INTRODUCTION — Intracranial hemorrhage, which includes intracerebral, intraventricular, subarachnoid, subdural, and epidural bleeding, is a potentially devastating occurrence associated with anticoagulant therapy. Reversal of anticoagulation in patients with anticoagulant-associated intracranial hemorrhage is a medical emergency, as anticoagulation is associated with greater hematoma growth, neurologic deterioration, and increased risk of death and major disability compared with no anticoagulation.
This topic discusses the reversal of anticoagulation in patients with anticoagulant-associated intracranial hemorrhage.
Other aspects of the management and prevention of intracranial hemorrhage are presented separately.
●Risks and prevention
•(See "Risks and prevention of bleeding with oral anticoagulants", section on 'Intracranial'.)
●Management
•Intracerebral hemorrhage (ICH) – (See "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis".)
•Intraventricular hemorrhage (IVH) – (See "Intraventricular hemorrhage", section on 'Management'.)
•Subarachnoid hemorrhage (SAH) – (See "Aneurysmal subarachnoid hemorrhage: Treatment and prognosis".)
•Subdural hematoma (SDH) – (See "Subdural hematoma in adults: Management and prognosis".)
•Epidural hematoma (EDH) – (See "Intracranial epidural hematoma in adults", section on 'Management'.)
TERMINOLOGY — The following terms are used herein:
●Intracranial hemorrhage – Intracranial hemorrhage is the broadest term for bleeding; it includes bleeding anywhere inside the skull, including intracerebral, intraventricular, subarachnoid, and subdural hemorrhage.
●Intracerebral hemorrhage – Intracerebral hemorrhage (ICH) is bleeding in the brain parenchyma.
●Intraventricular hemorrhage – Intraventricular hemorrhage refers to bleeding within the ventricular system in the brain.
●Subarachnoid hemorrhage – Subarachnoid hemorrhage is bleeding directly adjacent to the brain in the subarachnoid space (between the pia mater and arachnoid membrane); most of these are caused by ruptured saccular aneurysms.
●Subdural hematoma – Subdural hematoma is bleeding outside the subarachnoid space directly beneath the dura mater.
●Epidural hematoma – Epidural hematoma is bleeding in the potential space between the dura mater and the skull or in the epidural space in the spinal canal.
URGENT EVALUATION — The goal of the evaluation is to document as rapidly as possible, with brain imaging, whether the patient's symptoms are due to intracranial bleeding rather than another cause, such as poisoning (related to a suicide attempt or other toxic exposure), encephalopathy, or ischemia; to document that the patient is in fact anticoagulated; and to confirm which anticoagulant is responsible. This information determines whether a reversal agent is needed, and if so, the specific agent.
We request urgent evaluation by neurology and/or neurosurgery (and stroke team where available) when a patient presents with a suspected intracranial hemorrhage.
Laboratory testing for anticoagulation status and brain imaging should be obtained immediately, notifying the laboratory and the radiology department of the emergency nature of the testing and imaging. Patient history can be obtained while the patient is being examined and blood for laboratory testing is being drawn. Often, the neurologist or stroke team can meet the patient in the radiology department and see the images as they are being obtained or review them immediately after.
Rapid clinical assessment — We rapidly obtain the relevant history, which includes the onset of symptoms, major neurologic abnormalities, which anticoagulant and dose the patient is taking and for what indication, and when they took the most recent dose.
●Presenting symptoms – Patients with symptomatic intracranial hemorrhage generally present with acute onset of neurologic impairment consistent with stroke, although some patients with subarachnoid hemorrhage (SAH) may be neurologically intact and present with isolated sudden-onset, severe headache (ie, thunderclap headache).
Subdural hematoma in the setting of anticoagulation may present with acute neurologic deterioration or milder symptoms initially, depending on the size and acuity.
Clinical manifestations of epidural hematoma (EDH) are highly variable and include altered consciousness, headache, vomiting, drowsiness, confusion, aphasia, seizures, and hemiparesis. Some patients with acute EDH and transient loss of consciousness have a "lucid interval" with recovery of consciousness followed by deterioration due to hematoma enlargement.
Symptoms of intracranial hemorrhage may include headache, nausea, vomiting, or neurologic deficit(s). It is important to distinguish between symptomatic hemorrhage and incidentally discovered hemorrhage on neuroimaging; both are potentially clinically serious, but the urgency and aggressiveness needed to manage symptomatic bleeding may be greater than that for an incidentally discovered subdural bleed that may already be stabilized or resolving. (See 'Neuroimaging' below.)
Additional details of the presentation and diagnosis of intracranial hemorrhage are reviewed separately by etiology:
•Intracerebral hemorrhage (see "Spontaneous intracerebral hemorrhage: Pathogenesis, clinical features, and diagnosis", section on 'Clinical presentation' and "Spontaneous intracerebral hemorrhage: Pathogenesis, clinical features, and diagnosis", section on 'Evaluation and diagnosis')
•Intraventricular hemorrhage (see "Intraventricular hemorrhage", section on 'Clinical presentation' and "Intraventricular hemorrhage", section on 'Diagnostic evaluation')
•Subarachnoid hemorrhage (see "Aneurysmal subarachnoid hemorrhage: Clinical manifestations and diagnosis", section on 'Clinical presentation' and "Aneurysmal subarachnoid hemorrhage: Clinical manifestations and diagnosis", section on 'Evaluation and diagnosis')
•Subdural hematoma (see "Subdural hematoma in adults: Etiology, clinical features, and diagnosis", section on 'Clinical manifestations' and "Subdural hematoma in adults: Etiology, clinical features, and diagnosis", section on 'Diagnosis and evaluation')
•Epidural hematoma (see "Intracranial epidural hematoma in adults", section on 'Clinical manifestations' and "Intracranial epidural hematoma in adults", section on 'Diagnostic evaluation')
●Trauma – We ask if there was direct trauma to the head. If the patient was injured in a motor vehicle accident or other trauma (eg, a fall), we determine how fast the vehicle was going and whether neurologic deterioration started before or after the accident. If the patient fell, we determine the distance fallen.
●Anticoagulant and other medications – We ask the name of the anticoagulant and confirm that it is the same as that listed in the medical record. We also ask if the patient is taking other medication(s) that could affect hemostasis, such as an antiplatelet agent (eg, aspirin, clopidogrel) or a drug that may affect the metabolism of their anticoagulant. The tables provide examples of drug interactions for warfarin (table 1) and the direct oral anticoagulants (DOACs) (table 2). (See "Biology of warfarin and modulators of INR control", section on 'Drug interactions'.)
●Dose and timing – The indication for the anticoagulant should be assessed as this helps in confirming the specific agent, the dose, and the risk of thrombosis if anticoagulation is stopped or reversed. We ask what anticoagulant dose the patient is using and when the last dose was taken. The strength and timing of the last dose taken may impact whether reversal is necessary and, especially for andexanet alfa, the amount of reversal agent needed. We also assess whether it is possible that the patient may be anticoagulated to a greater degree than expected with routine dosing. This includes questioning about how they are taking the anticoagulant to ascertain possible overdose or incorrect dosing. For patients on warfarin, we ask about the timing and result of the most recent international normalized ratio (INR).
