Fatal warfarin-associated intracranial hemorrhage in atrial fibrillation inpatients.

Ischemic stroke and major bleeding, mostly due to intracranial hemorrhage (ICH), cause about the same rates of death in pivotal randomized trials of direct oral anticoagulants (DOACs) versus warfarin for stroke prevention in atrial fibrillation (AF). We analyzed our AF inpatient database to determine whether any ICH-related deaths were potentially preventable. Among 5008 patients admitted to our institution between May 2008 and September 2014 with a diagnosis of AF, eight had fatal ICH between admission and 90 days follow-up. The mean age of these patients was 85 years; 62% were male. Localization of the ICH was intraparenchymal in 62% and subdural in 38%. CHA2DS2-VASc scores ranged from 4 to 7, and the HAS-BLED scores ranged from 3 to 7. Three of the eight fatal ICHs were directly due to falls. All 8 patients were taking warfarin. One was taking concomitant aspirin. At the time ICH was diagnosed, one patient had an INR of 5.4. Five patients had an INR within the target therapeutic range of 2.0-3.0, and two had an INR less than 2.0. After multivariate adjustment, a history of falls was the sole independent predictor of fatal ICH (OR 22.3; 95% CI 2.5-60.3). In conclusion, most patients had achieved the target INR at the time of ICH, and the primary precipitant of fatal ICH was often a fall. Using DOACs instead of warfarin and implementing structured fall-prevention programs in high-risk patients could further reduce mortality from ICH in AF.