Akira Tempaku1. 1. Department of Neurosurgery, Hokuto Hospital, Japan.
Abstract
Objective: Oral anticoagulants are widely administered to patients with atrial fibrillation in order to prevent the onset of cardiogenic embolisms. However, intracranial bleeding during anticoagulant therapy often leads to fatal outcomes. Accordingly, the use of novel oral anticoagulants (NOACs), which less frequently have intracranial bleeding as a complication, is expanding. A nationwide survey of intracranial bleeding and its prognosis in Japan reported that intracranial bleeding of advanced severity was not common after NOAC administration. In this report, two cases from our institute are presented. Patients: Case 1 was an 85-year-old man with a right frontal lobe hemorrhage while under dabigatran therapy. Case 2 was an 81-year-old man who had cerebellar hemorrhage while under rivaroxaban therapy. Result: In both patients, the clinical course progressed without aggravation of bleeding or neurological abnormalities once anticoagulant therapy was discontinued. Conclusion: These observations suggest that intracranial hemorrhage during NOAC therapy is easily controlled by discontinuation of the drug. NOAC administration may therefore be appropriate despite the risk of such severe complications. Further case studies that include a subgroup analysis with respect to each NOAC or patient background will be required to establish appropriate guidelines for the prevention of cardiogenic embolisms in patients with atrial fibrillation.
Objective: Oral anticoagulants are widely administered to patients with atrial fibrillation in order to prevent the onset of cardiogenic embolisms. However, intracranial bleeding during anticoagulant therapy often leads to fatal outcomes. Accordingly, the use of novel oral anticoagulants (NOACs), which less frequently have intracranial bleeding as a complication, is expanding. A nationwide survey of intracranial bleeding and its prognosis in Japan reported that intracranial bleeding of advanced severity was not common after NOAC administration. In this report, two cases from our institute are presented. Patients: Case 1 was an 85-year-old man with a right frontal lobe hemorrhage while under dabigatran therapy. Case 2 was an 81-year-old man who had cerebellar hemorrhage while under rivaroxaban therapy. Result: In both patients, the clinical course progressed without aggravation of bleeding or neurological abnormalities once anticoagulant therapy was discontinued. Conclusion: These observations suggest that intracranial hemorrhage during NOAC therapy is easily controlled by discontinuation of the drug. NOAC administration may therefore be appropriate despite the risk of such severe complications. Further case studies that include a subgroup analysis with respect to each NOAC or patient background will be required to establish appropriate guidelines for the prevention of cardiogenic embolisms in patients with atrial fibrillation.
Anticoagulants are effective in preventing cerebral embolisms that may complicate atrial
fibrillation[1],[2],[3]). Conventionally, warfarin has been the only oral
anticoagulant used. However, its administration is associated with problems, such as the
adjustments needed to maintain an effective blood concentration. In addition, dietary
restrictions are necessary, because vitamin K levels influence warfarin’s efficacy. In
recent years, novel oral anticoagulants (NOACs) have been developed. Those have demonstrated
encouraging outcomes, as well as effects comparable to those of warfarin[4],[5],[6],[7]).
Further a lower reported incidence of intracranial and other massive bleeding[8],[9],[10]). NOACs are increasingly being used for primary and secondary
prevention with respect to cardiogenic embolisms. With the aging of society, elderly
patients suffering from atrial fibrillation make up a growing population. As a result, the
prevention of cardiogenic embolisms is becoming a major issue for social and health
economics.Unfortunately, intracranial bleeding is a serious complication of anticoagulant treatment.
Surgical treatments such as hematoma removal may be required in some cases. Since warfarin
has a long half-life, controlling intracranial bleeding is very difficult. There are some
reports that the incidence of intracranial bleeding after NOAC administration is not as
frequent as in patients treated with warfarin[4],[5],[6],[7]).
Other reports indicate that the clinical course following intracranial hemorrhage is better
in NOACpatients than in warfarinpatients[11], [12]). We investigated the progress and prognosis of cases at our
institute in which conservative therapy was selected, with intracranial bleeding as a
complication during NOAC administration.
Materials and Methods
NOACs were administered to 313 patients (dabigatran: 173, rivaroxaban: 140) at our
institute between 2011 and July 2014. All patients were diagnosed with non-valvular atrial
fibrillation and NOAC medication was started for the prevention of cardiogenic embolization.
Patients were allocated to dabigatran or rivaroxaban on the basis of their application time,
medication compliance, and other factors. Random assignment was not performed. Among these
patients, intracranial bleeding occurred in 2 cases, which are described below.