●Comorbidities – We ask if the individual has renal or hepatic disease, which may affect clearance or metabolism of DOACs, as well as any intercurrent infection, which may affect the INR on warfarin. Disorders that alter gastrointestinal absorption of drugs may explain lack of anticoagulant effect in some cases. Concomitant uremia or thrombocytopenia may increase bleeding risk.
The underlying condition necessitating anticoagulation may also be a factor in decision-making, especially for those at highest risk of thromboembolic complications (recent pulmonary embolism, mechanical heart valve).
Patients with moderate-to-severe neurologic deficits due to acute stroke may be unable to provide a reliable history. In such cases, it is important to review the patient's medical record and ask family members, friends, and caregivers whether the patient is receiving an anticoagulant and if so, when it was last administered. In some cases, it may be reasonable to ask if there was a suspected suicide attempt. If no history is available, we base our assessment on the results of coagulation testing and imaging.
Neuroimaging — As noted above, neuroimaging is performed in individuals taking an anticoagulant who develop symptoms consistent with intracranial bleeding, including headache, nausea, vomiting, or neurologic defects. The need for intervention may be more urgent in those with symptomatic bleeding than for those with incidentally discovered bleeding. (See 'Rapid clinical assessment' above.)
Acute intracranial hemorrhage must be confirmed by neuroimaging, typically with urgent noncontrast computed tomography (CT), or, less frequently, with magnetic resonance imaging (MRI), before undertaking interventions to reverse anticoagulation. It is not possible to distinguish bleeding from ischemia by history and physical examination alone. This is especially true for patients with atrial fibrillation, who are at increased risk for ischemic stroke, but it applies to all individuals. As an example, neurologic symptoms following a motor vehicle accident may be due to traumatic intracranial hemorrhage or due to an ischemic stroke leading to loss of control of the vehicle.
In selected cases in which the time delay in obtaining intracranial imaging could be life threatening, the clinicians may decide to treat empirically; this is a challenging decision that weighs the risks and benefits of empiric treatment versus delaying therapy to confirm the diagnosis. As an example, anticoagulant reversal may be reasonable for an individual with head trauma who has papilledema and for whom brain imaging is not available. However, papilledema has causes aside from ICH and by itself (without brain imaging) is not enough evidence for reversal.
Laboratory testing — We perform the following testing as rapidly as possible:
●All patients – PT with INR, activated partial thromboplastin time (aPTT), and complete blood count (CBC) with platelet count.
●Patients on dabigatran – Thrombin time (TT); diluted thrombin time (dTT) or ecarin clotting time (ECT) may be performed but they are not widely available. Serum creatinine should also be measured, and creatinine clearance calculated, as dabigatran is primarily eliminated by renal clearance.
●Patients on an oral factor Xa inhibitor or low molecular weight (LMW) heparin – Anti-factor Xa activity, creatinine, and calculated creatinine clearance.
●Selected patients – Liver function tests, basic metabolic panel, and other testing to address suspected comorbidities or other causes of neurologic deterioration.
TT and anti-factor Xa activity may not be routinely available at all institutions, although most hospitals can perform a TT.
Obtaining a platelet count is important to ensure that the patient does not have concomitant thrombocytopenia, which might contribute to bleeding.
The usefulness of these coagulation tests varies according to the anticoagulant agent:
●Warfarin – A prolonged PT or INR (eg, INR ≥1.4) indicates that the patient is anticoagulated. The aPTT is also prolonged with warfarin. In some cases, an individual with a PT or INR at the high end of the normal range may be slightly anticoagulated, especially if the value is higher than his or her baseline before starting warfarin. The PT and INR are monitored daily to determine if the warfarin effect is persistent and additional vitamin K is needed. (See 'Warfarin' below.)
●Dabigatran – High plasma levels of dabigatran may prolong the aPTT. However, the aPTT may be normal or near normal in patients with therapeutic levels of dabigatran. Therefore, a prolonged aPTT indicates the need for reversal, but normal aPTT values cannot be used to establish the lack of dabigatran effect. The TT, dTT, and ECT are more sensitive to the anticoagulant effects of dabigatran than the aPTT. Therefore, we would not use a reversal agent if the TT, dTT, and/or ECT are normal. If the TT, dTT, or ECT are not available, anticoagulation status is determined from the history of dabigatran ingestion (eg, dose, time since last dose). (See 'Dabigatran' below.)
●Direct factor Xa inhibitors – Direct factor Xa inhibitors (apixaban, edoxaban, rivaroxaban) are detected by anti-factor Xa activity calibrated to the specific agent. In some cases, anti-factor Xa activity calibrated to a different agent may show drug effect, although this is not optimal. The PT and aPTT may be helpful if prolonged but are not considered reliable indicators of anticoagulant effect. An anti-factor Xa activity that indicates a drug level of >30 ng/mL is evidence of anticoagulation. If anti-factor Xa activity is not available and the PT and aPTT are normal, anticoagulation status is determined from the history of anticoagulant ingestion. (See 'Apixaban, edoxaban, and rivaroxaban' below.)
●Unfractionated heparin – Unfractionated heparin prolongs the aPTT and has anti-factor Xa activity. Heparin has a short half-life of approximately one hour after intravenous infusion. A normal aPTT indicates that the anticoagulant effect has resolved. (See 'Unfractionated heparin' below.)
●LMW heparin – LMW heparin has anti-factor Xa activity. The assay should be calibrated to LMW heparin. A detectable LMW heparin level ≥0.3 international units/mL indicates LMW heparin effect. There is no anticoagulant effect if there is no detectable anti-factor Xa activity or if the anti-factor Xa activity indicates a LMW heparin level less than 0.3 international units/mL. (See 'LMW heparin' below.)
The expected effect of anticoagulants on commonly used clotting tests is summarized in the table (table 3).
For warfarin, the extent of INR prolongation at the time of warfarin-associated ICH correlates with initial hematoma size, progressive hematoma enlargement after admission, functional outcome, and mortality [1-6]. Most episodes of warfarin-associated ICH occur in patients with a therapeutic level of anticoagulation (INR 2.0 to 3.5) [1,7-11]. However, even patients with a therapeutic INR can have an increased risk of bleeding, especially those older than 70 years of age. (See "Management of warfarin-associated bleeding or supratherapeutic INR", section on 'Mitigating bleeding risk'.)
While similar data have not been obtained for all the other anticoagulants, it is likely that the intensity of the anticoagulant effect correlates with the severity of intracerebral bleeding, and that any level of anticoagulation puts the patient at risk for adverse outcomes of intracranial bleeding compared with no anticoagulation.
INDICATIONS FOR REVERSAL AND GOALS OF TREATMENT
Acute intracranial hemorrhage — Urgent anticoagulant reversal is indicated for patients with acute intracerebral (ICH), intraventricular, subarachnoid, or subdural hemorrhage associated with active anticoagulation. These types of hemorrhage are assumed to be life threating regardless of the extent of hemorrhage visible on initial brain imaging, since ongoing bleeding and hemorrhage enlargement can cause neurologic deterioration, elevation in intracranial pressure, and poor functional outcome or death [12].
