Spontaneous intracerebral hemorrhage: Secondary prevention and long-term prognosis

INTRODUCTION — Spontaneous intracerebral hemorrhage (ICH) is often associated with long-term neurologic symptoms, and patients with ICH have an elevated risk of recurrence. Prevention of recurrent ICH (ie, secondary prevention) may reduce accumulating neurologic disability as well as societal burden of ICH.

ICH may be categorized as either spontaneous or traumatic. ICH following traumatic brain injury is reviewed separately. (See "Traumatic brain injury: Epidemiology, classification, and pathophysiology".)

This topic will review the epidemiology, secondary prevention, and long-term prognosis in adults with spontaneous ICH. Other aspects of ICH are discussed elsewhere.

●(See "Spontaneous intracerebral hemorrhage: Pathogenesis, clinical features, and diagnosis".)

●(See "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis".)

●(See "Cerebral amyloid angiopathy".)

●(See "Hemorrhagic stroke in children".)

●(See "Stroke in the newborn: Management and prognosis".)

RISK OF RECURRENCE

Incidence — The incidence of recurrent ICH varies from 2 to 7 percent per year, depending on risk factors [1,2]. Patients with a history of ICH have a risk of recurrent ICH that is higher than the risk of recurrence in patients with ischemic stroke [1-3]. In addition, the rate of recurrent ICH was 6.6-fold higher than for a first ICH in a study of patients with prior ischemic stroke [1].

The risk of ICH recurrence may be highest in the first 12 months after the initial ICH but persists for years after the first event, particularly after lobar ICH [4]. The cumulative risk of ICH recurrence varies from 1.3 to 8.9 percent after one year and ranges from 7.4 to 13.7 percent after five years in different populations [2,3,5,6].

Risk factors

●Initial ICH location and etiology – Deep (nonlobar) ICH involving the basal ganglia, thalamus, cerebellum, or brainstem is associated with a lower risk of recurrence than ICH in lobar locations. The annual risk of ICH recurrence after deep ICH is approximately 2 to 3 percent, versus 7 to 14 percent after lobar ICH [2,7-9]. ICH involving deep nuclei is often attributed to hypertensive microvascular disease and lobar ICH is often attributed to cerebral amyloid angiopathy (CAA), but the risk of recurrence appears to be independent of ICH etiology, at least in part.

CAA is a major cause of incident and recurrent lobar ICH [10]. A meta-analysis of 10 prospective cohorts of ICH patients found that the annual risk of ICH recurrence after CAA-related ICH was 7.4 percent (95% CI 3.2-12.6), versus 1.1 percent (95% CI 0.5-1.7) for non-CAA-related ICH [11]. (See "Cerebral amyloid angiopathy".)

Other secondary causes of ICH, such as brain arteriovenous malformation, moyamoya syndrome, sickle cell disease, and brain tumors are also associated with elevated risks of recurrent ICH based on the nature of the underlying causes and the presence of specific associated high-risk features. These are discussed separately. (See "Brain arteriovenous malformations" and "Moyamoya disease and moyamoya syndrome: Etiology, clinical features, and diagnosis" and "Prevention of stroke (initial or recurrent) in sickle cell disease" and "Overview of the clinical features and diagnosis of brain tumors in adults".)

●Other imaging features – The presence and number of cerebral microbleeds (CMBs) and/or superficial siderosis on brain magnetic resonance imaging (MRI) identifies patients at high risk for recurrent ICH [11,12]. In one study, the presence of >1 CMB was associated with increased risk of recurrent ICH in patients with CAA-related ICH while a higher threshold, >10 CMBs, identified increased risk of a recurrent event in patients with non-CAA-related ICH [11].

Hyperintensities on diffusion-weighted imaging (DWI) brain MRI sequences found in some patients with ICH may be a marker of severe small-vessel disease associated with the risk of ICH recurrence [13,14]. In one study of 247 patients with ICH, those with DWI hyperintensities had a higher risk of recurrent ICH but not subsequent ischemic stroke at two years [15].

●Hypertension – Hypertension (HTN) is the most consistent risk factor for ICH recurrence. HTN predisposes to ICH recurrence of both deep and lobar ICH [4]. Inadequate control of HTN is common and increases the risk of recurrent ICH [4,16,17]. In a longitudinal study of 1145 patients with ICH, each 10 mmHg increase in systolic blood pressure was associated with an incremental increase in risk of recurrent lobar ICH (hazard ratio [HR] 1.33, 95% CI 1.02-1.76) and recurrent deep ICH (HR 1.54, 95% CI 1.03-2.30) [18].

The risks of ICH related to inadequate blood pressure control and management to mitigate these risks are discussed below. (See 'Blood pressure management' below.)

●Age – The risk of ICH recurrence with age is based largely on the association of advancing age with the risk of initial ICH [19]. A meta-analysis of more than 8100 patients with ICH assessed the age-related incidence of ICH over a 28-year period [20]. Using the age group of 45 to 54 years as reference, the incidence ratio increased from 0.10 (95% CI 0.06-0.14) for those under 45 years up to 9.6 (95% CI 6.6-13.9) for patients older than 85 years. These findings are likely true for ICH recurrence as well.

Older age is also associated with higher prevalence of CAA and higher use of antithrombotic drugs for accumulating cardiovascular comorbidities. (See "Spontaneous intracerebral hemorrhage: Pathogenesis, clinical features, and diagnosis", section on 'Risk factors'.)

●Medications – Antithrombotic medications (including antiplatelet agents and anticoagulants) and statins may be associated with risk of ICH recurrence. In addition, several other substances including selective serotonin reuptake inhibitors and nonsteroidal anti-inflammatory drugs have been linked to increased risk of bleeding in general, including ICH and ICH recurrence (table 1). However, studies examining the associations between these drugs and risk of ICH recurrence have yielded inconsistent results [3,18,21,22].