Case Presentations
Case 1
An 85-year-old man with a history of high-blood pressure, diabetes, and episodic atrial
fibrillation (cardiac pacemaker) was staying at a health and welfare institution for the
elderly because of dementia. He was taking a hypoglycemic agent, β-inhibitor,
anti-dementia drug, aspirin (100 mg per day), and dabigatran (220 mg per day). He was
transported to our institute after the occurrence of left hemiparesis. On initial
examination, his level of consciousness was 10 points on the Japan Coma Scale (JCS 10) and
pupil diameters were equal at 2.5 mm bilaterally; however, right conjugate deviation was
observed. Paralysis was observed in the left upper and lower limbs, and he scored 3/5 on a
manual muscle test. His renal function indicated a creatinine clearance (Ccr) of 49.6
mL/min. An evaluation of the risk factors for cerebral stroke gave 3 points on the
congestive heart failure, hypertension, age, diabetes mellitus, and stroke/TIA
(CHADS2) scale[13]), and 4 points on the congestive heart failure,
hypertension, age, diabetes mellitus, stroke/TIA, vascular disease, age and sex category
(CHA2DS2-VASc) scale[14]), making him eligible for anticoagulant therapy. At the same
time, his risk factors for bleeding during anticoagulant therapy produced a hypertension,
abnormal renal/liver function, stroke, bleeding history or predisposition, labile INR,
elderly, drug/alcohol use (HAS-BLED) [15], [16]) score of 2 points. Mild midline shift was observed on a
head computed tomography (CT) scan as a complication of subcortical bleeding in the right
frontal lobe (Figure 1A). Anticoagulant therapy (dabigatran) and antiplatelet therapy (aspirin) were
discontinued and conservative therapy was applied. The level of consciousness shifted to
JCS 3 without aggravation of the cerebral hemorrhage (Figure 1B). Rehabilitation was continued with no aggravation of the central
nervous system disorders. The patient was transferred to a nursing home 10 months after
the onset of cerebral hemorrhage and the discontinuation of anticoagulant therapy.
Figure 1
Case 1: the patient’s head computed tomography on admission (A) and 7 days later
(B).
Case 1: the patient’s head computed tomography on admission (A) and 7 days later
(B).
Case 2
An 81-year-old man with a history of cerebral infarction, diabetes, and atrial
fibrillation was treated with insulin and rivaroxaban (10 mg per day). A residual
disability caused by cerebral infarction left him requiring total assistance in his
day-to-day living. He was admitted because of torso tilt and vomiting. On initial
examination, his consciousness level was JCS 3 with no apparent immobility of the pupils,
deviation, or paralysis of the four limbs. Ccr was 39.3 mL/min. Regarding risk factors for
cerebral stroke, his CHADS2 score was 4 points and
CHA2DS2-VASc was 5 points. At the same time, the HAS-BLED score
was 2 points, with the risk of bleeding during anticoagulant therapy determined as
moderate. A head CT scan indicated a left cerebellar hemorrhage (Figure 2A). Anticoagulant therapy (rivaroxaban) was discontinued and conservative therapy was
applied. No increase in the amount of hematoma was observed (Figure 2B). The clinical course progressed without aggravation of
the neurological abnormalities. The patient was transferred to a geriatric health services
facility 3 months after the onset of cerebellar hemorrhage while still receiving no
anticoagulant therapy.
Figure 2
Case 2: the patient’s head computed tomography on admission (A) and 5 days later
(B).
Case 2: the patient’s head computed tomography on admission (A) and 5 days later
(B).
Results
Among the 313 patients treated with NOAC at our institute between 2011 and July 2014
(dabigatran: 173, rivaroxaban: 140) the average age was 74.6 years, with 96 patients aged up
to 70 years and 217 aged 71 years or older. There were 220 men and 93 women. Regarding
CHADS2, 54 patients scored 0–1 point, while 259 scored 2 points or more. The
CHA2DS2-VASc score was 0–1 point in 24 cases and 2 points or more in
289 cases. Serious complications requiring discontinuation of anticoagulant therapy occurred
in 9 cases, of which intracranial bleeding was observed in 2. In addition, recurrence of
cerebral infarction was observed in 8 cases. The intracranial bleeding involved hemorrhage
in the cerebral parenchyma (1 case of cerebral hemorrhage in the cerebrum and 1 case of
cerebellar hemorrhage), with no cases of subdural bleeding. Conservative therapy was
selected in both intracranial hemorrhage cases; no aggravation of hematoma was observed as a
result of the antihypertensive therapy, the administration of a hemostatic drug, and the
discontinuation of anticoagulants.
Discussion
I have described above two cases in which the administration of anticoagulants was
indicated (CHADS2 and CHA2DS2-VASc scores of 2 points or
more) and intracranial hemorrhage was observed as a complication during NOAC administration.
