Send to

Choose Destination
Acta Neurol Scand. 1995 Mar;91(3):208-14.

Intracranial traumatic and non-traumatic haemorrhagic complications of warfarin treatment.

Author information

Department of Neurosurgery, Karolinska Institute and Hospital, Stockholm, Sweden.


An increased referral of patients with anticoagulation related haemorrhages necessitated an analysis of causes and outcome of these complications in a patient group that reflected Swedish therapeutic traditions of anticoagulation treatment. Prospectively, all patients from Stockholm evaluated for warfarin related intracranial haemorrhage occurring during 1987 were analysed and their 6 month outcome recorded. Sixty-eight patients were included. The results of intracranial haemorrhagic complications were catastrophic with a 77% mortality rate. Their incidence was much higher than expected. Forty-one patients had non-traumatic intracerebral haematomas, 1 had a non-traumatic subarachnoid haemorrhage and 26 had traumatic injuries. In the 42 patients with non-traumatic haemorrhages, the indications for anticoagulation were cerebral ischaemic events in a majority (27/42). The remaining 15 patients had different indications for anticoagulation. They also had an increased frequency of hypertension (p < 0.05). In the 26 patients with traumatic haematomas, only 6/26 patients had previous cerebral ischaemic injuries (p < 0.01). Valvular heart prosthesis was their most common indication (11/26) for anticoagulation. Caution in instituting anticoagulation therapy in patients with hypertension or cerebrovascular disease, which is an important indication for anticoagulation in Sweden, is mandatory. Adherence to strict treatment regimens and their continuous reevaluation may help to avoid complications. The finding of more patients than expected with haemorrhagic complications is not compatible with the risk evaluations used to justify anticoagulation therapy in the patient groups studied. Clinical practice must have changed with time, showing that risk evaluations from controlled trials or retrospectively collected clinical data from selected patients are not necessarily applicable for long-term clinical practice.

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Wiley
Loading ...
Support Center