The potential risks of recurrent ICH due to antithrombotic and statin medications are discussed below. (See 'Management of antithrombotic therapy' below and 'Management of statins' below.)

●Other risk factors

•Race and ethnicity – There are racial and ethnic disparities in the risk of ICH recurrence. Black American, Hispanic American, and Asian American patients with a history of ICH seem to be at higher risk for a recurrent event than White American patients [16,23]. The prevalence of HTN in these groups does not fully account for this elevated risk of ICH. In one study, patients remained at higher risk of ICH recurrence after adjusting for blood pressure measurements and variability [16].

•Chronic kidney disease – Chronic kidney disease can be a marker of atherosclerotic disease and may further contribute to the risk of ICH through renally mediated impairment of cerebral autoregulation [24]. A large population-based study in Denmark evaluated 15,270 patients with ICH and found that patients with kidney failure at the time of the initial ICH were at higher risk for ICH recurrence (relative risk 1.72, 95% CI 1.34-2.17) [3].

•Prior ischemic stroke or ICH – The risk of a future ICH is higher in patients with history of prior ICH and those with history of a prior ischemic stroke [1].

•Genetic features – Certain genetic features associated with CAA are associated with increased risk of ICH recurrence [25]. Patients with ICH who are carriers of apolipoprotein-E (APOE) e2 or e4 genotypes, frequently associated with CAA, are at elevated risk of ICH [9].

Additionally, patients with genetic bleeding disorders also are at risk for ICH. These disorders are discussed separately. (See "Clinical presentation and diagnosis of von Willebrand disease" and "Clinical manifestations and diagnosis of hemophilia" and "Rare inherited coagulation disorders".)

FOLLOW-UP NEUROIMAGING — All patients whose symptoms unexpectedly fail to improve or worsen during the recovery period require neuroimaging to evaluate for a recurrent hemorrhage.

In addition, follow-up imaging studies can help to identify or confirm the cause of the ICH, which in turn determines the risk of recurrence and may help guide preventive measures. For some patients, neuroimaging studies performed during the acute hospitalization identify the etiology such that further imaging studies are not required. For other patients, the initial imaging study does not sufficiently exclude other causes of ICH and follow-up studies are required. When acute bleeding and surrounding edema from the acute ICH shroud and distort underlying brain structures, delayed imaging performed after bleeding and edema have resolved may identify patients who are at high risk for recurrence due to an underlying structural cause (algorithm 1). (See "Spontaneous intracerebral hemorrhage: Pathogenesis, clinical features, and diagnosis", section on 'Subsequent imaging'.)

●Patients with suspected hypertensive ICH – Patients with ICH attributed to hypertension (HTN) who continue to improve clinically during recovery may not warrant additional imaging.

Clinical and imaging features on head computed tomography (CT) or brain magnetic resonance imaging (MRI) suggestive of hemorrhage related to HTN or other atherosclerotic risk factors associated with deep penetrating vasculopathy include:

•Hematoma or cerebral microbleeds (CMBs) in basal ganglia or thalamus, cerebellar nuclei, or brainstem (image 1)

•Known history or new diagnosis of HTN

•No prior ICH (unless in setting of uncontrolled HTN)

•No atypical clinical or neuroimaging features (table 2)

•Patient age ≥65 years

Clinically stable patients who meet most or all of the criteria listed above likely do not require a follow-up imaging study. For patients with only some of these features, we suggest repeating a brain MRI with gadolinium contrast in 12 to 16 weeks after the ICH to assess for alternative secondary causes.

Additional associated imaging features found in some patients with a hypertensive ICH include evidence of prior chronic ischemic stroke attributed to small vessel (penetrating artery) and CMBs located in the basal ganglia or thalamus evident on T2*-weighted brain MRI sequences.

●Patients with suspected CAA-related ICH – Some patients with ICH attributed to cerebral amyloid angiopathy (CAA) whose symptoms continue to improve during recovery may not require additional imaging in the ambulatory setting. Imaging features on brain MRI suggestive of ICH related to CAA include lobar location and evidence of lobar CMBs or cortical superficial siderosis in an older patient (image 1 and image 2). The approach to confirming that diagnosis is discussed separately. (See "Cerebral amyloid angiopathy", section on 'Diagnostic approach'.)

●Other patients — We obtain follow-up imaging for patients with clinical or imaging features of the ICH suggestive of an underlying cause but not identified or excluded during the acute hospitalization (table 2).

Clinical features that raise suspicion for an underlying cause of ICH include:

•Age <65 years

•No history or new diagnosis of HTN

•History of protracted new-onset headaches

•History of new-onset neurologic symptoms preceding ICH

•Thunderclap headache at onset of hemorrhage

•History of prior ICH (unless attributed to uncontrolled HTN or CAA)

Imaging features of the hemorrhage on imaging (head CT or brain MRI) raising suspicion for other secondary causes of ICH (such as a vascular lesion, primary or metastatic brain tumor, cerebral venous thrombosis, or hemorrhagic transformation of ischemic infarct) include:

•Early perihematomal edema out of proportion to the size of the ICH (image 3)

•Hemorrhage appears to be in arterial vascular territory suggesting primary ischemic infarction (image 4)

•Enhancement of intracranial vessels around ICH (image 5)

•Multifocal hemorrhage (image 6)

•Isolated intraventricular hemorrhage (image 7)

Specific underlying etiologies of nontraumatic ICH are discussed separately. (See "Spontaneous intracerebral hemorrhage: Pathogenesis, clinical features, and diagnosis", section on 'Specific etiologies'.)

We prefer brain MRI with gadolinium for most patients who undergo post-acute testing to identify underlying cause of ICH. MRI should include T2*-weighted (gradient echo [GRE] or susceptibility-weighted imaging [SWI]) sequences. In an observational study in 400 patients with spontaneous ICH, MRI performed within 30 days improved diagnostic accuracy regarding ICH etiology over CT, changing the diagnostic impression in approximately 14 percent and management in 20 percent of cases [26]. These findings were confirmed in a subsequent study of 123 patients where MRI was most useful for establishing the diagnosis of ICH secondary to cerebral venous sinus thrombosis, hemorrhagic transformation of an ischemic infarct, neoplasms, and vascular malformations [27].