An exception is that urgent reversal may not be necessary for a clinically stable patient with a small, chronic subdural hemorrhage and no evidence of elevated intracranial pressure; in such a case, the potential benefit of reversing anticoagulation (reduced risk of hematoma enlargement) must be weighed against the risk of thrombosis related to the underlying need for anticoagulation. As an example, anticoagulation with warfarin dose adjusted to achieve a therapeutic international normalized ratio (INR) may be continued in an individual with a mechanical heart valve and a small subdural hematoma with no signs of increased intracranial pressure, since the risk of valve thrombosis may be a greater concern than hematoma expansion. Another option in such an individual is to switch to heparin anticoagulation, which is easier to reverse quickly, while observing for clinical improvement. This subject is discussed in detail separately. (See "Anticoagulation for prosthetic heart valves: Management of bleeding and invasive procedures" and "Antithrombotic therapy for mechanical heart valves".)
As noted above, reversal is only appropriate after intracerebral bleeding has been documented on an imaging study. Empiric treatment for suspected intracranial hemorrhage in the absence of confirmation by one of these methods is not advised unless the patient is in extremis and imaging is not available. This is because reversal agents are potentially prothrombotic, and their use may cause harms without benefit if they are given to an individual who did not actually have a hemorrhage [12].
As noted above, some patients who are receiving an anticoagulant may not actually be anticoagulated, especially if they are on a short-acting agent and took their last dose several half-lives ago. The lack of anticoagulation was demonstrated in some of the trials evaluating reversal agents for the direct oral anticoagulants. (See "Management of bleeding in patients receiving direct oral anticoagulants", section on 'Anticoagulant reversal'.)
The main goals of treatment are to rapidly reverse anticoagulation effects and to maintain reversal for a minimum of 72 hours, thereby limiting hemorrhage enlargement. This is important because hematoma growth, particularly within the first 24 hours after ICH, is an independent predictor of mortality and poor outcome. (See "Spontaneous intracerebral hemorrhage: Pathogenesis, clinical features, and diagnosis", section on 'Predicting hemorrhage expansion' and "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis", section on 'Clinical risk factors'.)
Reversal of anticoagulation, along with blood pressure control, is believed to improve outcomes in warfarin-associated ICH. In one observational study, reduced rates of hematoma enlargement were noted for patients who had reversal of INR <1.3 within four hours of admission (20 versus 42 percent) [13]. The combination of successful reversal of INR <1.3 and blood pressure control <160 mmHg within four hours was associated with lowered rates of hematoma enlargement (18 versus 44 percent) and in-hospital mortality (13 versus 21 percent).
Lumbar puncture in suspected subarachnoid hemorrhage — When there is strong suspicion for subarachnoid hemorrhage (SAH) despite a normal head computed tomography (CT), lumbar puncture is generally required, except for selected patients with isolated headache, a normal neurologic examination, and high-quality brain imaging performed within six hours of headache onset that is negative for hemorrhage. The sensitivity of CT for detecting SAH decreases over the ensuing hours to days. The evaluation for SAH is discussed in detail separately. (See "Aneurysmal subarachnoid hemorrhage: Clinical manifestations and diagnosis", section on 'Evaluation and diagnosis'.)
Anticoagulation is considered a contraindication to lumbar puncture. Anticoagulation should be reversed for patients who require a lumbar puncture to rule out SAH while anticoagulated, as assessed by criteria described above (eg, INR ≥1.5 for warfarin; prolonged TT for dabigatran; anti-factor Xa activity assay that indicates a level of apixaban, rivaroxaban, or edoxaban over 30 ng/mL; history of ingestion in the preceding 24 to 48 hours if laboratory testing is inconclusive or unavailable). (See 'Laboratory testing' above.)
GENERAL MEASURES FOR ALL ANTICOAGULANTS
Discontinue all antithrombotic agents — All anticoagulant and antiplatelet therapy should be discontinued. This includes discontinuation of the anticoagulant the patient is taking, and avoidance of other anticoagulants (eg, "routine" orders for heparin or low molecular weight (LMW) heparin administration for venous thromboembolism prophylaxis). These measures must be clearly stipulated in the medical record and ordering system.
If an individual has another hemostatic defect such as moderate to severe thrombocytopenia (platelet count <50,000 to 100,000/microL) or use of dual antiplatelet therapy, consideration should be given to concomitant therapy for that defect as well. (See 'Other supportive care' below and 'Limited role of platelet transfusions' below.)
Potential exceptions are noted above. (See 'Indications for reversal and goals of treatment' above.)
Admit to intensive care — Patients with acute anticoagulant-associated intracranial hemorrhage should be managed initially in an intensive care setting where frequent neurologic checks can detect neurologic deterioration, and hemodynamic monitoring can allow tighter blood pressure control and management.
Control blood pressure — Elevated blood pressure may predispose to hematoma expansion in patients with intracerebral hemorrhage (ICH), as already noted. (See 'Indications for reversal and goals of treatment' above.)
We target a systolic blood pressure below 140 mmHg for ICH and below 160 mmHg for subarachnoid hemorrhage. Lower blood pressure targets in the acute setting increase the risk of hypoperfusion and infarction. We generally administer antihypertensive agents for those with a systolic blood pressure above 150 mmHg. Specific antihypertensive agents and the frequency of monitoring are discussed separately. (See "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis", section on 'Blood pressure management' and "Aneurysmal subarachnoid hemorrhage: Treatment and prognosis", section on 'Blood pressure control'.)
Other supportive care — Other interventions include rapid and regular assessment of hemodynamic status and airway, along with optimization of body temperature, pH, and electrolyte balance.
Rarely, transfusions may be needed, such as platelet transfusions for thrombocytopenia or red blood cell transfusions for anemia. (See 'Limited role of platelet transfusions' below.)
Supportive measures are discussed in more detail separately. (See "Management of bleeding in patients receiving direct oral anticoagulants", section on 'Overview of management'.)
Supportive care related to the site of bleeding is presented elsewhere:
●Intracerebral hemorrhage – (See "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis".)
●Intraventricular hemorrhage – (See "Intraventricular hemorrhage", section on 'Management'.)
●Subarachnoid hemorrhage – (See "Aneurysmal subarachnoid hemorrhage: Treatment and prognosis".)
●Subdural hematoma – (See "Subdural hematoma in adults: Management and prognosis".)
●Epidural hematoma – (See "Intracranial epidural hematoma in adults", section on 'Management'.)
Limited role of platelet transfusions — Platelet transfusions generally are not indicated in the setting of intracranial bleeding, even for patients on concomitant antiplatelet therapy.
●Thrombocytopenia – Platelet transfusion is appropriate for an individual with platelet counts <100,000/microL or with a known platelet function defect. (See "Platelet transfusion: Indications, ordering, and associated risks", section on 'Actively bleeding patient' and "Platelet transfusion: Indications, ordering, and associated risks", section on 'Platelet function disorders'.)
●Antiplatelet agents – The available data from trials of patients with spontaneous ICH (eg, the PATCH trial) suggest that empiric platelet transfusions in those without thrombocytopenia may be hazardous and should generally be avoided. (See "Platelet transfusion: Indications, ordering, and associated risks", section on 'Antiplatelet agents'.)
