Spontaneous intracerebral hemorrhage: Pathogenesis, clinical features, and diagnosis

INTRODUCTION — Intracerebral hemorrhage (ICH) is the second most common cause of stroke after ischemic stroke and is a substantial cause of morbidity and mortality.

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", section on 'Primary brain injury'.)

The pathogenesis, epidemiology, clinical features, and diagnosis of spontaneous (atraumatic) ICH will be reviewed here. Other aspects of ICH are discussed separately.

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

●(See "Spontaneous intracerebral hemorrhage: Secondary prevention and long-term prognosis".)

●(See "Hemorrhagic stroke in children".)

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

PATHOGENESIS AND ETIOLOGIES

Mechanisms of brain injury — There are several mechanisms of brain injury in ICH. These include:

●Primary mechanical injury to brain parenchyma occurs via hematoma expansion and perilesional edema. Both hematoma volume and edema contribute to the mass effect and increased intracranial pressure (ICP), which in turn can cause reduced cerebral perfusion and ischemic injury, and, in very large ICH, cerebral herniation [1].

●Secondary brain injury from the breakdown of the blood-brain barrier after the initial hemorrhage includes excitotoxic and inflammatory processes; however, the exact mechanism(s) underlying this remain uncertain.

Enlargement of the hemorrhage is associated with neurologic deterioration, the development of increased intracranial pressure, and worse outcomes. In most cases, the bulk of hemorrhage expansion occurs in the first several hours after onset of ICH [2-5].

Perihematomal edema is frequent in ICH and may be related to mass effect, local neuronal ischemia, or the accumulation of cytotoxic factors [6]. Edema is present on computed tomography (CT) or magnetic resonance imaging (MRI) in at least half of patients when the patient is first imaged and progresses, reaching maximum volume 7 to 12 days after onset; the most rapid expansion occurs in the first 48 hours [7-9]. Hemorrhage volume, higher admission hematocrit, and a prolonged partial thromboplastin time appear to correlate with peak edema volume.

The perihematomal region exhibits delayed perfusion and increased diffusivity mixed with areas of reduced diffusion, suggesting the presence of both vasogenic and cytotoxic (ischemic) edema [10]. Acute blood pressure changes with impaired cerebral autoregulation in patients with ICH may contribute to perihematomal ischemia [11]. Underlying hypertensive vasculopathy or cerebral amyloid angiopathy (CAA) may also affect autoregulation in some patients. Increased intracranial pressure and the resulting reduction in cerebral perfusion pressure may play a role; this phenomenon may be exacerbated by blood pressure lowering.

Breakdown of the blood-brain barrier due to an inflammatory response to the ICH may be identified by contrast enhancement of brain tissue. In patients with acute and subacute ICH, postcontrast enhancement may be noted in the perihematomal area on CT and MRI [12,13]. Contrast enhancement remote from the hematoma may also be identified on imaging. One MRI study described a pattern of punctate areas of contrast enhancement in the sulcal areas both contiguous and remote from the hemorrhage suggesting that acute ICH may be associated with a more diffuse inflammatory process leading to widespread breakdown of the blood-brain barrier [14].

MRI hyperintensities on diffusion-weighted imaging (DWI) may also be seen with ICH. In some cases, restricted diffusion adjacent to the acute hematoma may be due to distortion of the magnetic field from paramagnetic blood products [15]. However, acute ischemic lesions discontiguous with the hemorrhage may also be identified on DWI [16-19]. In a 2019 meta-analysis of 11 studies and 1910 patients with ICH, the prevalence of diffusion lesions was approximately 20 percent and did not differ between ICH related to hypertensive vasculopathy or ICH related to CAA [20]. The mechanism causing this phenomenon and the implication of these findings for clinical brain injury and prognosis are as yet undefined [21]. One study with follow-up imaging identified new ischemic lesions beyond the acute ICH period [19].