The bleeding risk in both cases was moderate (HAS-BLED score 2 points). However, no increase
in the quantity of hematoma or aggravation of the abnormal neurological findings were
observed as a result of conservative therapy. Whereas the half-life of warfarin in serum is
considered to be 20 to 60 hours, the half-lives of 12–17 hours for dabigatran and 5–9 hours
for rivaroxaban mean that the anticoagulant effect of NOAC is likely to disappear within a
very short period of time following withdrawal. In addition, in NOAC the mechanism of
coagulation inhibition and hemostasis differs from that of warfarin, being limited to a
single cause (Figure 3). Furthermore, the brain system is rich in tissue factor, which contributes to the
activation of the extrinsic clotting system due to factor VII[4], [8], [17],[18],[19],[20]). Unlike warfarin, NOAC never inhibit factor VII function. The
microenvironment of the hemorrhaged brain system contains abundant active factor VII in
patients taking NOAC. Therefore, the rapid functional restoration of the coagulation and
hemostasis mechanisms occurs in parallel with the decline in blood concentration once the
NOAC intake is stopped. Thus, the increase in hematoma size can be controlled by rapid
withdrawal alone, even after the onset of intracranial bleeding.
Figure 3
Mechanism of hemostasis and targets of warfarin and novel oral anticoagulants.
Mechanism of hemostasis and targets of warfarin and novel oral anticoagulants.A comparative outcome analysis has indicated that NOAC have a lower incidence of
complications such as massive bleeding or intracranial bleeding. Substantial anticoagulant
effects in preventing cerebral infarction and systemic embolisms have been
reported[4],[5],[6],[7],
[17], [21]). Although intracranial bleeding
during anticoagulant therapy is likely to become serious, no sudden increase in the hematoma
is observed, because of the short half-lives of NOAC in blood compared to that of warfarin.
In cases with a high anticoagulant therapy recommendation score (CHADS2 and
CHA2DS2-VASc), anticoagulant therapy is recommended even if the
bleeding risk is high. Currently, in Japan, the suitability of drugs is limited to
inhibiting the occurrence of ischemic cerebral stroke and systemic embolisms in patients
with non-valvular atrial fibrillation. Based on our results and a previous report[20]), NOAC are more likely than
warfarin to inhibit an increase in severity at the onset of hemorrhagic complications.The HAS-BLED score is usually used for the prediction of the risk of intracranial
hemorrhage during anticoagulation therapy. These two cases scored 2 points on the HAS-BLED
scale. This moderate risk had been predicted before the onset of intracranial hemorrhage.
Neither of these cases showed any increase in intracranial hematoma. However, a moderate
HAS-BLED score has not been established as a prognostic factor. The relationship between
HAS-BLED score and hematoma size or prognosis needs to be investigated by accumulating more
patient cases and performing a statistical analysis. Further study would be desirable.With the rapid aging of society, the number of elderly patients developing atrial
fibrillation as underlying disease is expected to increase. Accordingly, for the purposes of
primary and secondary prevention, the expansion of the range of diseases suitable for
anticoagulant therapy cannot be avoided. Against such a social background, it has been
predicted that cases of intracranial bleeding during anticoagulant therapy will continue to
increase in the future. Although the use of NOAC may be limited by social and health
economic backgrounds, along with the limitation of pharmaceutical registration, the types of
disease suitable for anticoagulant treatment are likely to increase. Since NOAC have the
advantage of not worsening the severity of serious complications, they may be useful in the
management of patients with atrial fibrillation.
Authors: Stuart J Connolly; Michael D Ezekowitz; Salim Yusuf; Paul A Reilly; Lars Wallentin Journal: N Engl J Med Date: 2010-11-04 Impact factor: 91.245
Authors: John W Eikelboom; Lars Wallentin; Stuart J Connolly; Mike Ezekowitz; Jeff S Healey; Jonas Oldgren; Sean Yang; Marco Alings; Scott Kaatz; Stefan H Hohnloser; Hans-Christoph Diener; Maria Grazia Franzosi; Kurt Huber; Paul Reilly; Jeanne Varrone; Salim Yusuf Journal: Circulation Date: 2011-05-16 Impact factor: 29.690
Authors: Robert G Hart; Hans-Christoph Diener; Sean Yang; Stuart J Connolly; Lars Wallentin; Paul A Reilly; Michael D Ezekowitz; Salim Yusuf Journal: Stroke Date: 2012-04-05 Impact factor: 7.914
Authors: Robert P Giugliano; Christian T Ruff; Eugene Braunwald; Sabina A Murphy; Stephen D Wiviott; Jonathan L Halperin; Albert L Waldo; Michael D Ezekowitz; Jeffrey I Weitz; Jindřich Špinar; Witold Ruzyllo; Mikhail Ruda; Yukihiro Koretsune; Joshua Betcher; Minggao Shi; Laura T Grip; Shirali P Patel; Indravadan Patel; James J Hanyok; Michele Mercuri; Elliott M Antman Journal: N Engl J Med Date: 2013-11-19 Impact factor: 91.245