Some clinicians use platelet transfusions for patients taking long-acting antiplatelet agents such as clopidogrel or prasugrel, but there is limited evidence to support this approach. Input from the neurologist and hematologist may be helpful in determining an individualized approach. Platelet transfusion is not considered useful for patients taking ticagrelor because ticagrelor is a reversible inhibitor of the adenosine diphosphate (ADP) receptor on the platelet surface and will bind to transfused platelets.
REVERSAL STRATEGY FOR SPECIFIC ANTICOAGULANTS — The reversal agent depends on which anticoagulant the patient is receiving. This should be confirmed from the history and in some cases, the results of coagulation testing. (See 'Rapid clinical assessment' above and 'Laboratory testing' above.)
As noted above, reversal should be done as rapidly as possible but should only be done when intracerebral bleeding is documented (see 'Neuroimaging' above), in order to avoid giving potentially prothrombotic drugs unnecessarily to a person with an increased thrombosis risk.
Our approach for specific anticoagulants (outlined in the following sections) is largely consistent with general guidelines for reversing anticoagulation in the setting of severe bleeding as well as guidelines specific to hemorrhagic stroke [14-19]. A complete listing of society guidelines is presented separately. (See "Society guideline links: Anticoagulation" and "Society guideline links: Stroke in adults".)
Warfarin — Intracranial bleeding associated with warfarin anticoagulation should be treated with a rapid source of functional clotting factors as well as vitamin K to allow endogenous production of functional vitamin K-dependent factors.
Warfarin works by interfering with carboxylation of the vitamin K-dependent clotting factors (factors II [prothrombin], VII, IX, and X). Carboxylation of these factors is essential for their function. The principal means of reversing warfarin rapidly is to replace these fully functional factors. The most rapidly acting source of functional factors is a 4-factor prothrombin complex concentrate (4-factor PCC).
Convincing evidence of warfarin anticoagulation is based on a prolonged prothrombin time (PT) and an international normalized ratio (INR) outside the normal range (≥1.4 in most cases). (See 'Laboratory testing' above.)
Reversal and monitoring strategy — Our approach (described in the following sections) is similar to the 2022 guideline from the American Heart Association (AHA)/American Stroke Association (ASA), the 2018 guideline from the American Society of Hematology (ASH), and the 2012 guideline from the American College of Chest Physicians (ACCP), which recommend the following for serious or life-threatening bleeding associated with warfarin (table 4) [16,18,19]:
●Hold warfarin. We also make sure that warfarin has been discontinued and that this is clearly stated in the medical record.
●Administer a 4-factor PCC; if a PCC is not available, a plasma product such as Fresh Frozen Plasma (FFP) or Thawed Plasma may be used. (See 'Reversal agent' below.)
●Administer intravenous vitamin K. (See 'Intravenous vitamin K' below.)
The effects of these treatments is summarized in the table (table 5) and an overview of warfarin-associated bleeding and reversal is presented separately. (See "Management of warfarin-associated bleeding or supratherapeutic INR".)
We recheck the PT/INR at approximately 30 minutes following PCC (or plasma) administration and periodically thereafter (eg, INR checked every four to six hours for the first 24 hours, and then checked daily for a few days) to ensure that the INR has become normal (<1.4 in most laboratories) and is maintained in the normal range [11,20].
If the INR remains elevated, additional doses of PCC or plasma may be given [21,22]. Details are presented separately. (See "Management of warfarin-associated bleeding or supratherapeutic INR", section on 'Serious/life-threatening bleeding'.)
Reversal agent — For warfarin-associated ICH, we recommend reversal with 4-factor prothrombin complex concentrate (4-factor PCC) rather than plasma. Options if 4-factor PCC is unavailable include 3-factor PCC supplemented with a source of factor VII, such as recombinant activated factor VII (rFVIIa), or a plasma product such as Fresh Frozen Plasma or Thawed Plasma. However, use of plasma may delay warfarin reversal, leading to a greater risk of complications of hematoma expansion, and may cause a transfusion reaction.
We generally do not recommend the use of rFVIIa alone for treatment of warfarin-associated ICH because it does not replace factors II, IX, or X and may give a false sense of security by normalizing the INR without fully reversing warfarin effect. rFVIIa acts rapidly (eg, normalization of the INR within 10 minutes), but the half-life is only two to three hours, and repeated dosing or administration of another product would be required [23,24]. Although the INR is normalized rapidly with rFVIIa, the bleeding risk may persist due to dysfunction of other vitamin K-dependent factors. Thus, the normal INR may give a false sense of security and deprive the patient of other, more effective treatments. Additionally, 4-factor PCC is likely to carry a lower risk of thrombosis than products containing activated factors (rFVIIa or activated PCC, which contains activated factor VII).
Prothrombin complex concentrate — All four vitamin K-dependent factors are present in 4-factor PCC, which can be administered rapidly in a small volume (table 6). Thus, 4-factor PCC is the preferred treatment (table 4). Supporting evidence is summarized below. (See 'Efficacy of PCC versus plasma' below.)
All institutions that treat patients with anticoagulant-associated hemorrhage should stock a 4-factor PCC.
The available strategies for factor replacement are presented in order of preference; only one of these should be used, along with vitamin K:
●4-factor PCC – We generally give 4-factor PCC (Kcentra in the United States and Japan; Beriplex or Octaplex in Canada; Octaplex, Cofact, or Proplex in many European countries) as an initial fixed dose of 1500 to 2000 international units at a rate of 100 units/minute (table 7). Alternatively, the initial dose may be calculated by body weight and INR. In a randomized trial involving 199 patients with warfarin-associated extracranial bleeding, the rate of effective hemostasis in those assigned to an initial fixed-dose PCC (1000 units) was similar to those assigned a calculated dose that incorporated body weight and INR (87 versus 90 percent) [25]. The median initial dose in the calculated dose group was 1750 units. An additional dose of PCC was given to four patients in the fixed-dose group compared with one in the weight-based group, but the door-to-needle time was shorter in the fixed-dose group (109 versus 142 minutes). The proportion of patients in each group reaching an INR ≤2.0 within 60 minutes was similar (91 versus 92 percent).
If weight-based or INR-based dosing is used, institutional protocols should be followed. An example is 25 units/kg for INR 2 to 4; 35 units/kg for INR 4 to 6; and 50 units/kg for INR >6; with a maximum dose of 5000 units [26].
●3-factor PCC plus a supplement – If a 4-factor PCC is not available, a 3-factor PCC (Profilnine in the United States; Bebulin was discontinued in 2018) can be used. However, 3-factor PCC does not contain factor VII. Therefore, most experts recommend supplementing 3-factor PCC with rFVIIa at a dose of 20 mcg/kg or with plasma (eg, FFP, two units).
If rFVIIa is used to supplement 3-factor PCC, we prefer a lower dose (20 mcg/kg) rather than higher doses. This is based on data from a randomized trial in 841 patients with spontaneous intracerebral hemorrhage (ICH; not warfarin-associated) that compared two doses of rFVIIa (80 and 20 mcg/kg) with placebo; PCC was not administered [27]. The primary outcome (severe disability or death) was similar among the three groups, despite less expansion of the hemorrhage in those receiving rFVIIa. Overall, thromboembolic events were similar in the three groups, but severe events (eg, cerebral infarction, myocardial infarction [MI]) were more common in those who received high-dose rFVIIa compared with placebo (8 versus 4 percent).