For patients who are unable to undergo MRI, head CT with contrast is a reasonable but less sensitive alternative.

When an underlying vascular cause is suspected, additional noninvasive vascular imaging with CT or MR angiogram should be performed. Digital subtraction angiography (DSA) is performed when CT or MR angiography is inconclusive or is negative and clinical suspicion for an underlying vascular lesion remains high (image 5) [28,29]. In addition, we generally obtain DSA to evaluate for a vascular lesion such as an arteriovenous malformation. CT or MR venography is performed for suspected venous lesions, such as cerebral venous thrombosis. (See "Brain arteriovenous malformations" and "Cerebral venous thrombosis: Etiology, clinical features, and diagnosis".)

The optimal time to obtain follow-up imaging depends on the indication and the clinical recovery of the patient.

•For patients who did not undergo evaluation during the acute hospitalization for potential underlying secondary causes of the ICH, we obtain early imaging in four weeks to identify potential underlying structural sources amenable to early treatment.

•For other patients with a suspected secondary cause where the acute evaluation did not identify or exclude an underlying source, we obtain imaging after six to eight weeks to allow for better visualization of underlying brain tissue after some resorption of the hematoma.

•For patients without worrisome clinical or imaging features of the ICH for whom the acute evaluation for underlying causes did not identify the etiology, repeat imaging is indicated to exclude underlying structural sources. For these patients, we typically delay repeat imaging for 12 to 16 weeks after the ICH to promote optimal visualization of underlying brain tissue and reduce the risk of identifying confounding abnormal imaging findings that may be attributed to healing of the ICH in the late subacute time period.

BLOOD PRESSURE MANAGEMENT — Blood pressure (BP) control is an important aspect of reducing the risk of recurrent ICH. Hypertension (HTN) is one of the single most important modifiable risk factors for initial ICH and ICH recurrence. (See 'Risk factors' above.)

Small improvements in BP control help to reduce the risk of ICH. The relationship between BP control and ICH recurrence has been studied somewhat indirectly in patients with either ischemic stroke or ICH. In the Perindopril Protection Against Recurrent Stroke Study (PROGRESS) trial, patients with prior stroke (hemorrhage or infarction) were randomly assigned to an angiotensin-converting enzyme inhibitor (perindopril) with or without a diuretic (indapamide) versus placebo. A modest BP reduction rate by 9/4 mmHg in patients assigned to active treatment reduced the risk of ICH from 2.4 to 1.2 percent, corresponding to a 50 percent relative risk reduction (95% CI 26-67 percent) [30]. The relative risk reduction for recurrent stroke was 49 percent (95% CI 18-68 percent) among patients whose qualifying event was ICH [30-32].

Blood pressure goals — We suggest aiming for BP <130/80 mmHg as a long-term target to reduce the risk of recurrence after ICH (table 3) [33]. The risk of ICH is reduced with each incremental reduction in BP, but the benefit may be greatest for patients whose BP reaches intensive BP-lowering targets [18,31].

The benefit for intensive BP reduction has not been demonstrated specifically for ICH recurrence; however, indirect evidence from patients with other cerebrovascular conditions suggests a likely benefit. In the Secondary Prevention of Small Subcortical Strokes (SPS3) trial, 3020 patients with prior ischemic stroke were assigned either to a systolic BP target of 130 to 149 mmHg or <130 mmHg [34]. During a mean follow-up of 3.7 years, there were fewer ICH events in those assigned to the intensive BP target (6 versus 16), corresponding to a lower rate of ICH at 0.1 percent per patient-year in the intensive group compared with 0.3 percent per patient-year in the higher target.

A more intensive BP target was assessed in the Recurrent Stroke Prevention Clinical Outcome (RESPECT) trial, in which 1266 patients with ischemic stroke were randomly assigned to intensive BP control (<120/80 mmHg) or to standard treatment (<140/90 mmHg) [35]. There was a trend toward fewer strokes in patients assigned to the intensive group; however, limitations in this study prevent firm conclusions. The actual mean BP achieved in the intensive group was 127/77, not very different from 133/78 mmHg in the standard group. The trial was stopped early with relatively few (12) ICH events, most in the standard treatment group (11 versus 1).

In a meta-analysis of these studies along with two additional trials in patients after ischemic stroke, more intensive BP lowering (variably defined) was associated with a reduced risk of ICH (relative risk [RR] 0.25, 95% CI 0.07-0.90) [35].

Blood pressure lowering has additional benefits in reducing the risk of other vascular events including ischemic stroke, although in the population of patients with prior ICH, the absolute benefits in this regard are likely to be small [35].

When to begin antihypertensive therapy — The approach to BP control after ICH is stepwise. However, high quality data to specify when to safely implement BP control after ICH are lacking.

In the acute (typically hospital) setting, initial steps to control the BP are begun immediately to prevent hematoma expansion. The benefit of acute control of elevated blood pressure must be balanced against the competing risk of cerebral or other organ hypoperfusion. Initial target systolic blood pressure is 140 to 160 mmHg for most patients. (See "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis", section on 'Blood pressure management'.)

Further reductions in BP toward normotension are made in a stepwise fashion. For most patients, we aim to achieve the BP target of <130/80 mmHg within 10 to 14 days after the onset of ICH. We also advise close outpatient monitoring to ensure that BP control is maintained.

Selection of antihypertensive agent — For patients without a clinical indication for a specific agent, we typically start an angiotensin-converting enzyme (ACE) inhibitor. For those unable to tolerate or who prefer an agent other than an ACE inhibitor, we offer an angiotensin receptor blocker, thiazide diuretic, or calcium channel blocker, based on analyses of risk reduction in patients with atherosclerotic disease [36,37]. The choice of agent should be guided by efficacy in achieving target BP. Clinical indications to guide individualized medication selection are discussed elsewhere (table 4). (See "Choice of drug therapy in primary (essential) hypertension".)

