Curr Cardiol Rep (2014) 16:480 DOI 10.1007/s11886-014-0480-9
NEW THERAPIES FOR CARDIOVASCULAR DISEASE (KW MAHAFFEY, SECTION EDITOR)
Intracranial Hemorrhage and Novel Anticoagulants for Atrial Fibrillation: What Have We Learned? Graeme J. Hankey
Published online: 20 March 2014 # Springer Science+Business Media New York 2014
Abstract Intracranial hemorrhage (ICH) affects 0.2-0.5 % of atrial fibrillation (AF) patients taking a novel oral anticoagulant (NOAC) each year. About two thirds of ICHs are intracerebral and one quarter subdural. The 30-day case fatality of NOAC-associated ICH was similar to that of warfarinassociated ICH in two trials. Consistent predictors of ICH are increasing age, a history of prior stroke or TIA, and concomitant use of an antiplatelet drug. Compared to warfarin, the NOACs significantly reduce the risk of ICH by half (risk ratio=0.44; 95 % CI: 0.37 to 0.51). Compared to aspirin, apixaban has a similar risk of ICH (risk ratio=0.84; 95 % CI, 0.38 to 1.87). Current treatments for NOAC-associated ICH include nonactivated and activated prothrombin complex concentrate, which reverse the anticoagulant effects of the NOACs, but their effects on bleeding and patient outcome are not known. Future treatments for NOACassociated ICH promise to include specific antidotes to dabigatran (e.g., aDabi-Fab, PER977) and factor Xa inhibitors (e.g., r-Antidote PRT064445, PER977).
Keywords Intracranial hemorrhage . Novel oral anticoagulants . Atrial fibrillation
This article is part of the Topical Collection on New Therapies for Cardiovascular Disease G. J. Hankey (*) School of Medicine and Pharmacology, The University of Western Australia, Room 222, Harry Perkins Institute of Medical Research QQ block, QE II Medical Centre, 6 Verdun Street, Nedlands 6009, Australia e-mail: [email protected] G. J. Hankey Department of Neurology, Sir Charles Gairdner Hospital, Nedlands, Perth, Australia 6009
Introduction Intracranial hemorrhage (ICH) is the most feared complication of anticoagulation in patients with atrial fibrillation (AF) and contributes substantially to the morbidity and mortality that accompanies warfarin-associated intracranial and extracranial hemorrhage [1]. Recent evidence from phase III clinical trials of the NOACs compared to warfarin or aspirin for stroke prevention in AF report that the NOACs substantially reduce the incidence of ICH compared to warfarin, and apixaban does not increase the incidence of ICH compared to aspirin [2–10]. This paper reviews the relevant literature that has been published in recent years regarding ICH among AF patients taking NOACs.
Rate Clinical Trials The rate of ICH among AF patients allocated NOACs in the phase III clinical trials of the NOACs (vs warfarin or aspirin) was 0.2 to 0.5 per 100 patient years (Table 1) [2–10]. The rates of ICH among 5851 AF patients who were assigned dabigatran in the RE-LY trial and who continued treatment with dabigatran for a median of 2.3 years after the RE-LY trial, as part of the Long-term Multicenter Extension of Dabigatran Treatment in Patients with Atrial Fibrillation (RELY-ABLE) study, were 0.33 % per year with dabigatran 150 mg bid and 0.25 % per year with dabigatran 110 mg bid [11•]. These estimates are similar to those observed during the RE-LY trial [11•]. General Population Following the approval and marketing of dabigatran for stroke prevention in AF, the reported rates of ICH among AF patients treated with dabigatran in “everyday clinical practice” (i.e.,
480, Page 2 of 9
Curr Cardiol Rep (2014) 16:480
Table 1 Features of ICH among AF patients taking NOACs in four phase III clinical trials Characteristics