●Activated PCC – Activated PCC (aPCC) contains factors II, VII, IX, and X; factor VII is mostly present in the activated form (VIIa). The only available aPCC is factor eight inhibitor bypassing activity (FEIBA). FEIBA generally is not used for reversing warfarin anticoagulation, because the activated factor VII is potentially more prothrombotic compared with the factor VII in 4-factor PCC (discussed above) that is not in the activated state. However, retrospective data suggest that treatment with FEIBA does not increase the risk of thrombotic events compared with plasma products [28].
Plasma products if PCC is unavailable — If prothrombin complex concentrate (PCC) is unavailable, a plasma product (eg, FFP, Thawed Plasma) can be used to provide clotting factors. However, plasma requires a much larger volume of administration, often delaying the time to normalization of the INR, during which time hemorrhage expansion can continue. This was illustrated in a series of 45 patients with ICH, in which the median time interval between admission to a neuro-intensive care unit and INR normalization with FFP was 30 hours (range: 14 to 50 hours) [29]. Thus, plasma is not a very effective intervention for reducing the expansion of the hemorrhage [12].
Plasma also carries risks of transfusion reactions. A reasonable approach is to give two units of plasma, recheck the INR, and give additional units if needed. The infusion rate for plasma depends on the patient's ability to tolerate the volume load.
FFP and other plasma products such as Thawed Plasma are equally effective in reversing the effects of anticoagulation and are considered interchangeable. Thawed Plasma (FFP that has been stored by refrigeration at 1 to 6 degrees Celsius) has the advantage of being available for immediate use. Eight units (2 liters) of FFP or Thawed Plasma are often required to fully reverse anticoagulation in patients treated with warfarin or other vitamin K antagonists; the total number of units required depends on the extent of INR prolongation. (See "Clinical use of plasma components", section on 'Plasma products'.)
Efficacy of PCC versus plasma — Prothrombin complex concentrates (PCCs) normalize the INR more rapidly than infusion of plasma or vitamin K alone, often within 10 minutes of administration [11,21,30-33]. However, vitamin K should be administered concomitantly because the effect of PCC is transient (hours) [11]. (See "Plasma derivatives and recombinant DNA-produced coagulation factors", section on 'PCCs'.)
Evidence supporting the use of 4-factor PCC in warfarin-associated intracranial bleeding mostly consists of small, randomized trials and observational studies in ICH. As examples:
●PCC versus plasma – Two randomized trials comparing PCC with FFP in intracranial hemorrhage associated with vitamin K antagonists. One trial was stopped early due to safety concerns when the 4-factor PCC group appeared to have a lower rate of death (8 of 23 [35 percent] receiving FFP versus 5 of 27 [19 percent] receiving PCC) and a lower rate of hematoma expansion [34]. The other trial, which randomly assigned 202 patients who presented with major bleeding associated with a vitamin K antagonist to receive 4-factor PCC or plasma, found that PCC was associated with a trend towards greater hemostatic efficacy (72 percent with PCC versus 65 percent with plasma) and a greater likelihood of INR correction within the first half hour after the infusion (62 versus 10 percent); serious adverse events were similar [26]. Prospective and retrospective observational studies have consistently shown superior or equivalent outcomes with PCC compared with plasma; in some studies, PCCs were also associated with fewer serious adverse events [22,28,31,35-39]. Despite the efficacy of PCC in reversing warfarin effect, the mortality of anticoagulant-associated ICH remains high (42 percent in one series) [40]. This is likely due to early hematoma expansion and the delay in anticoagulant reversal during transportation to the hospital, intracranial imaging, and administration of reversal products.
●4-factor versus 3-factor PCC – These two products have not been directly compared for patients with intracerebral or any other type of bleeding. A systematic review of 18 observational studies involving 654 patients with warfarin-associated bleeding (mostly ICH or gastrointestinal bleeding) or requiring urgent surgery found that INR correction was more reliable with 4-factor than with 3-factor PCC [41]. Dosing was variable: Some studies used fixed dosing; some used weight-based dosing. A reduction in INR to ≤1.5 within one hour of PCC administration was seen in six of nine studies employing 3-factor PCCs, and in 12 of 13 studies employing 4-factor PCCs; the single 4-factor PCC study that did not meet this endpoint used a lower-than-average PCC dose (200 international units). Fixed-dose PCC might be easier logistically to stock and deliver in emergencies, but supplemental PCC doses may be required for those with warfarin-associated ICH who have a higher INR (>3).
●PCC versus rFVIIa or aPCC – A case series of 101 patients with warfarin-associated intracranial hemorrhage treated with rFVIIa (mean dose 52 mcg/kg) reported thromboembolic complications in eight patients (seven deep vein thromboses, one stroke) [42]. Similar risks were found in a multicenter randomized trial of rFVIIa in 841 patients with ICH not associated with warfarin [27]. There was no benefit with treatment on the primary clinical outcomes of death and disability, and higher rates of arterial thromboembolic serious adverse events (eg, stroke, MI) were found in patients assigned to the higher dose (80 mcg/kg) treatment group. Other studies have also suggested an association between rFVIIa and serious thromboembolic events. These and other studies are discussed in more detail separately. (See "Recombinant factor VIIa: Administration and adverse effects", section on 'Thromboembolic complications' and "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis", section on 'Reverse anticoagulation'.)
In a retrospective study comparing outcomes with 4-factor PCC versus aPCC in 342 patients with ICH, treatment with 4-factor PCC was associated with a higher likelihood of an INR ≤1.5 [43].
Case reports suggest that incomplete INR correction is associated with clinical worsening in patients treated with PCC. Observational studies show that in some cases ICH can continue to expand, even in patients for whom anticoagulation is reversed, although more rapid reversal with a PCC appears to correlate with a lower risk of expansion [4,11,44].
Additional supporting data for PCC are presented separately. (See "Management of warfarin-associated bleeding or supratherapeutic INR", section on 'PCC products'.)
Thrombotic events have complicated infusion of PCC, but this risk is difficult to quantify due to varying preparations, doses, and differing patient populations in available reports. Among most series, thrombotic complications occurred in 1.5 to 10 percent of patients [20,31,40,45-47]. The risk may be substantially higher in individuals with prosthetic heart valves or valvular heart disease [44].
Intravenous vitamin K — Vitamin K should be given because the half-life of PCC is very short (hours).
Vitamin K 10 mg is given by slow intravenous infusion, no faster than 1 mg/min to minimize anaphylactic risk [11,31]. If the INR is ≤1.5, a lower dose (eg, 5 mg) may be given if desired. The effect of vitamin K on the INR takes approximately 12 to 24 hours; thus, all patients should also receive PCC. (See 'Prothrombin complex concentrate' above.)
Intravenous vitamin K administration is preferred over oral or subcutaneous administration because it results in more rapid correction of the INR and because oral administration can be problematic in the setting of neurologic deficits or conditions that affect gastrointestinal absorption. Oral vitamin K may be used in individuals who are awake and have normal gastrointestinal function.
Vitamin K administration can be repeated every 12 hours for persistent INR elevation, and daily INR should be obtained to assess for this need [48]. (See 'Laboratory testing' above.)
