In contrast to the previous Figure, this Figure shows the subsets of major bleeding.  For the 2 critical subsets, bleeding into a critical organ and fatal bleeding (light blue highlight in the Figure), both of these were lower in the rivaroxaban arm than in the warfarin arm.[112][116] These are similar to the data with dabigatran in RE-LY, except that the benefit for fatal bleeding with rivaroxaban is even more dramatic. The reduction in “bleeding into a critical organ” refers primarily to intracranial bleeding, and here again the reduction seen with rivaroxaban is similar to that seen with dabigatran.  Indeed, a lower likelihood of bleeding into the head seems to be feature with the NOACs.

 Reiffel JA. Am J Med 2013; 126: 00-00.