RE-LY [2, 15••]
ROCKET AF [3, 16•]
No. of participants 18,113 14,264 Follow-up, years 2.0 (mean) 1.9 (median) Rate of intracranial hemorrhage (ICH) per 100 patient years Dabigatran 150 0.31 Dabigatran 110 0.23 Rivaroxaban 0.5 Edoxaban 60 Edoxaban 30 Apixaban Warfarin 0.76 0.7 Aspirin Site of ICH Intracerebral 71 (46 %) 128 (74 %) Subarachnoid 13 (8 %) 5 (3 %) Subdural 70 (45 %) 38 (22 %) Extradural 0 1 (1 %) Cause of ICH Trauma 46 (30 %) Mortality at 30 days by site All ICH 36 % Intracerebral 52 % Subarachnoid 67 % Subdural 31 % Extradural Mortality at 30 days by treatment Dabigatran 150 35 % Dabigatran 110 41 % Rivaroxaban Warfarin 36 % Predictors (RR or HR, 95 % CI) Age 1.1 / year Prior stroke /TIA 1.8 Race- White 0.68 Race – Asian Race – Black Warfarin 2.9 Aspirin also Thienopyridine Rivaroxaban Platelet count Serum albumin
ARISTOTLE [4]
AVERROES [5, 6]
ENGAGE [9]
18,201 1.8 (median)
5,599 1.1 (mean)
21,105 2.8 (median)
0.39 0.26 0.33 0.80
0.41 0.85 0.35 15 (62 %) 6 (25 %)
9 (7 %) 75 (43 %) 57 % 20 % 29 % 0%
50 % 48 % 1.34 / year 1.51 2.03 4.22
1.6 2.61 0.66 1.08 per 10×109/l decrease 1.42 per 0.5 g/dl decrease
“real world” populations) are about 0.1 to 0.3 per 100 patient years [12•, 13]. In the Danish Registry of Medicinal Product Statistics, the rate of ICH among 2239 anticoagulant-naïve AF patients treated with dabigatran etexilate 150 mg bid was 0.1 per 100 patient years, and among 2739 patients treated with dabigatran etexilate 110 mg bid the rate of ICH was 0.3 per
100 patient years [12•]. A review of reports of major bleeding among insurance-claim data and administrative data from the US FDA Mini-Sentinel database found that the rate of ICH among 54,000 individuals with AF who were taking dabigatran was 0.8 events per 100,000 days at risk, which equates to 0.29 per 100 patient years [13]. Although these
Curr Cardiol Rep (2014) 16:480
rates of ICH are similar to, if not lower than, the rates of ICH reported in the clinical trials of NOACs, they are not adjusted for possible differences in baseline risk of ICH or completeness of ICH ascertainment between the “real world” and clinical trial populations.
Anatomy About two thirds (half to three quarters) of NOAC-associated ICHs in clinical trials were intracerebral hemorrhages, and most of the remainder were subdural hematomas (Table 1).
Etiology The most common causes of spontaneous ICH are cerebral arteriopathies (subcortical hypertensive small vessel disease, amyloid angiopathy, arteriovenous malformations) and intense oral anticoagulation [14]. Some of the phase III trials of NOACs vs warfarin distinguished traumatic from nontraumatic ICH (Table 1), but causes of non-traumatic spontaneous ICH were not reported [15••, 16•].
Pathogenesis The pathophysiology of anticoagulant-associated intracerebral hemorrhage is uncertain. Experimental and observational evidence indicate that effective anticoagulation with NOACs increases the volume of intracerebral hemorrhage (and thereby a worse functional outcome) compared to a normal coagulation status [17–19]. The mechanism is impaired hemostasis. It is not known however if anticoagulation also precipitates ICH, perhaps by compromising the integrity of the cerebral vasculature. Animal studies of experimental acute ischemic stroke treated with thrombolysis, and not treated with thrombolysis, suggest that plasma concentrations of dabigatran similar to those used in humans do not increase hemorrhagic transformation of acutely infarcted brain whereas supra-therapeutic plasma concentrations of dabigatran and warfarin do [20–23].
