only 2% of INR measurements were greater than or equal to 4.0, consistent with the results for other clinical trials and probably better than the typical patient managed in routine practice. The primary outcome was the rate of major hemorrhage, defined as fatal bleeding, bleeding requiring hospitalization with transfusion, or bleeding involving a critical site like the brain, spine, eye, heart, or joint. We are not told exactly how these outcomes were measured (ie, patient recall or chart review). During the first year, the rate of major hemorrhage was 7.2 per 100 person-years (95% CI, 4.9 - 10.6) and the rate of intracranial hemorrhage was 2.5% (1.1 - 4.7). The subset of patients 80 years and older was at even greater risk: an impressive 13.1 major bleeding episodes per 100 person-years. The risk of bleeding was higher with increased age, with more time spent with an INR greater than or equal to 4.0, and during the first 90 days of therapy. It was also associated with a higher CHADS2 score (a risk score for stroke in patients with atrial fibrillation). Although 40% of patients were also taking aspirin, aspirin use did not increase the risk of major hemorrhage. Overall mortality and stroke incidence were not reported in this study.