Benefits and Risks of Anticoagulation Resumption Following Traumatic Brain Injury

Study Questions:

What are the risks of thrombotic and hemorrhagic events associated with warfarin therapy following a traumatic brain injury (TBI) in older adults?


Using the Medicare data set of patients ages 65 years and older between 2006 and 2009, 10,782 patients who were prescribed warfarin within 1 month prior to a TBI event were identified. The patients were then divided into groups based on any exposure to warfarin following the TBI event. The primary outcomes were hemorrhagic and thrombotic event rates following a hospitalization for TBI.


Approximately 14% of the 105,432 patients who were discharged alive following a TBI event were prescribed warfarin prior to the TBI. Mean length of follow-up was 594.9 days. Atrial fibrillation was present in 82% of the population. In the first month following hospital discharge, 26% of patients received warfarin with 46% receiving warfarin for at least 1 month during the follow-up period. Unadjusted thrombotic event rates per 1,000 beneficiaries were 113.5 (95% confidence interval [CI], 102.9-125.2) for those using warfarin and 155.9 (95% CI, 143.5-169.3) for those not using warfarin. Unadjusted hemorrhagic event rates per 1,000 beneficiaries were 119.8 (95% CI, 108.9-131.8) for beneficiaries using warfarin and 85.7 (95% CI, 76.7-95.8) for those without warfarin use. In adjusted regression models, warfarin use was associated with a lower risk of thrombotic events (relative risk [RR], 0.77; 95% CI, 0.67-0.88) and an increased risk of hemorrhagic events (RR, 1.51; 95% CI, 1.29-1.78). Warfarin reduced the combined risk of thrombotic and hemorrhagic events (RR, 0.83; 95% CI, 0.72-0.96) without significant influence on the time between TBI hospital discharge and post-hospital warfarin exposure or limiting the analysis only to patients with atrial fibrillation.


The authors concluded that in patients hospitalized for a TBI, post-hospitalization use of warfarin is beneficial despite the increased risk of hemorrhagic events. The authors highlighted that 82% of patients treated with warfarin prior to a TBI had atrial fibrillation, just only 55% of participants resumed warfarin at any time following the TBI. The authors concluded that despite concerns about increased risk of bleeding in TBI patients, resumption of warfarin for appropriate indications (e.g., atrial fibrillation) following a TBI hospitalization is important.


This population-based cohort study reinforces the message that despite increased risks of hemorrhagic complications, most patients with atrial fibrillation would still benefit from anticoagulation therapy. While concerns about selection bias are well founded in this analysis, the findings are consistent with many other observational studies. As practitioners, we are driven to first and foremost ‘do no harm.’ However, we must remember that omission of therapy (e.g., anticoagulation in atrial fibrillation) can be even more harmful than electing to prescribe therapy to a patient at risk for adverse events. These data are a powerful tool to be used when discussing the risks and benefits of anticoagulation therapy with patients and their family members.

Clinical Topics: Anticoagulation Management

Keywords: Thromboembolism, Follow-Up Studies, Selection Bias, Warfarin, Confidence Intervals, Risk Assessment, Medicare

< Back to Listings