Page 3 of 9, 480
Predictors Two prognostic models have examined predictors of anticoagulant-associated ICH among AF patients taking NOACs or warfarin in the RE-LY and ROCKET AF trials (Table 1) [15••, 16•]. In the RE-LY trial, the independent, significant predictors of ICH were assignment to warfarin (relative risk, 2.9; P
NEW THERAPIES FOR CARDIOVASCULAR DISEASE (KW MAHAFFEY, SECTION EDITOR)
Intracranial Hemorrhage and Novel Anticoagulants for Atrial Fibrillation: What Have We Learned? Graeme J. Hankey
Published online: 20 March 2014 # Springer Science+Business Media New York 2014
Abstract Intracranial hemorrhage (ICH) affects 0.2-0.5 % of atrial fibrillation (AF) patients taking a novel oral anticoagulant (NOAC) each year. About two thirds of ICHs are intracerebral and one quarter subdural. The 30-day case fatality of NOAC-associated ICH was similar to that of warfarinassociated ICH in two trials. Consistent predictors of ICH are increasing age, a history of prior stroke or TIA, and concomitant use of an antiplatelet drug. Compared to warfarin, the NOACs significantly reduce the risk of ICH by half (risk ratio=0.44; 95 % CI: 0.37 to 0.51). Compared to aspirin, apixaban has a similar risk of ICH (risk ratio=0.84; 95 % CI, 0.38 to 1.87). Current treatments for NOAC-associated ICH include nonactivated and activated prothrombin complex concentrate, which reverse the anticoagulant effects of the NOACs, but their effects on bleeding and patient outcome are not known. Future treatments for NOACassociated ICH promise to include specific antidotes to dabigatran (e.g., aDabi-Fab, PER977) and factor Xa inhibitors (e.g., r-Antidote PRT064445, PER977).
Keywords Intracranial hemorrhage . Novel oral anticoagulants . Atrial fibrillation
This article is part of the Topical Collection on New Therapies for Cardiovascular Disease G. J. Hankey (*) School of Medicine and Pharmacology, The University of Western Australia, Room 222, Harry Perkins Institute of Medical Research QQ block, QE II Medical Centre, 6 Verdun Street, Nedlands 6009, Australia e-mail: [email protected] G. J. Hankey Department of Neurology, Sir Charles Gairdner Hospital, Nedlands, Perth, Australia 6009
Introduction Intracranial hemorrhage (ICH) is the most feared complication of anticoagulation in patients with atrial fibrillation (AF) and contributes substantially to the morbidity and mortality that accompanies warfarin-associated intracranial and extracranial hemorrhage [1]. Recent evidence from phase III clinical trials of the NOACs compared to warfarin or aspirin for stroke prevention in AF report that the NOACs substantially reduce the incidence of ICH compared to warfarin, and apixaban does not increase the incidence of ICH compared to aspirin [2–10]. This paper reviews the relevant literature that has been published in recent years regarding ICH among AF patients taking NOACs.
Rate Clinical Trials The rate of ICH among AF patients allocated NOACs in the phase III clinical trials of the NOACs (vs warfarin or aspirin) was 0.2 to 0.5 per 100 patient years (Table 1) [2–10]. The rates of ICH among 5851 AF patients who were assigned dabigatran in the RE-LY trial and who continued treatment with dabigatran for a median of 2.3 years after the RE-LY trial, as part of the Long-term Multicenter Extension of Dabigatran Treatment in Patients with Atrial Fibrillation (RELY-ABLE) study, were 0.33 % per year with dabigatran 150 mg bid and 0.25 % per year with dabigatran 110 mg bid [11•]. These estimates are similar to those observed during the RE-LY trial [11•]. General Population Following the approval and marketing of dabigatran for stroke prevention in AF, the reported rates of ICH among AF patients treated with dabigatran in “everyday clinical practice” (i.e.,
480, Page 2 of 9
Curr Cardiol Rep (2014) 16:480
Table 1 Features of ICH among AF patients taking NOACs in four phase III clinical trials Characteristics
