What about warfarin and bleeding?  There is an impression that warfarin has an advantage over the NOACs because its anticoagulation effect can be counteracted in the case of a bleed.  I want to challenge this, as I have been studying warfarin during my entire career.  Figure 22 shows data from a study that we published 20 years ago, looking at every patient who came into our emergency room with an intracerebral hemorrhage while on warfarin.[15] 

All of these patients were treated aggressively.  They were given fresh frozen plasma or intravenous vitamin K, and yet despite doing everything we could do to stop this bleeding, about 30% to 40% of individuals experienced ongoing bleeding.  These statistics have not changed: cerebral hemorrhage on warfarin has a mortality rate of about 46%, and of those who survive, about 17% will have major neurological deficits, despite the fact that about two-thirds of these events occur when the INR is in the therapeutic range.  As discussed in Figure 3, this may be due to there simply not being enough factor VII in the brain, but whatever, this remains the most feared complication of warfarin therapy.