REVIEWS
ORAL ANTICOAGULATION RELATED INTRACEREBRAL
HEMORRHAGE: MORE QUESTIONS THAN ANSWERS
Elena Terecoasa1, Cristina Tiu1, Nuria Huertas2, Maria Alonso de Leciñana3
1
Neurology Department, Emergency University Hospital, Bucharest, Romania
2
Neurology Department, Severo Ochoa University Hospital, Madrid, Spain
3
Neurology Department, Ramon y Cajal University Hospital, Madrid, Spain
ABSTRACT
Intracranial hemorrhage (ICH) is the most feared and devastating complication of oral anticoagulant therapy. When
an ICH occurs, the patient’s situation hinges on the balance between how great is the embolic risk while not receiv-
ing anticoagulants, and how big is the threat of the hemorrhage if the anticoagulant effect is not reversed prompt-
ly. Although several studies which compared the use of different reversal agents failed to demonstrate any im-
provement in prognosis and survival, at the present moment the consensus seem to be that anticoagulation should
be rapidly reversed after an ICH. The second question to be answered is whether and when should be oral antico-
agulation treatment restarted. Although the risk of thromboembolism in patients off anticoagulation seems to be
higher than the risk of ICH recurrence, there is a marked paucity of prospective large studies on the real risk of ICH
recurrence when OAC is resumed, paucity that probably emphasizes the ethical challenge of prescribing patients
a medication to which they have an apparent contraindication. The little evidence available suggests that the opti-
mal time for resumption is between 10 days and 30 weeks.
Key words: intracerebral hemorrhage, oral anticoagulation, re-bleeding, atrial fibrillation,
reversal of anticoagulation, resumption of anticoagulation
Intracranial hemorrhage (ICH) is the most feared management of atrial fibrillation as a result of the
and devastating complication of anticoagulant RE-LY trial (2). Dabigatran can be administrated
treatment, leading to death or disability in two either 110 mg twice a day or 150 mg twice a day,
thirds of cases (1). the lower dose being non-inferior to warfarin but
Vitamin K antagonists (VKA, warfarin and having fewer hemorrhagic complications whereas
acenocumarol) have long been the mainstay of an- the higher dose is superior in terms of thrombopro-
ticoagulation therapy in atrial fibrillation. A strenu- phylaxis with the cost of a slightly higher incidence
ous effort has been made for many years in order to of bleeding, but not exceeding warfarin bleeding
develop new oral anticoagulants as effective as rate. The category of factor Xa antagonists is repre-
VKA but with a superior pharmacological profile. sented at the moment by two drugs: rivaroxaban
For the moment there are two directions of research: and apixaban, both of them being in Phase III of
the direct inhibitors of thrombin and the Xa factor development. In the double-blind randomized
antagonists. Unlike VKA, dietary restrictions and ROCKET-AF trial (3), based on a population of
frequent blood sampling to monitor the degree of over 14,000 patients with atrial fibrillation, rivar-
anticoagulation are unnecessary with the currently oxaban (20 mg once a day) was non-inferior to
available new agents. warfarin in terms of stroke and non-central nervous
The sole representative of the former category is system embolism prevention and had a lower rate
dabigatran etexilate, which was recently included of ICH but a higher risk of gastrointestinal bleed-
in both the ESC and AHA/ACC Guidelines for the ings. At the end of last year rivaroxaban was ap-
Author for correspondence:
Elena Terecoasa, Neurology Department, Emergency University Hospital, Splaiul Independentei, No. 169, Bucharest, Romania
e-mail:
ROMANIAN JOURNAL OF NEUROLOGY – VOLUME XI, NO. 1, 2012 13
, 14 ROMANIAN JOURNAL OF NEUROLOGY – VOLUME XI, NO. 1, 2012
proved for the prevention of non-valvular atrial fi- more effective prophylaxis of thromboembolic
brillation-related stroke and systemic embolism in events, the methods of recognition of those patients
USA and Europe. Apixaban has completed to the with a high risk of bleeding must be improved and
date two major trials. In the AVERROES trial (4), the therapeutic attitude carefully weighted.