RE-LY [2, 15••]
ROCKET AF [3, 16•]
No. of participants 18,113 14,264 Follow-up, years 2.0 (mean) 1.9 (median) Rate of intracranial hemorrhage (ICH) per 100 patient years Dabigatran 150 0.31 Dabigatran 110 0.23 Rivaroxaban 0.5 Edoxaban 60 Edoxaban 30 Apixaban Warfarin 0.76 0.7 Aspirin Site of ICH Intracerebral 71 (46 %) 128 (74 %) Subarachnoid 13 (8 %) 5 (3 %) Subdural 70 (45 %) 38 (22 %) Extradural 0 1 (1 %) Cause of ICH Trauma 46 (30 %) Mortality at 30 days by site All ICH 36 % Intracerebral 52 % Subarachnoid 67 % Subdural 31 % Extradural Mortality at 30 days by treatment Dabigatran 150 35 % Dabigatran 110 41 % Rivaroxaban Warfarin 36 % Predictors (RR or HR, 95 % CI) Age 1.1 / year Prior stroke /TIA 1.8 Race- White 0.68 Race – Asian Race – Black Warfarin 2.9 Aspirin also Thienopyridine Rivaroxaban Platelet count Serum albumin
ARISTOTLE [4]
AVERROES [5, 6]
ENGAGE [9]
18,201 1.8 (median)
5,599 1.1 (mean)
21,105 2.8 (median)
0.39 0.26 0.33 0.80
0.41 0.85 0.35 15 (62 %) 6 (25 %)
9 (7 %) 75 (43 %) 57 % 20 % 29 % 0%
50 % 48 % 1.34 / year 1.51 2.03 4.22
1.6 2.61 0.66 1.08 per 10×109/l decrease 1.42 per 0.5 g/dl decrease
“real world” populations) are about 0.1 to 0.3 per 100 patient years [12•, 13]. In the Danish Registry of Medicinal Product Statistics, the rate of ICH among 2239 anticoagulant-naïve AF patients treated with dabigatran etexilate 150 mg bid was 0.1 per 100 patient years, and among 2739 patients treated with dabigatran etexilate 110 mg bid the rate of ICH was 0.3 per
100 patient years [12•]. A review of reports of major bleeding among insurance-claim data and administrative data from the US FDA Mini-Sentinel database found that the rate of ICH among 54,000 individuals with AF who were taking dabigatran was 0.8 events per 100,000 days at risk, which equates to 0.29 per 100 patient years [13]. Although these
Curr Cardiol Rep (2014) 16:480
rates of ICH are similar to, if not lower than, the rates of ICH reported in the clinical trials of NOACs, they are not adjusted for possible differences in baseline risk of ICH or completeness of ICH ascertainment between the “real world” and clinical trial populations.
Anatomy About two thirds (half to three quarters) of NOAC-associated ICHs in clinical trials were intracerebral hemorrhages, and most of the remainder were subdural hematomas (Table 1).
Etiology The most common causes of spontaneous ICH are cerebral arteriopathies (subcortical hypertensive small vessel disease, amyloid angiopathy, arteriovenous malformations) and intense oral anticoagulation [14]. Some of the phase III trials of NOACs vs warfarin distinguished traumatic from nontraumatic ICH (Table 1), but causes of non-traumatic spontaneous ICH were not reported [15••, 16•].
Pathogenesis The pathophysiology of anticoagulant-associated intracerebral hemorrhage is uncertain. Experimental and observational evidence indicate that effective anticoagulation with NOACs increases the volume of intracerebral hemorrhage (and thereby a worse functional outcome) compared to a normal coagulation status [17–19]. The mechanism is impaired hemostasis. It is not known however if anticoagulation also precipitates ICH, perhaps by compromising the integrity of the cerebral vasculature. Animal studies of experimental acute ischemic stroke treated with thrombolysis, and not treated with thrombolysis, suggest that plasma concentrations of dabigatran similar to those used in humans do not increase hemorrhagic transformation of acutely infarcted brain whereas supra-therapeutic plasma concentrations of dabigatran and warfarin do [20–23].
Page 3 of 9, 480
Predictors Two prognostic models have examined predictors of anticoagulant-associated ICH among AF patients taking NOACs or warfarin in the RE-LY and ROCKET AF trials (Table 1) [15••, 16•]. In the RE-LY trial, the independent, significant predictors of ICH were assignment to warfarin (relative risk, 2.9; P