The use of combination antihypertensive regimen (ie, multiple antihypertensive drugs, instead of a single drug) may decrease the risk of adverse effects and improves tolerability and compliance [38]. (See "Choice of drug therapy in primary (essential) hypertension".)

Education of patients and their caregivers about target BP and their engagement in self-monitoring at home and communications with their medical providers are also key to achieve better BP control and to improve adherence to therapy [39]. Lifestyle modifications and management of obstructive sleep apnea and obesity are fundamental components of BP management. (See 'Lifestyle modifications' below.)

MANAGEMENT OF ANTITHROMBOTIC THERAPY — Many patients with ICH have comorbid cardiovascular conditions and may have indications for antiplatelet or oral anticoagulant agents. Whether to resume or discontinue these medications after ICH requires weighing the competing risks of thromboembolic events versus ICH recurrence. In the absence of high-quality trial data, observational reports and expert opinion guide risk/benefit assessment and decision-making.

Antiplatelet therapy — Antiplatelet therapy is typically withheld in the acute setting to mitigate the risk of hemorrhage expansion. (See "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis", section on 'Reverse anticoagulation'.)

We suggest resuming antiplatelet therapy after ICH for most patients who have a specific indication for such therapy. However, it is important to balance individual risks and benefits.

●Patients with established atherosclerotic disease – We resume antiplatelet therapy for most patients with nonlobar ICH and those with lobar ICH attributed to cerebral amyloid angiopathy (CAA) who have established atherosclerotic disease, in agreement with guidelines from the American Heart Association (AHA) [33]. Such indications may include prior cardiovascular disease, ischemic stroke, or peripheral arterial disease. These indications are discussed separately. (See "Prevention of cardiovascular disease events in those with established disease (secondary prevention) or at very high risk" and "Long-term antithrombotic therapy for the secondary prevention of ischemic stroke" and "Aspirin for the secondary prevention of atherosclerotic cardiovascular disease".)

The risks and benefits of resuming antiplatelet medications for patients with CAA are discussed in detail separately. (See "Cerebral amyloid angiopathy", section on 'Prevention of recurrent hemorrhage'.)

If lobar ICH was attributed to an alternative source, we base decisions on resuming antiplatelet therapy on the individual risks associated with the underlying etiology.

We prefer low-dose aspirin (81 mg per day), typically starting a few days after ICH, if neuroimaging confirms stability. Low-dose aspirin has been most studied in relationship to ICH recurrence. Some [4,40-42] but not all [43] small studies have reported no difference in the rates of ICH recurrence among patients with ICH who continued aspirin and those who discontinued it. In the Restart or Stop Antithrombotics Randomized Trial (RESTART), 537 patients who developed ICH while taking antithrombotic therapy were assigned either to continue or discontinue antiplatelet therapy [44]. Most patients (88 percent) were taking antithrombotic therapy for secondary prevention of atherosclerotic disease; 25 percent had atrial fibrillation, either as a comorbid condition or as their primary indication. After a median two years of follow-up, the risk of recurrent ICH was similar (nonsignificantly lower) in patients who continued versus discontinued antiplatelet therapy (4 versus 9 percent; adjusted hazard ratio [aHR] 0.51, 95% CI 0.25-1.03), while rates of major occlusive and thromboembolic events were higher and were similar among treatment groups (15 versus 14 percent; aHR 1.02, 95% CI 0.65-1.60). These findings were sustained in an extended follow-up at a median time of three years (interquartile range two to five years) [45]. Hemorrhagic and thromboembolic outcomes were similar among patients who continued versus those who discontinued antiplatelet therapy, and the rate of recurrent ICH remained lower than the rate of major vascular events (9 versus 30 percent).

●Primary prevention of atherosclerotic disease – For patients with ICH without established atherosclerotic disease, we consider individual risk factors to weigh overall benefits of antiplatelet therapy against the risk of hemorrhage. As examples, we typically would resume aspirin for patients with ICH and hypertension (HTN), hyperlipidemia, and diabetes mellitus or those with carotid atherosclerotic disease. For such patients, ICH may be a marker of atherosclerotic risk. In a Danish cohort study, patients with prior ICH had a higher risk of subsequent cardiovascular events than age- and sex-matched controls, including ischemic stroke (1.5 versus 0.6 per 100 person-years) and major adverse cardiovascular events (4.2 versus 1.4 per 100 person-years) [46]. (See "Aspirin in the primary prevention of cardiovascular disease and cancer".)

For most of these patients in whom antiplatelet therapy is resumed, we prefer low-dose aspirin and restart such therapy several days after the ICH has stabilized.

For patients with lobar ICH and suspected CAA without high risk of ischemic stroke or cardiovascular events, we avoid antiplatelet therapy. (See "Cerebral amyloid angiopathy".)

●Patients with intravascular stents – Patients with symptomatic atherosclerotic disease who have undergone intravascular stent placement are typically prescribed antiplatelet therapy for several months to prevent vascular occlusion from thrombosis at the site of the stent. We typically resume these medications in patients with ICH because of the thrombotic risks related to their discontinuation and typically start within a few days after ICH if neuroimaging confirms stability. Whenever feasible, we prefer single antiplatelet therapy over dual antiplatelet therapy. (See "Antithrombotic therapy for elective percutaneous coronary intervention: General use" and "Long-term antiplatelet therapy after coronary artery stenting in stable patients" and "Overview of carotid artery stenting" and "Endovascular techniques for lower extremity revascularization", section on 'Antiplatelet therapy'.)

●Patients taking nonsteroidal antiinflammatory drugs – We prefer nonacetylated salicylates (eg, magnesium salicylate) over other nonsteroidal antiinflammatory medications with antithrombotic properties that impair platelet function.