which included almost 6,000 patients unsuitable
for VKA treatment, apixaban was significantly su- RISK FACTORS FOR OAC-RELATED ICH
perior to aspirin in terms of reducing the risk of
stroke and systemic embolism with no increase of Many factors contribute to the variability of the
major bleeding or ICH rates. The results of ARIS- anticoagulation effect and accordingly, to the risk
TOTLE trial (5), a double-blind randomized trial of bleeding. OAC – drug interactions have been
which included over 18,000 patients, were also shown to increase the risk of serious bleeding (14).
made public last year. In this study apixaban was Warfarin interaction with at least one drug was con-
found superior to warfarin for the prevention of sidered in a retrospective study recently published
thromboembolism in patients with atrial fibrilla- (15) the main contributor to bleeding in almost half
tion. Besides, apixaban had a lower rate of major of the cases.
bleeding and of all-cause mortality. To the date Non-steroidal anti-inflammatory drugs, lipid-
apixaban is under reviewing for approval. lowering drugs, acetaminophen, selective serotonin
The enthusiasm for these agents, however, must reuptake inhibitors, amiodarone, omeprazole, anti-
be tempered by two notables concerns: there are no fungal agents and cimetidine are some of the wide-
readily available means for assessing the degree of ly used drugs prone to interact with oral anticoagu-
anticoagulation and there is no readily available a lants (warfarin or acenocoumarol). The main
reversal strategy. Currently, the only reversal op- mechanism of interaction implies the inhibition of
tion for dabigatran is emergency dialysis which can CYP2C9, which increases the plasma concentra-
be a challenge when it comes to a patient with a tion of the drug and leads to stronger anticoagula-
threatening ICH. tion effect and higher bleeding risk. Several studies
The new oral anticoagulants-related ICHs may have shown that antiplatelet drugs raise the bleed-
have different epidemiology, mechanism, charac- ing risk associated with anticoagulation therapy.
teristics, acute and long term management from the A number of predisposing factors for cerebral
vitamin K antagonists-related ICHs and are beyond hemorrhage in patients receiving OAC has been
the purpose of this article. This subject will be re- identified. These factors are summarized in Table 1
viewed in a future paper. (16) (9) (17).
Further on we will refer only to cerebral hemor-
rhage related to oral anticoagulation with vitamin Table 1. Risk factors for ICH during oral anticoagulation
K antagonists. Established:
Advanced age (especially > 75 years)
Hypertension (especially systolic blood pressure >
EPIDEMIOLOGY 160 mmHg)
History of cerebrovascular disease
Approximately 1% of the european population
Intensity of anticoagulation
is currently receiving oral anticoagulation treat-
Possible:
ment (OAC) with vitamin K antagonists, and this Concomitant use of aspirin
proportion has increased to 1.7% in some countries Increased variation of INR
(6) (7). The rate of ICH in patients undergoing Cerebral amyloid angiopathy
long-term oral anticoagulation is about 2-9 per Tobacco smoking
100,000 population/year, an incidence 7- to 11- fold Heavy alcohol consumption
higher than in the not treated population of similar Diabetes
age (8) (9) (10) (11). About 5-12% of ICH are re- Serious heart disease
lated to OAC (12) (9). The incidence of OAC-relat- Liver disease
ed ICH is increasing and this continuous increment Malignancy
can be explained by the larger number of elderly Imaging and genetic markers:
patients that receive OAC, the increased use of Leukoaraiosis detected by brain CT/MRI
Microbleeds by T2*-weighted MRI
combined anticoagulant and antiplatelet regimens,
APO II or IV genotype
or the increasing use of OAC for secondary stroke
CT: computed tomography; MRI magnetic resonance
prevention (13). Although the greater number of imaging
bleeding complications is counterbalanced by a
ORAL ANTICOAGULATION RELATED INTRACEREBRAL
HEMORRHAGE: MORE QUESTIONS THAN ANSWERS
