Should anticoagulation be resumed after intracerebral hemorrhage?
ABSTRACTIntracerebral hemorrhage (ICH) is the most feared and the most deadly complication of oral anticoagulant therapy, eg, with warfarin (Coumadin). After such an event, clinicians wonder whether their patients should resume anticoagulant therapy. The authors review the management of anticoagulation during and after anticoagulation-associated ICH.
KEY POINTS
- Given the high risk of hematoma expansion in the early phase of acute ICH, most experts recommend reversing anticoagulation immediately.
- Many clinicians start subcutaneous heparinoids in low doses 24 to 72 hours after ICH to prevent deep vein thrombosis, and after the first few days or a week, consider either increasing the dose to a full anticoagulation dose or making a transition to oral anticoagulants.
- Many patients with lobar hemorrhage or cerebral amyloid angiopathy may remain at higher risk of anticoagulant-related ICH recurrence than thromboembolic events and would therefore be best managed without anticoagulants.
- Those with deep hemispheric ICH, hypertension that can be well controlled, and a high risk of disabling thromboembolism may receive net benefit from restarting anticoagulation.
Risk of thromboembolism after ICH: Ongoing and cumulative
Thromboembolism after ICH is a major concern, for two main reasons.
First, patients on oral anticoagulation typically have a preexisting risk factor and are thus at higher risk of a thromboembolic event, particularly while they are off anticoagulation. Patients with atrial fibrillation or a mechanical valve are at risk of arterial events such as ischemic stroke, whereas patients with a known venous thromboembolic condition such as deep venous thrombosis or pulmonary embolism are at risk of extension of the thrombosis or recurrence of a venous thrombotic event.
Second, ICH itself increases the risk of arterial and venous thromboembolic events. Including patients not previously on anticoagulation, this risk is as high as 7% during the initial hospitalization and 9% during the first 90 days.45,46 Worth noting is that patients who previously received anticoagulant drugs (and who are off this therapy in the acute phase) are at no higher risk of thromboembolism compared with those who never received anticoagulants.45
However, while the risk of hematoma expansion is highest at presentation and then decreases with time, the risk of thromboembolism (particularly venous thromboembolism) is ongoing and cumulative. Arterial thromboembolism is more likely to occur early, within the first week, whereas venous thromboembolism can occur later.45
Overall, studies have estimated the short-term risk of pulmonary embolism to be 1% to 2%, deep venous thrombosis 1% to 4%, myocardial ischemia about 2%, and cerebral ischemia 2% to 3%.45,46 However, when patients are actively screened, the incidence of asymptomatic deep venous thrombosis is found to be as high as 16% in the first 10 days,47 and evidence of myocardial ischemia can be detected in up to 27% of patients.48
Therefore, the risk of hematoma expansion appears to be high and the risk of thromboembolism appears to be low during the first day after ICH. Over the next days, as the risk of hematoma expansion recedes, this ratio shifts.
Studies of in-hospital anticoagulation after ICH
The data on restarting oral anticoagulation in the acute phase are sparse. In practice, clinicians typically start heparinoids in low subcutaneous doses to prevent deep venous thrombosis and, after the first few days or a week, consider increasing to a full anticoagulation dose or starting an oral anticoagulant and subsequently discontinuing the heparin when the INR is in the therapeutic range (see discussion below).
ICH patients in general may benefit from starting prophylactic-dose heparin therapy early. One randomized trial found that starting heparin in a low subcutaneous dose the day after an ICH decreased the risk of thromboembolism without increasing the risk of rebleeding.49 Another study also found no increased risk of rebleeding with early prophylactic-dose subcutaneous heparin.50
As the benefit appears to outweigh the risk, national guidelines suggest starting subcutaneous heparin early in all ICH patients, including those not previously on warfarin.42,43
Commonly used heparinoid regimens include unfractionated heparin 5,000 units subcutaneously twice a day; enoxaparin (Lovenox) 40 mg once a day; and dalteparin (Fragmin) 5,000 units once a day.51 In addition, all patients should receive optimal mechanical thromboprophylaxis, including graduated compression stockings or intermittent pneumatic compression stockings, or both.
LONG-TERM MANAGEMENT: ICH RECURRENCE VS THROMBOEMBOLISM
Risk of ICH recurrence on warfarin is not precisely known
Overall, the risk of ICH recurrence is about 1% at 3 months, and warfarin likely increases this risk.42,52 Unfortunately, the risk of ICH recurrence in patients on anticoagulation therapy after a first ICH is not clear, and no population-based study has clarified this risk. Therefore, the best we can do at present is to try to estimate the risk of recurrent warfarin-related ICH by separately examining two issues:
- The risk of ICH recurrence in general
- The risk of major bleeding (including ICH) in the general population of patients on warfarin.
The risk of ICH recurrence in general is about 2% to 4% per patient-year.52–54 However, this risk appears to be a function of the underlying vasculopathy. ICH location is often used as a surrogate for underlying cause. Most ICHs in deep hemispheric (basal ganglia, thalamus) or brainstem territories are likely caused by hypertensive vasculopathy, whereas lobar ICH is often associated with cerebral amyloid angiopathy.52–54 Presumably because of this distinction, ICH in a deep location recurs in about 2% of cases per year, compared with 4% for lobar ICH.53 The presence and number of microbleeds on T2-weighted gradient-echo magnetic resonance imaging appear to predict ICH recurrence; microbleeds likely are markers of more severe or widespread underlying vasculopathy.55–57
A genetic risk factor for the recurrence of lobar ICH is apolipoprotein E genotype58; future studies may highlight genetic variations that specifically modify the risk of warfarin-related ICH.59 Unfortunately, there is currently no way to modify the risk of ICH associated with cerebral amyloid angiopathy. On the other hand, in patients with hypertensive hemorrhage, antihypertensive therapy likely reduces the risk of recurrent ICH. One randomized controlled trial showed that such therapy decreased the risk of ICH by more than half.60
The risk of major bleeding in the general population of patients on warfarin may be 2% to 3% per year and is likely higher in the first month.11 The risk is higher in older patients and if the INR rises above 4.0.11,17 For some patients, it is possible to estimate the likelihood of major bleeding using validated decision-support tools that include factors such as age, sex, and medical history.11,61–64
Given the lack of data specifically addressing the risk of ICH recurrence on warfarin, the clinician is left to try to extrapolate this risk from available data using specific patient characteristics that modify the presumed risk. For example, one can combine factors such as ICH location (or better yet, the underlying cause) with decision-support tools that predict the risk of major bleeding. Close control of both blood pressure and the INR appears especially critical for patients receiving anticoagulation after ICH.11,60,65 Still, the risk does not disappear with good INR control, and most patients with anticoagulation-related ICH present with INRs within the therapeutic range.5,10,65