Anticoagulation — Anticoagulation is typically withheld, and the effects are reversed acutely for patients with ICH, to reduce the risks of hemorrhagic expansion and associated morbidity. (See "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis", section on 'Reverse anticoagulation'.)

Many patients with ICH benefit from resuming anticoagulation when the thromboembolic risk is higher than the risk of recurrent ICH. However, it is important to balance individual risks and benefits.

●Individualize decision to resume or discontinue − We balance the risks of recurrent ICH with the risk of thromboembolism to help make decisions about resuming anticoagulation for individual patients (algorithm 2). Only limited observational data and expert opinion are available to support clinical decisions regarding resuming or withholding anticoagulation [47-50]. These decisions should be made along with the patient after weighing the individualized risks, benefits, and exploring alternative options whenever possible.

●Timing of resumption − The optimal time for restarting oral anticoagulation after ICH has not been established and may depend on the underlying indication for anticoagulation. Early resumption of anticoagulation within several days after stabilization of the ICH may be indicated for select patients with a compelling indication (eg, mechanical prosthetic heart valve) [33]. (See 'Mechanical prosthetic heart valves' below.)

For most other patients who resume anticoagulation, we generally suggest delaying restarting oral anticoagulants for four to eight weeks after onset of the ICH, in agreement with AHA guidelines [33]. We use hemorrhage size and thromboembolic risks to guide the specific timing of resumption for an individual patient. In one study of 177 patients with intracranial hemorrhage and an indication for anticoagulation, the combined risk of recurrent intracranial hemorrhage or ischemic stroke reached a nadir when warfarin was resumed after 10 weeks, suggesting that the optimal timing for resumption of oral anticoagulation is after 10 weeks [51]. The risk of ischemic stroke was lowest and the risk of recurrent hemorrhage was highest within the first five weeks after the initial hemorrhage, suggesting anticoagulation may be delayed during this time interval.

Mechanical prosthetic heart valves — Resumption of warfarin is recommended for most patients with mechanical prosthetic valves who develop ICH while taking warfarin because the ongoing risk of thromboembolic events is higher than the risk of recurrent ICH, regardless of hemorrhage etiology. In a meta-analysis of more than 13,000 patients with mechanical heart valves, the incidence of major embolism was four times higher among those not on antithrombotic therapy versus those taking warfarin (4 versus 1 per 100 patient-years) [52]. A prosthesis in the mitral position increased the thromboembolic risk almost twice as compared with the aortic position. (See "Antithrombotic therapy for mechanical heart valves".)

CAA-related ICH — Many patients with lobar ICH do not resume anticoagulation because of the associated risk of recurrent ICH attributed to CAA outside a compelling indication such as a mechanical heart valve. The specific risks of future ICH for individual patients with CAA should be weighed against the benefits of resuming anticoagulation. These risks and benefits of anticoagulation for patients with CAA are discussed separately. (See "Cerebral amyloid angiopathy", section on 'Prevention of recurrent hemorrhage'.)

Atrial fibrillation — In the absence of high-quality trial data, the decision to resume or withhold anticoagulation in patients with ICH and atrial fibrillation requires balancing future hemorrhagic and thromboembolic risks at an individual level.

For many patients with atrial fibrillation, ischemic stroke is more common than recurrent ICH and the risk-benefit analysis favors resuming anticoagulation after ICH [6]. In a 2017 meta-analysis of eight studies including 5306 patients with anticoagulation-associated ICH, restarting anticoagulation after ICH was associated with a lower risk of thromboembolic complications and no excess risk of ICH recurrence [53]. Most patients resumed warfarin and atrial fibrillation was the most common indication for restarting anticoagulation. Resumption of oral anticoagulation was also associated with reduced risk of all-cause stroke and mortality at 12 months in an analysis of 1012 patients with warfarin-associated lobar and nonlobar ICH [54]. Another meta-analysis of 50,470 patients with spontaneous or anticoagulation-associated intracranial hemorrhage and atrial fibrillation also found that resuming anticoagulation was associated with lower risk of subsequent thromboembolism without excess risk of recurrent intracranial hemorrhage [55]. However, interpretation of these meta-analyses is limited by heterogeneity of included studies, their retrospective and observational nature, and inherent selection, indication, and prescription biases.

Estimating bleeding and thromboembolic risks — Several clinical prediction scores have been developed to help quantify individual risks of future bleeding and thromboembolism. The CHADS2 or CHA2DS2-VASc scores to assess thromboembolic risks are used widely (table 5 and algorithm 2).

Other scores have been developed to help estimate bleeding risk. Among these, the HAS-BLED score incorporates clinical risk factors associated with bleeding to help to assess initial hemorrhagic risk (scored 1 to 9) in patients with atrial fibrillation (table 6) [56]. However, its generalizability is limited by the small number of patients with risks who scored 5 to 9 and may also be restricted to the assessment of initial ICH risk among patients taking warfarin. Additionally, subjective clinical assessment of bleeding risk may have a similar predictive accuracy to bleeding scores [57]. (See "Risks and prevention of bleeding with oral anticoagulants", section on 'Bleeding risk scores' and "Atrial fibrillation in adults: Selection of candidates for anticoagulation", section on 'CHA2DS2-VASc score'.)

As examples:

●For a typical patient with nonlobar ICH attributed to HTN without additional hemorrhagic risk factors who has atrial fibrillation and a CHA2DS2-VASc score ≥2, we may resume anticoagulation once HTN is controlled.

●For a typical patient with lobar ICH attributed to CAA without additional hemorrhagic risk factors who has atrial fibrillation and a CHA2DS2-VASc score ≥2, we may pursue alternatives to anticoagulation.

●For a typical patient with non-lobar ICH of undetermined source who has atrial fibrillation a CHA2DS2-VASc score ≥2, we would exclude underlying sources prior to considering resuming anticoagulation.