Elena Terecoasa1, Cristina Tiu1, Nuria Huertas2, Maria Alonso de Leciñana3
1
Neurology Department, Emergency University Hospital, Bucharest, Romania
2
Neurology Department, Severo Ochoa University Hospital, Madrid, Spain
3
Neurology Department, Ramon y Cajal University Hospital, Madrid, Spain
ABSTRACT
Intracranial hemorrhage (ICH) is the most feared and devastating complication of oral anticoagulant therapy. When
an ICH occurs, the patient’s situation hinges on the balance between how great is the embolic risk while not receiv-
ing anticoagulants, and how big is the threat of the hemorrhage if the anticoagulant effect is not reversed prompt-
ly. Although several studies which compared the use of different reversal agents failed to demonstrate any im-
provement in prognosis and survival, at the present moment the consensus seem to be that anticoagulation should
be rapidly reversed after an ICH. The second question to be answered is whether and when should be oral antico-
agulation treatment restarted. Although the risk of thromboembolism in patients off anticoagulation seems to be
higher than the risk of ICH recurrence, there is a marked paucity of prospective large studies on the real risk of ICH
recurrence when OAC is resumed, paucity that probably emphasizes the ethical challenge of prescribing patients
a medication to which they have an apparent contraindication. The little evidence available suggests that the opti-
mal time for resumption is between 10 days and 30 weeks.
Key words: intracerebral hemorrhage, oral anticoagulation, re-bleeding, atrial fibrillation,
reversal of anticoagulation, resumption of anticoagulation
Intracranial hemorrhage (ICH) is the most feared management of atrial fibrillation as a result of the
and devastating complication of anticoagulant RE-LY trial (2). Dabigatran can be administrated
treatment, leading to death or disability in two either 110 mg twice a day or 150 mg twice a day,
thirds of cases (1). the lower dose being non-inferior to warfarin but
Vitamin K antagonists (VKA, warfarin and having fewer hemorrhagic complications whereas
acenocumarol) have long been the mainstay of an- the higher dose is superior in terms of thrombopro-
ticoagulation therapy in atrial fibrillation. A strenu- phylaxis with the cost of a slightly higher incidence
ous effort has been made for many years in order to of bleeding, but not exceeding warfarin bleeding
develop new oral anticoagulants as effective as rate. The category of factor Xa antagonists is repre-
VKA but with a superior pharmacological profile. sented at the moment by two drugs: rivaroxaban
For the moment there are two directions of research: and apixaban, both of them being in Phase III of
the direct inhibitors of thrombin and the Xa factor development. In the double-blind randomized
antagonists. Unlike VKA, dietary restrictions and ROCKET-AF trial (3), based on a population of
frequent blood sampling to monitor the degree of over 14,000 patients with atrial fibrillation, rivar-
anticoagulation are unnecessary with the currently oxaban (20 mg once a day) was non-inferior to
available new agents. warfarin in terms of stroke and non-central nervous
The sole representative of the former category is system embolism prevention and had a lower rate
dabigatran etexilate, which was recently included of ICH but a higher risk of gastrointestinal bleed-
in both the ESC and AHA/ACC Guidelines for the ings. At the end of last year rivaroxaban was ap-
Author for correspondence:
Elena Terecoasa, Neurology Department, Emergency University Hospital, Splaiul Independentei, No. 169, Bucharest, Romania
e-mail:
ROMANIAN JOURNAL OF NEUROLOGY – VOLUME XI, NO. 1, 2012 13
, 14 ROMANIAN JOURNAL OF NEUROLOGY – VOLUME XI, NO. 1, 2012
proved for the prevention of non-valvular atrial fi- more effective prophylaxis of thromboembolic
brillation-related stroke and systemic embolism in events, the methods of recognition of those patients
USA and Europe. Apixaban has completed to the with a high risk of bleeding must be improved and
date two major trials. In the AVERROES trial (4), the therapeutic attitude carefully weighted.
