Purposes: Intracranial haemorrhage (ICH) is a rare but potentially devastating complication of oral anticoagulants (OAC). This raises the difficult clinical choice between either permanent cessation of OAC, or continuing OAC and if so, when to restart. To make this choice, one needs to balance the thrombo-embolic risk after cessation of OAC against the risk of recurrent intracranial haemorrhage when OAC are restarted. There are few published data to base this difficult clinical decision on.
Methods: We present an observational study of a consecutive series of 108 patients, collected prospectively and admitted to our department, with an OAC-related intracranial haemorrhage, in whom we assessed the thrombotic event rate and the recurrent intracranial bleeding rate during follow-up.
Results: In the 25 patients in whom OAC were reinstituted no new thrombo-embolic events occurred (0/506 unprotected patient-days). In the group of patients in whom OAC were not restarted (n = 81), the thrombo-embolic event rate was 8/11590 unprotected patient-days, of which only 2 were cerebrovascular thrombo-embolisms. The overall risk of a thrombo-embolic complication can be estimated to be 0.66 events/1000 patient-days at risk (95% exact confidence limits of 0.3 to 1.3 events/1000 patient-days at risk). In three patients the thrombo-embolic event was fatal. We saw recurrent intracranial bleeding in eight patients, 2 of which were fatal. Seven of these occurred before the restarting of the OAC.
Conclusions: In OAC-related intracranial haemorrhages, OAC can be stopped safely for a considerable period, with a very low overall thrombotic event rate. The recurrent bleeding risk after restarting OAC is low. Recurrent bleeding mostly occurred before restarting OAC and is probably caused by insufficient or unsustained correction of the initial coagulation deficit. Immediate reversal of anticoagulation provides the patient with the best possible treatment options including surgery. OAC-related intracranial haemorrhages can therefore be actively treated.