Background: The frequency of use of warfarin anticoagulation increases significantly in the elderly population. It remains controversial whether this puts these patients at increased risk for hemorrhagic complications after trauma.
Methods: We prospectively evaluated consecutive trauma patients who were taking warfarin and compared their outcomes to a group of age-matched patients with head injuries but not taking warfarin.
Results: One hundred fifty-nine trauma patients on warfarin were evaluated, 94 (59%) with some type of head trauma; 25 of these 94 patients (27%) had documented intracranial trauma. Fifteen patients died (9.4%); they had an international normalized ratio of 3.3 +/- 1.6 versus 3.0 +/- 2.1 for survivors in the warfarin group (p = 0.585). Twelve deaths were in the group of 25 patients with intracranial injuries (48%). Three patients without head injury died (5%) of other causes not related to warfarin or hemorrhage at a mean of 13 days after admission. Ten of 12 patients on warfarin with intracranial injuries who died had documented loss of consciousness (LOC); two patients who died secondary to an isolated intracranial injury had no LOC. Of 70 age-matched patients with head trauma not taking warfarin, 47 (67%) had intracranial injury and 5 of these died (10%) (p < 0.001 for both values compared with study patients). There were no significant differences for patients with intracranial injury comparing those on warfarin and those who were not in terms of age, gender, mechanism of injury, Injury Severity Score, or Glasgow Come Scale score.
Conclusion: We conclude that the preinjury use of warfarin does not place the trauma patient at increased risk for fatal hemorrhagic complications in the absence of head trauma. Furthermore, the presence of a head trauma alone is not predictive of mortality. However, the presence of intracranial injury is strongly associated with a mortality rate that is significantly higher than patients with head trauma who are not taking warfarin. LOC is also associated with mortality, but the absence of loss of consciousness does not reliably indicate the absence of intracranial injury or risk of death.