which included almost 6,000 patients unsuitable
for VKA treatment, apixaban was significantly su- RISK FACTORS FOR OAC-RELATED ICH
perior to aspirin in terms of reducing the risk of
stroke and systemic embolism with no increase of Many factors contribute to the variability of the
major bleeding or ICH rates. The results of ARIS- anticoagulation effect and accordingly, to the risk
TOTLE trial (5), a double-blind randomized trial of bleeding. OAC – drug interactions have been
which included over 18,000 patients, were also shown to increase the risk of serious bleeding (14).
made public last year. In this study apixaban was Warfarin interaction with at least one drug was con-
found superior to warfarin for the prevention of sidered in a retrospective study recently published
thromboembolism in patients with atrial fibrilla- (15) the main contributor to bleeding in almost half
tion. Besides, apixaban had a lower rate of major of the cases.
bleeding and of all-cause mortality. To the date Non-steroidal anti-inflammatory drugs, lipid-
apixaban is under reviewing for approval. lowering drugs, acetaminophen, selective serotonin
The enthusiasm for these agents, however, must reuptake inhibitors, amiodarone, omeprazole, anti-
be tempered by two notables concerns: there are no fungal agents and cimetidine are some of the wide-
readily available means for assessing the degree of ly used drugs prone to interact with oral anticoagu-
anticoagulation and there is no readily available a lants (warfarin or acenocoumarol). The main
reversal strategy. Currently, the only reversal op- mechanism of interaction implies the inhibition of
tion for dabigatran is emergency dialysis which can CYP2C9, which increases the plasma concentra-
be a challenge when it comes to a patient with a tion of the drug and leads to stronger anticoagula-
threatening ICH. tion effect and higher bleeding risk. Several studies
The new oral anticoagulants-related ICHs may have shown that antiplatelet drugs raise the bleed-
have different epidemiology, mechanism, charac- ing risk associated with anticoagulation therapy.
teristics, acute and long term management from the A number of predisposing factors for cerebral
vitamin K antagonists-related ICHs and are beyond hemorrhage in patients receiving OAC has been
the purpose of this article. This subject will be re- identified. These factors are summarized in Table 1
viewed in a future paper. (16) (9) (17).
Further on we will refer only to cerebral hemor-
rhage related to oral anticoagulation with vitamin Table 1. Risk factors for ICH during oral anticoagulation
K antagonists. Established:
Advanced age (especially > 75 years)
Hypertension (especially systolic blood pressure >
EPIDEMIOLOGY 160 mmHg)
History of cerebrovascular disease
Approximately 1% of the european population
Intensity of anticoagulation
is currently receiving oral anticoagulation treat-
Possible:
ment (OAC) with vitamin K antagonists, and this Concomitant use of aspirin
proportion has increased to 1.7% in some countries Increased variation of INR
(6) (7). The rate of ICH in patients undergoing Cerebral amyloid angiopathy
long-term oral anticoagulation is about 2-9 per Tobacco smoking
100,000 population/year, an incidence 7- to 11- fold Heavy alcohol consumption
higher than in the not treated population of similar Diabetes
age (8) (9) (10) (11). About 5-12% of ICH are re- Serious heart disease
lated to OAC (12) (9). The incidence of OAC-relat- Liver disease
ed ICH is increasing and this continuous increment Malignancy
can be explained by the larger number of elderly Imaging and genetic markers:
patients that receive OAC, the increased use of Leukoaraiosis detected by brain CT/MRI
Microbleeds by T2*-weighted MRI
combined anticoagulant and antiplatelet regimens,
APO II or IV genotype
or the increasing use of OAC for secondary stroke
CT: computed tomography; MRI magnetic resonance
prevention (13). Although the greater number of imaging
bleeding complications is counterbalanced by a