One study using a decision-analysis model to compare warfarin resumption versus discontinuation after ICH found that resumption improves quality-adjusted life (QoL) expectancy in some patients. For patients with lobar ICH, warfarin discontinuation improves QoL expectancy by 1.9 years and is therefore preferred unless the rate of ICH recurrence is estimated to be <1.4 percent per year [58]. By contrast, for patients with deep ICH, resumption of warfarin may be preferred if the rate of recurrent ICH is low (eg, <1.6 percent per year) and the rate of ischemic stroke is high (eg, >7 percent per year). This analysis was limited to patients taking warfarin.

Therapeutic options — Therapeutic options for patients with ICH and atrial fibrillation include:

●Resumption of anticoagulation – For patients with ICH who have atrial fibrillation, anticoagulation may be resumed when the associated risk of thromboembolism is higher than the future hemorrhagic risk. (See 'Estimating bleeding and thromboembolic risks' above.)

For long-term anticoagulation, options include a direct oral anticoagulant (DOAC) or a vitamin K antagonist, such as warfarin. For most patients with atrial fibrillation who develop ICH while on an oral anticoagulant and in whom oral anticoagulation is resumed, we suggest a DOAC over warfarin because of a favorable hemorrhagic risk profile. For select patients, warfarin may be preferred for specific indications, described immediately below.

•Direct oral anticoagulants – DOACs are at least as effective as warfarin for prevention of thromboembolic events in patients with atrial fibrillation [59]. Because they are associated with a lower risk for ICH, they are an appealing option to help reduce the risk of recurrent ICH [60]. In a study of patients with atrial fibrillation or venous thromboembolism resuming anticoagulation, there was a trend toward fewer ICH recurrences in those taking a DOAC [61]. The incidence rate of recurrent ICH was 2.5 per 100 patient-years with warfarin and 1.3 per 100-patient years with a DOAC (risk ratio 1.9, 95% CI 0.6-7.4) [61]. There were too few recurrent events to assess a difference among specific DOACs.

•Warfarin – Much of the experience with resumption of anticoagulation has been from patients taking warfarin [48,50,51]. Warfarin may be chosen based on cost or availability and is indicated for patients with specific indications, including some patients with atrial fibrillation associated with valvular heart disease and those with mechanical prosthetic heart valves. (See 'Mechanical prosthetic heart valves' above and "Atrial fibrillation in adults: Use of oral anticoagulants", section on 'Patients with valvular heart disease'.)

Warfarin may be preferred by patients previously taking the medication whose international normalized ratio (INR) is well-controlled. Additionally, warfarin may be preferred for other patients, including some with bodyweight <60 kg or age ≥80 years, and those unable to take a DOAC (eg, drug interaction, creatinine clearance <30 mL/minute). (See "Direct oral anticoagulants (DOACs) and parenteral direct-acting anticoagulants: Dosing and adverse effects", section on 'Settings in which a heparin or vitamin K antagonist may be preferable'.)

●Left atrial appendage occlusion – Percutaneous left atrial appendage occlusion (LAAO) may be a viable treatment option for patients with ICH and nonvalvular atrial fibrillation who are at high risk for recurrent ICH and thromboembolic events and in whom resumption of oral anticoagulation is not resumed or contraindicated [33]. In a meta-analysis of trials comparing LAAO closure with oral anticoagulation, the rates of both systemic embolism and major bleeding were similar after a mean follow-up of 39 months [62]. The risk of ICH was lower for patients assigned to LAAO (0.5 versus 2.4 percent; relative risk 0.22, 95% CI 0.02-0.58). In the Amplatzer Cardiac Plug multicenter registry, the subsequent annual major bleeding rate was 0.7 percent, corresponding to a relative risk reduction of 89 percent in patients with prior intracranial hemorrhage compared with those with other indications for LAAO [63,64]. LAAO is discussed in further detail separately. (See "Risks and prevention of bleeding with oral anticoagulants" and "Atrial fibrillation: Left atrial appendage occlusion".)

●Antiplatelet therapy – For patients with ICH and atrial fibrillation in whom anticoagulation is not resumed, antiplatelet therapy is used as an alternative. Because the benefit of aspirin to prevent thromboembolism in this population has not been established, we reserve antiplatelet therapy for patients with other indications. (See 'Antiplatelet therapy' above.)

Other patients — Anticoagulation may be used for patients with selected indications associated with a very high risk of thromboembolism and inadequate alternative choices. In these circumstances, risks and benefits should be discussed with patients and decisions should be individualized based on risk analysis and patient values. Such indications may include:

●Cancer-related thrombophilia with high risk of or prior venous thromboembolism (see "Risk and prevention of venous thromboembolism in adults with cancer")

●Hypercoagulable (acquired or inherited) conditions (see "Evaluating adult patients with established venous thromboembolism for acquired and inherited risk factors")

●Venous thromboembolism with high risk of recurrence (see "Selecting adult patients with lower extremity deep venous thrombosis and pulmonary embolism for indefinite anticoagulation")

●Other forms of cardiovascular disease (see "Prevention of cardiovascular disease events in those with established disease (secondary prevention) or at very high risk" and "Antithrombotic therapy in patients with heart failure")

●Other temporary high-risk indications for anticoagulation (see "Atrial fibrillation: Left atrial appendage occlusion", section on 'Postprocedure management' and "Overview of the treatment of proximal and distal lower extremity deep vein thrombosis (DVT)" and "Prevention of venous thromboembolism in adults undergoing hip fracture repair or hip or knee replacement" and "Left ventricular thrombus after acute myocardial infarction")

Alternatives to intravenous thrombolytic therapy — We do not routinely offer intravenous thrombolytic therapy to patients with ICH who develop conditions such as ischemic stroke, myocardial infarction, or pulmonary embolism, consistent with AHA guideline recommendations [65]. Endovascular and mechanical thrombectomy procedures may be an option for such patients. (See "Mechanical thrombectomy for acute ischemic stroke", section on 'Patient selection'.)