Evidence supporting the efficacy of vitamin K and comparison of the routes of administration are presented separately. (See "Management of warfarin-associated bleeding or supratherapeutic INR", section on 'Vitamin K dose, route, formulation'.)
Dabigatran — Intracranial bleeding associated with dabigatran can be treated with idarucizumab or aPCC. We make sure that dabigatran has been discontinued and that this is clearly stated in the medical record. (See "Management of bleeding in patients receiving direct oral anticoagulants", section on 'Dabigatran reversal'.)
Convincing evidence of dabigatran anticoagulation may be based on a clinical history of ingestion within the previous 3.5 days and/or laboratory evidence of dabigatran effect (eg, prolonged activated partial thromboplastin time [aPTT], thrombin time [TT], diluted thrombin time [dTT], or ecarin clotting time [ECT]). An aPTT in the normal range cannot be used to justify withholding of idarucizumab, because drug effect may still be present. However, idarucizumab should not be given to patients who have a normal TT, dTT, and/or ECT. (See "Direct oral anticoagulants (DOACs) and parenteral direct-acting anticoagulants: Dosing and adverse effects", section on 'Laboratory testing and monitoring (dabigatran)'.)
Idarucizumab — Idarucizumab is an emergency reversal agent for dabigatran. It is an anti-dabigatran monoclonal antibody fragment.
For patients with acute intracranial hemorrhage and convincing evidence of dabigatran anticoagulation, we suggest idarucizumab if available, rather than clotting factor products (aPCC, PCC, or plasma).
The dose of idarucizumab is 5 grams (two 2.5-gram vials), administered either as two consecutive infusions or as a bolus (ie, injecting both vials consecutively via syringe). Repeat dosing is generally not required but may be appropriate in selected cases (eg, overdose, persistently prolonged aPTT), although data are limited.
We do not combine idarucizumab with other prohemostatic products such as a PCC, aPCC, or rFVIIa.
Treatment with idarucizumab may be associated with thrombosis due to the patient's underlying thrombotic risk factors. Evidence for the efficacy and safety of idarucizumab in dabigatran-associated bleeding is presented separately. (See "Management of bleeding in patients receiving direct oral anticoagulants", section on 'Dabigatran reversal'.)
Alternative options
Activated PCC — If idarucizumab is not available, we suggest administering activated prothrombin complex concentrate (aPCC; FEIBA) at a dose of 50 to 80 units/kg. The activated factor VII in this product activates the free factor X and may be sufficient to bypass dabigatran and promote clotting.
If aPCC is not available, 4-factor or 3-factor PCC at a dose of 50 units/kg would be reasonable. Three-factor PCC may be supplemented with rFVIIa or plasma. Dosing of these supplements is described above. (See 'Reversal agent' above.)
Activated charcoal and dialysis — Unabsorbed dabigatran can be removed from the gastrointestinal tract using oral activated charcoal. This is generally appropriate if the last dose was within the previous two hours. Dosing and contraindications are presented separately. (See "Gastrointestinal decontamination of the poisoned patient".)
Dabigatran can also be removed by hemodialysis if the consulting specialist believes this would be useful.
These interventions are discussed in more detail separately. (See "Management of bleeding in patients receiving direct oral anticoagulants", section on 'Overview of management'.)
Apixaban, edoxaban, and rivaroxaban — Intracranial bleeding associated with a direct oral factor Xa inhibitor (apixaban, edoxaban, or rivaroxaban) can be treated with andexanet alfa (a reversal agent for factor Xa inhibitors) or 4-factor PCC (table 8). We make sure that the factor Xa inhibitor has been discontinued and that this is clearly stated in the medical record.
Convincing evidence of factor Xa inhibitor anticoagulation may be based on a clinical history of ingestion within a period of five half-lives and/or laboratory evidence of anticoagulant effect (eg, increased anti-factor Xa activity, ideally calibrated for the specific drug). The PT, aPTT, and anti-Xa calibrated for other drugs may be useful if abnormal but are less reliable. (See "Direct oral anticoagulants (DOACs) and parenteral direct-acting anticoagulants: Dosing and adverse effects", section on 'Direct factor Xa inhibitors'.)
The half-lives of these agents and the number of days elapsed in five half-lives are listed separately (see "Management of bleeding in patients receiving direct oral anticoagulants", section on 'Interval since last dose'); most are within two to three days.
Reversal agent options — For patients with acute intracranial hemorrhage and convincing evidence of anticoagulation with a factor Xa inhibitor, we suggest andexanet alfa or 4-factor PCC. These agents have not been directly compared in a randomized trial. We prefer andexanet alfa given the drug's specificity and available data [49,50].
However, some experts consider there to be more equipoise in the choice between andexanet alfa and PCC. In a retrospective chart review of 109 adults with intracranial hemorrhage who were taking either apixaban or rivaroxaban at presentation, the rate of effective hemostasis was similar (71 percent for patients who received andexanet alfa and those who received 4-factor PCC) [51]. At baseline, the mean intracerebral hemorrhage score was 1 and Glasgow Coma Scale score was 14. After treatment, the median change in hematoma volume on repeat brain imaging and the rate of thrombotic complications were also similar between groups, but the total cost of treatment was more than three times higher with andexanet alfa. These results are limited by the relative small sample size, the potential of treatment bias, and uncertain applicability to patients with more severe hemorrhagic events. A randomized trial comparing andexanet versus usual care in patients with intracranial bleeding is ongoing [52]. Further discussion of this subject and evidence for the efficacy and safety of andexanet are reviewed separately. (See "Management of bleeding in patients receiving direct oral anticoagulants", section on 'Andexanet alfa'.)
Andexanet alfa — Andexanet alfa (andexanet) is an emergency reversal agent for factor Xa inhibitors. It is a recombinantly produced, catalytically inactive form of factor Xa that acts as a "decoy" to bind and sequester the factor Xa inhibitor anticoagulant.
Andexanet alfa is given at one of two dose levels based on the dose and timing of the factor Xa inhibitor.
●The low dose is given as a bolus of 400 mg at 30 mg/min over 15 minutes, followed by an infusion of 480 mg given at 4 mg/min for up to 120 minutes. This is used in patients who received a lower dose of factor Xa inhibitor (eg, rivaroxaban ≤10 mg, apixaban ≤5 mg, or edoxaban ≤30 mg) or if eight hours or more have elapsed since the last dose of factor Xa inhibitor.
●The high dose is given as a bolus of 800 mg at 30 mg/min over 30 minutes, followed by an infusion at 960 mg given at 8 mg/min for up to 120 minutes. This is used for those who received a higher dose of factor Xa inhibitor (eg, rivaroxaban >10 mg, apixaban >5 mg, edoxaban >30 mg) or unknown dose within the previous eight hours.
We do not use anti-factor Xa assays to assess the extent of anticoagulation reversal. Routine anti-factor Xa levels obtained after treatment with andexanet alfa may be falsely elevated due to dilutional effects [53].
We do not combine andexanet with other prohemostatic products such as PCC, aPCC or rFVIIa.