MANAGEMENT OF STATINS — The decision to resume statins (hydroxymethylglutaryl [HMG] CoA reductase inhibitors) for patients with ICH requires balancing the benefits of therapy with the potential hemorrhagic risks.

●We suggest resuming statins in most patients with ICH who have a strong indication for therapy. This includes patients with diabetes and coronary artery disease and those with recent myocardial infarction or baseline severe atherosclerotic arterial disease.

●We discontinue statins for patients with ICH with less compelling indications such as isolated dyslipidemia and for those with multiple (recurrent) lobar ICHs who are at high risk for ICH recurrence [66].

For patients with ICH who resume statins, we avoid high doses and prefer hydrophilic statins (eg, pravastatin or rosuvastatin). For patients who discontinue statins after ICH, we use the indication for therapy to help select an alternative medication. (See "Statins: Actions, side effects, and administration" and "Management of low density lipoprotein cholesterol (LDL-C) in the secondary prevention of cardiovascular disease" and "Hypertriglyceridemia in adults: Management".)

Statins appear to increase the propensity for ICH by inhibiting platelets, decreasing thrombus formation, and enhancing fibrinolysis [67,68]. In addition, hyperlipidemia appears to be protective against ICH. In a case-control study including 3492 patients with ICH, a reduced risk of ICH was associated with each increase in serum cholesterol by 5 mg/dL (odds ratio 0.87, 95% CI 0.86-0.88) [69]. Additionally, lower levels of low-density lipoprotein cholesterol are also associated with increased risk for ICH [70,71].

●Risk of initial ICH – The contribution of statin drugs to the risk of initial ICH is supported by trials and several observational studies [69,72,73]. In the Stroke Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) trial, patients with prior transient ischemic attack or ischemic stroke assigned to atorvastatin had fewer major cardiovascular events than those assigned to placebo (3.5 percent absolute risk reduction) [74]. However, the incidence rate of ICH was significantly higher in patients who received atorvastatin 80 mg daily (2.3 versus 1.4 percent; risk ratio [RR] 1.68, 95% CI 1.09-2.5).

●ICH recurrence risk – However, data on the risk of recurrent ICH are less certain [75,76]. Outcomes associated with statin use were evaluated in a 2018 systematic review of 15 studies that included more than 50,000 patients with prior ICH [25]. Among studies reporting ICH recurrence, the risk associated with statin use was similar to controls (RR 1.04, 95% CI 0.86-1.25). However, statin use was associated with improved functional outcome and reduced mortality in patients with prior ICH. These findings may reflect variability of individual risk related to several factors including statin dose or formulation, prescription bias, the burden of atherosclerotic disease, and the cause of the prior ICH.

•Effect of statin dose – Statin dose may modify the risk of hemorrhagic complications such as ICH [77]. A meta-analysis of seven randomized controlled trials found that higher doses of statins were associated with increased risk for ICH compared with placebo (RR 1.53, 95% CI 1.16-2.01) [78]. However, a large 10-year nationwide cohort study from Taiwan found no association between statin dose and risk of recurrent ICH [79].

•Effect of statin formulation – Statins with lipophilic solubility (atorvastatin, lovastatin, simvastatin, cerivastatin, and fluvastatin) have a greater ability to penetrate across blood-brain barrier and have been associated with increased odds of recurrent ICH compared with hydrophilic statins [79].

•ICH etiology – Some cohorts suggest the hemorrhagic risk with statin use is associated with patients with lobar ICH [80-82]. The protective effect of hyperlipidemia also appears to be higher for patients with lobar versus nonlobar ICH [69].

A case-control study that included 984 ICH patients who had a recurrent ICH and 3755 matched controls also found that statin exposure was associated similar risk of recurrent ICH for cases and controls (39 versus 41 percent) [83]. In addition, the recurrent ICH risk was similar for cases and controls when results were stratified by duration and intensity of statin therapy as well as by the presence of a history of ischemic stroke. Randomized trials evaluating the use of statins in patients with a history of ICH are ongoing.

LIFESTYLE MODIFICATIONS — We advise lifestyle modifications based on their association as risk factors of stroke, including ICH [33]. (See "Spontaneous intracerebral hemorrhage: Pathogenesis, clinical features, and diagnosis", section on 'Risk factors'.)

These include:

●Regular physical activity (see "The benefits and risks of aerobic exercise")

●Maintenance of healthy body weight (see "Obesity in adults: Overview of management")

●Healthy diet (see "Healthy diet in adults")

●Cessation of smoking and excessive alcohol use (see "Cardiovascular risk of smoking and benefits of smoking cessation")

●Avoidance of sympathomimetic medications (see "Spontaneous intracerebral hemorrhage: Pathogenesis, clinical features, and diagnosis", section on 'Risk factors')

LONG-TERM PROGNOSIS — Prognosis after ICH involves both initial prognosis and long-term prognosis. The clinical and imaging determinants of initial prognosis occur within the acute hospitalization and recovery periods, typically comprising the first 180 days after ICH. Long-term prognosis focuses on sequelae of ICH, which persist beyond 180 days.

Acute prognosis in ICH is discussed separately. (See "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis", section on 'Early prognosis'.)

Functional recovery — The rate of recovery after ICH may be highest in the first few weeks to months. In one study, recovery was greatest within the first 30 days after ICH [84]. However, recovery after ICH is often delayed and can be slow; many patients report functional improvements for 6 to 12 months [85,86]. In a post-hoc analysis of individual patient data from two clinical trials of patients with intracerebral or intraventricular hemorrhage, poor functional outcome at 30 days was reported in 715 of 999 patients (72 percent) [87]. By one year, 308 of these patients (46 percent) had achieved good functional outcome, including 30 percent who were functionally independent. Older age, larger hemorrhage volumes, and baseline conditions such as diabetes mellitus and severe leukoaraiosis on imaging were associated with poor one-year outcomes. In addition, common complications of acute ICH were also predictors of poor outcome at one year including sepsis, new ischemic stroke, prolonged mechanical ventilation, hydrocephalus, and the need for gastrostomy feeding tube. Aggressive acute treatment to prevent these acute complications may help avoid premature withdrawal of support and improve long-term outcomes. Early rehabilitation and sustained support are recommended to maximize functional recovery [84,88,89]. Educating patients and caregivers regarding secondary prevention strategies and addressing lifestyle changes, depression, and caregiver burden is an important part of post-ICH rehabilitation program. (See "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis", section on 'Initial aggressive care'.)