4-factor PCC — A 4-factor prothrombin complex concentrate (PCC) is an alternative to andexanet for reversing factor Xa inhibitors. In a retrospective series involving 663 individuals who had an intracranial hemorrhage while receiving apixaban or rivaroxaban, 4-factor PCC was associated with good or excellent hemostasis in 354 of 433 evaluated for efficacy (82 percent) [54]. Thrombosis within 14 days of PCC administration occurred in 22 of the 663 (3.3 percent). These efficacy and safety outcomes are like those seen with andexanet and with 4-factor PCC in other studies of oral factor Xa inhibitor reversal after bleeding in other sites. (See "Management of bleeding in patients receiving direct oral anticoagulants", section on 'Overview of factor Xa inhibitor reversal'.)
PCC can be given at a dose of 50 units/kg, or a fixed-dose regimen (2000 or 2500 units) can be used (eg, dosing like that used for warfarin reversal). (See 'Reversal agent' above.)
If PCC is used, the patient should not be treated with andexanet.
Activated charcoal — Unabsorbed anticoagulant can be removed from the gastrointestinal tract using oral activated charcoal.
We use the following intervals from the most recent dose to decide if charcoal may be helpful (see "Management of bleeding in patients receiving direct oral anticoagulants", section on 'Factor Xa inhibitors'):
●Apixaban – Within six hours
●Edoxaban – Within two hours
●Rivaroxaban – Within six to eight hours
Dosing and contraindications to oral activated charcoal are presented separately. (See "Gastrointestinal decontamination of the poisoned patient".)
Direct factor Xa inhibitors cannot be removed by hemodialysis.
Unfractionated heparin — For patients with intracranial bleeding associated with therapeutic doses of unfractionated heparin, protamine sulfate can be given for heparin reversal. (See "Heparin and LMW heparin: Dosing and adverse effects", section on 'Unfractionated heparin'.)
We make sure that all sources of heparin have been discontinued and that this is clearly stated in the medical record.
Protamine sulfate — Protamine sulfate (protamine) is an emergency reversal agent for unfractionated heparin.
For patients with acute intracranial hemorrhage and convincing evidence of anticoagulation with unfractionated heparin (prolonged aPTT and/or administration within the previous two hours), we recommend protamine.
A fixed dose of 50 mg or 25 mg may be given; some experts give 50 mg and others give 25 mg followed by an additional dose of 25 mg if needed (eg, based on a prolonged aPTT). The 50 mg dose would be more appropriate for an individual with a greater aPTT prolongation; however, this dose may be associated with a greater risk for relative hypotension. Alternatively, the dose can be calculated as 1 mg protamine per 100 units of heparin. The number of units of heparin is estimated based on the previous dose and the interval since it was administered, with an estimated half-life of heparin in the range of one to two hours (eg, if a dose of 5000 units was given one hour ago, the number of units would be 2500 and the dose of protamine would be 25 mg).
Protamine must be given by slow intravenous infusion, as rapid infusion may cause hypotension, particularly at high doses (eg, 50 mg). The infusion rate should not exceed 20 mg/min and the total dose should not exceed 50 mg in any 10-minute period. Repeat doses may be given for a persistently prolonged aPTT. If heparin had been given by subcutaneous injection, repeated small doses of protamine may be required because of prolonged heparin absorption from subcutaneous sites.
Protamine is derived from fish sperm and may elicit an allergic reaction, especially in previously exposed individuals. In the United States, the protamine sulfate label contains a boxed warning that the drug can cause severe hypotension, cardiovascular collapse, noncardiogenic pulmonary edema, catastrophic pulmonary vasoconstriction, and pulmonary hypertension. Risk factors include high dose or overdose, rapid administration, repeated doses, previous administration of protamine or protamine-containing insulin (eg, neutral protamine hagedorn [NPH] or protamine zinc insulin [PZI]), and certain beta-blockers. Allergy to fish, previous vasectomy, severe left ventricular dysfunction, and abnormal preoperative pulmonary hemodynamics also may be risk factors. Vasopressors and resuscitation equipment should be immediately available in case of a severe reaction to protamine.
Evidence for the efficacy and safety of protamine include a number of preclinical and observational studies that demonstrate effective reversal, as summarized in a 2016 guideline [14]. Studies evaluating clinically significant outcomes are lacking, as discussed separately. (See "Heparin and LMW heparin: Dosing and adverse effects", section on 'Urgent reversal (protamine)'.)
LMW heparin — Intracranial bleeding associated with low molecular weight (LMW) heparin (eg, enoxaparin, dalteparin, nadroparin, tinzaparin) is uncommon but can occur, particularly in patients with cancer and brain metastases. For intracranial bleeding associated with therapeutic dose LMW heparin, we suggest andexanet alfa rather than protamine sulfate. However, protamine sulfate is a reasonable alternative if andexanet is not available.
Evidence of LMW heparin effect may be based on the interval since the last dose and/or anti-factor Xa activity indicating a LMW heparin level of ≥0.3 international units/mL. (See 'Laboratory testing' above.)
We make sure that the LMW heparin has been discontinued and that this is clearly stated in the medical record.
Andexanet alfa — Despite limited data, andexanet alfa is a reasonable approach for the treatment of intracranial bleeding associated with LMW heparin anticoagulation. Studies evaluating andexanet alfa for reversal of anticoagulation suggested efficacy for achieving hemostasis in a small number of individuals receiving therapeutic dose LMW heparin (16 individuals receiving enoxaparin in the ANNEXA-4 study) [55]. Additional data are needed.
For patients with bleeding associated with therapeutic dose LMW heparin, the dosing of andexanet alfa in ANNEXA-4 was the higher dose level, with an 800 mg bolus given at 30 mg/minute over 30 minutes, followed by an infusion of 960 mg given at 8 mg/minute for up to 120 minutes [55]. The lower dose level (400 mg bolus and 480 infusion at 4 mg/minute) may be sufficient in individuals receiving prophylactic dose LMW heparin, although this has not been demonstrated, and it may be safer to use the higher dose level.
If andexanet is unavailable, protamine sulfate should be given. (See 'Protamine sulfate if andexanet alfa is unavailable' below.)
Protamine sulfate if andexanet alfa is unavailable — For patients with acute intracranial hemorrhage and evidence of LMW heparin anticoagulation (increased anti-factor Xa activity [preferred] or administration within the previous 12 hours), protamine sulfate is a reasonable treatment option if andexanet is not available.
Unlike its effect with unfractionated heparin, protamine is less effective in reversing the effect of LMW heparin. This is because protamine acts by reversing the inhibitory effect of LMW heparin on thrombin but only reverses approximately 60 percent of the inhibitory effect on factor Xa due to decreased binding to the shorter heparin chains in LMW heparin.
Dosing of protamine sulfate in LMW heparin-associated bleeding is discussed separately. (See "Heparin and LMW heparin: Dosing and adverse effects", section on 'Urgent reversal (protamine)'.)
Fondaparinux — Fondaparinux is a synthetic pentasaccharide analog of the natural pentasaccharide found in heparin. Fondaparinux acts by binding to and inducing a conformational change in antithrombin that causes selective inhibition of factor Xa. The half-life of fondaparinux is 17 to 21 hours.
For patients with intracranial hemorrhage associated with fondaparinux anticoagulation, there is little information to guide management. Andexanet alfa is a reasonable option if it is available. Although data are limited, we would use the higher dose level, with an 800 mg bolus given at 30 mg/minute over 30 minutes, followed by an infusion of 960 mg given at 8 mg/minute for up to 120 minutes.
Protamine sulfate is ineffective for fondaparinux reversal. Other options for reversal and additional information about the use of these agents are presented separately. (See "Fondaparinux: Dosing and adverse effects", section on 'Bleeding/emergency surgery'.)
RESUMPTION OF ANTICOAGULATION — In many cases, anticoagulation can be resumed after bleeding resolves, provided the patient remains stable and the risk-benefit calculation clearly favors reinitiating anticoagulation. However, the decision must be individualized. Considerations related to this decision and evidence to support the benefit of restarting anticoagulation are presented separately. (See "Management of bleeding in patients receiving direct oral anticoagulants", section on 'Resumption of anticoagulation'.)
Likewise, the optimal timing for restarting anticoagulation therapy following an intracranial bleed is unknown. The size and cause of the bleeding (eg, traumatic versus atraumatic) and patient-specific factors that increase bleeding risk may play a role in the calculation. (See "Risks and prevention of bleeding with oral anticoagulants", section on 'Prognosis and reinitiation of anticoagulation'.)
For individuals treated with high doses of vitamin K, there may be a period of refractoriness after resuming warfarin.
The decision process to reinitiate anticoagulation and the timing of reinitiation are discussed separately. (See "Spontaneous intracerebral hemorrhage: Secondary prevention and long-term prognosis", section on 'Management of antithrombotic therapy'.)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Anticoagulation" and "Society guideline links: Stroke in adults" and "Society guideline links: COVID-19 – Anticoagulation" and "Society guideline links: COVID-19 – Index of guideline topics".)
SUMMARY AND RECOMMENDATIONS
●Urgent evaluation – We perform a rapid clinical assessment to identify important features of the history and neuroimaging to distinguish hemorrhage from ischemia. Lumbar puncture (LP) is generally required when there is strong suspicion for subarachnoid hemorrhage (SAH) despite a normal head computed tomography (CT).
All patients should have a complete blood count (CBC) with platelet count, prothrombin time (PT) with international normalized ratio (INR), and activated partial thromboplastin time (aPTT). Liver function tests and a metabolic panel may also be appropriate. (See 'Urgent evaluation' above.)
Additional laboratory evaluation varies by anticoagulant:
•Dabigatran – Thrombin time (TT), creatinine determination, and calculation of the creatinine clearance
•Factor Xa inhibitors (apixaban, edoxaban, rivaroxaban, low molecular weight [LMW] heparin) – Anti-factor Xa activity calibrated to the drug, creatinine, and calculated creatinine clearance
TT and anti-factor Xa activity may not be immediately available.
●General measures for all patients – General measures for anticoagulant-associated intracranial hemorrhage (ICH) include discontinuation of all antithrombotic agents, admission with intensive monitoring, and blood pressure control. Other interventions may be needed to treat severe thrombocytopenia, anemia, and metabolic abnormalities. (See 'General measures for all anticoagulants' above.)
●Indications for anticoagulant reversal – Reversal of the anticoagulant effect is indicated in virtually all cases of documented acute ICH and in patients requiring an urgent LP to exclude SAH or infection. A rare exception is an individual for whom the risk of thrombosis is clinically more serious than the risk of hematoma expansion (eg, individual with a mechanical heart valve and a small, stable subdural hematoma). (See 'Indications for reversal and goals of treatment' above.)
●Specific reversal strategies – Reversal should be done as rapidly as possible to limit hemorrhage enlargement, which can be fatal.
•Warfarin – For warfarin-associated ICH, we recommend reversal with 4-factor prothrombin complex concentrate (4-factor PCC) rather than plasma (Grade 1B). PCC is typically given as a fixed dose of 1500 to 2000 international units at a rate of 100 units/minute (table 7). (See 'Prothrombin complex concentrate' above.)
Options if 4-factor PCC is unavailable include 3-factor PCC supplemented with a source of factor VII or a plasma product such as Fresh Frozen Plasma or Thawed Plasma. Use of plasma may delay warfarin reversal, leading to a greater risk of complications of hematoma expansion, and may cause a transfusion reaction.
Vitamin K is also given to all patients because PCC and plasma act transiently. The dose is 10 mg intravenously (which acts within several hours). Repeat vitamin K dosing may be appropriate if the PT or INR remains elevated. (See 'Intravenous vitamin K' above.)
•Dabigatran – For dabigatran-associated ICH, we suggest idarucizumab (Grade 2C). The dose is 5 g (two 2.5 g vials). If idarucizumab is not available, activated PCC (aPCC; factor eight inhibitor bypassing activity [FEIBA]) may be used. Oral activated charcoal may be given if the patient can take oral medications and the last dose of dabigatran was within the prior two hours. Dabigatran may also be removed by hemodialysis. (See 'Dabigatran' above.)
•Direct factor Xa inhibitors (eg, apixaban, edoxaban, rivaroxaban) – For direct factor Xa inhibitor-associated ICH, we suggest andexanet alfa rather than 4-factor PCC (Grade 2C). 4-factor PCC is a reasonable alternative if andexanet is unavailable. (See 'Apixaban, edoxaban, and rivaroxaban' above.)
There are two dose levels for andexanet; the choice between them depends on the anticoagulant being reversed, how much was taken, and the timing of the most recent dose. (See 'Andexanet alfa' above.)
-The higher dose level uses a bolus of 800 mg at 30 mg/minute followed by an infusion of 960 mg at 8 mg/minute.
-The lower dose level uses a bolus of 400 mg at 30 mg/minute followed by an infusion of 480 mg at 4 mg/minute.
Oral activated charcoal may be given if the patient can take oral medications and the last dose of the anticoagulant was recent (edoxaban within two hours; apixaban within six hours; rivaroxaban within eight hours). The direct factor Xa inhibitors cannot be dialyzed. (See 'Apixaban, edoxaban, and rivaroxaban' above.)
•UFH – Unfractionated heparin (UFH)-associated ICH is treated with protamine sulfate (protamine). The dose is 1 mg protamine per 100 units of heparin; if rapid calculation is not possible, a single intravenous dose of 25 or 50 mg can be given by slow infusion over at least 20 or 30 minutes. Protamine can cause allergic reactions, particularly in individuals previously exposed to protamine sulfate, protamine-containing insulin, or those with a fish allergy. (See 'Unfractionated heparin' above.)
•LMW heparin (enoxaparin, dalteparin, tinzaparin) – For LMW heparin-associated ICH, we suggest andexanet alfa rather than protamine (Grade 2C). The higher dose level (800 mg at 30 mg/minute followed by an infusion of 960 mg at 8 mg/minute) is appropriate. Protamine sulfate is a reasonable alternative if andexanet is unavailable. (See 'LMW heparin' above.)
•Fondaparinux – Treatment of fondaparinux-associated ICH is individualized. Andexanet alfa at the higher dose level is a reasonable option. (See 'Fondaparinux' above and "Fondaparinux: Dosing and adverse effects", section on 'Bleeding/emergency surgery'.)
ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Maria I Aguilar, MD, who contributed to an earlier version of this topic review.