The main clinical determinants associated with reduced functional recovery after ICH include increasing age, baseline comorbidities, and severity of ICH as assessed by Glasgow Coma Scale score at presentation. Key imaging determinants include ICH volume, the presence of intraventricular hemorrhage, and specific ICH locations (including brainstem, posterior limb of internal capsule, or thalamus) [90,91].

Functional outcome may be assessed using varying performance thresholds or clinical scoring tools. The modified Rankin Scale (mRS) is frequently used (table 7). In several trials, patients with ICH achieving a score of 0 to 3 have been described as having a good functional outcome; poor outcome included those scoring 4 to 6. In a retrospective study of 1499 patients, 51 percent of patients with a first ICH had good functional recovery after 90 days, compared with 31 percent of patients after recurrent ICH [92].

A decline in long-term functional status has been observed among patients after ICH. In a single-center observational study of 560 patients, 23 percent of those with a good functional outcome at six months had declined over a median nine-year follow-up [93]. Advanced age, higher initial ICH volume, lower six-month functional status, and new diagnoses of dementia or stroke during follow-up were predictors of functional decline.

Cognitive impairment — Cognitive impairment is frequent among patients after ICH [94,95]. A systematic review reported that the prevalence of cognitive impairment ranged between 14 and 88 percent after ICH [96]. Predictive factors were previous stroke, ICH volume and location, and markers of cerebral amyloid angiopathy. Cognitive deficits after ICH were common across multiple domains. The most frequently impaired domains were naming, processing speed, executive functioning, memory, visuospatial abilities, and attention.

Long-term mortality — In population-based cohorts of patients hospitalized after ICH the 10-year survival rate ranged from 18 to 25 percent [5,97]. Life expectancy among patients after ICH was decreased compared with age- and sex-matched controls in the general population [98-100]. In a cohort of 219 patients with ICH, the major cause of death within five years was cardiovascular disease, largely due to recurrent ICH and its sequelae (10 percent) irrespective of ICH location [5]. The risk of death was similar in patients with lobar versus nonlobar ICH and higher in patients with anticoagulation-associated ICH.

A retrospective multicenter 10-year study of 1499 patients with ICH reported recurrent ICH in 9.5 percent of patients and found that 30-day mortality was similar (approximately 14 percent) after first and recurrent ICH [92].

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: Stroke in adults".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or email these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: Intracerebral hemorrhage (The Basics)")

●Beyond the Basics topic (see "Patient education: Hemorrhagic stroke treatment (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Risk factors for recurrent ICH – Several risk factors have been implicated in the risk of recurrent intracerebral hemorrhage (ICH). These include the location and etiology of the initial ICH, hypertension, older age, several medications (table 1), race/ethnicity, chronic kidney disease, prior stroke, and specific genetic features. (See 'Risk of recurrence' above.)

●Follow-up imaging to evaluate for underlying causes – We obtain follow-up imaging for patients with clinical or imaging features of the ICH suggestive of an underlying cause (algorithm 1 and table 2). For most patients, we prefer brain magnetic resonance imaging (MRI) with gadolinium to evaluate for an underlying cause. (See 'Follow-up neuroimaging' above.)

●Blood pressure management – Blood pressure control is an important feature of secondary prevention for all patients with ICH. We suggest a long-term target <130/80 mmHg to lower the risk of ICH recurrence (Grade 2B). (See 'Blood pressure management' above.)

●Management of antiplatelet therapy – We suggest resuming antiplatelet therapy for most patients with ICH who have a specific indication for such therapy (Grade 2C). For patients with ICH in whom antiplatelet therapy is being resumed, we typically start aspirin within a few days after the ICH has stabilized. (See 'Antiplatelet therapy' above.)

●Management of anticoagulation – We balance the risks of recurrent ICH and thromboembolism to help make decisions about resuming anticoagulation for each patient (algorithm 2). (See 'Anticoagulation' above.)

•Early resumption of anticoagulation may be indicated for select patients with a compelling indication (eg, mechanical prosthetic heart valve). For most other patients who resume anticoagulation, we generally suggest waiting for at least four weeks after onset of the ICH to restart the anticoagulant (Grade 2C).

•For patients with atrial fibrillation, prediction models such as the HAS-BLED score may be used to help assess bleeding risk (table 6) and CHADS2 or CHA2DS2-VASc scores used to help assess thromboembolic risks (table 5). For many patients with atrial fibrillation, the risk-benefit analysis favors resuming anticoagulation after ICH.

•For most patients with atrial fibrillation who develop ICH while on warfarin and in whom oral anticoagulation is resumed, we suggest a direct oral anticoagulant (DOAC) over warfarin (Grade 2B). DOACs generally have a lower risk of bleeding, including ICH, than warfarin. Warfarin may be selected for patients with valvular atrial fibrillation, a mechanical heart valve, or an inability to take a DOAC.

●Management of statins – We suggest resuming statins for most patients with lobar and nonlobar ICH who have atherosclerotic disease (Grade 2C). We discontinue statins for most patients with multiple (recurrent) lobar ICHs. (See 'Management of statins' above.)

●Long-term prognosis – The rate of recovery after ICH may be highest in the first few months. However, the greatest extent of recovery may not be evident until 6 to 12 months after ICH. (See 'Functional recovery' above.)