Specific etiologies — There are several underlying pathological conditions associated with ICH; hypertension, amyloid angiopathy, and ruptured vascular malformation are most common [22,23].

Hypertensive vasculopathy — Hypertensive hemorrhages occur in the territory of penetrator arteries that branch off major intracerebral arteries, often at 90 degree angles to the parent vessel (image 1). These small penetrating arteries may be particularly susceptible to the effects of hypertension, as they are directly exposed to the pressure of the much larger parent vessel, without the protection of a preceding gradual decrease in vessel caliber [24].

The blood vessels that give rise to hypertensive hemorrhage are generally the same as those affected by hypertensive occlusive disease and diabetic vasculopathy, which cause lacunar strokes. These vessels supply the pons and midbrain (penetrators branching from the basilar artery), thalamus (thalamostriate and thalamogeniculate penetrators branching from the P1 and P2 segments of the posterior cerebral arteries), putamen, caudate, and globus pallidus (lenticulostriate penetrators branching from the M1 segment of the middle cerebral artery), and cerebellar nuclei (dentate nucleus penetrators branching from the cerebellar arteries). Cerebellar hemorrhage is more common than lacunar infarction in the cerebellum. Hypertensive vasculopathy is also believed to play a role in the development of white matter small vessel disease (leukoaraiosis), which may account for the association between white matter disease and risk of ICH [25].

Pathologic examination of the blood vessels in patients with chronic hypertension and in those with ICH has led to a theory of how hypertensive hemorrhage occurs. The penetrator vessels in patients with chronic hypertension develop intimal hyperplasia with hyalinosis in the vessel wall; this predisposes to focal necrosis, causing injury to the wall of the vessel. These microscopic "pseudoaneurysms" have been associated with small amounts of blood identified in the extravascular space. Pseudoaneurysm formation with subclinical leaks of blood may be relatively common; massive hemorrhage can occur when the clotting system is unable to compensate for these disruptions in the vessel wall.

Cerebral microbleeds (CMBs) are markers of bleeding-prone microangiopathy that may be found on imaging of patients with ICH due to hypertensive vasculopathy. T2*-weighted gradient echo and susceptibility-weighted MRI sequences can detect CMBs as punctate foci of hemosiderin deposition that represent remnants of clinically silent cerebral microhemorrhage [26,27]. The anatomic distribution of microbleeds varies with their etiology, with hypertensive microbleeds arising in the basal ganglia, thalamus, pons, and cerebellar nuclei in contrast with CAA-related microbleeds, which are found in more superficial lobar regions of the cerebral hemispheres (image 2). This regional distribution is consistent with the usual location of ICH in these conditions.

Cerebral amyloid angiopathy — CAA is an important cause of primary lobar ICH in older adults. CAA is characterized by the deposition of congophilic material in small- to medium-sized blood vessels of the brain and leptomeninges. This weakens the structure of the vessel walls and makes them prone to bleeding. CAA usually manifests with spontaneous lobar hemorrhage (image 3). The presence of CMBs restricted to the lobar region is associated with CAA. CAA is described in detail separately. (See "Cerebral amyloid angiopathy".)

Other etiologies — Several other less common underlying etiologies of nontraumatic ICH include [23,28-32]:

●Arteriovenous and other vascular malformations – Rupture of vascular malformations including arteriovenous malformations (AVMs) and cavernous malformations (CMs) may be the cause of ICH. AVMs are characterized by connection of high-flow arterial pressure to the venous circulation without intervening capillary network. The hemorrhage is thought to result from rupture of a weakened vascular segment such as an intranidal aneurysm and most frequently occurs in lobar, intraventricular, or subarachnoid regions (image 4 and image 5). Perilesional brain ischemia from vascular steal may also occur in patients with AVMs. By contrast, the hemorrhage due to spontaneous rupture of a CM is less common than with AVMs because these lesions are slow-flow vascular malformations comprised of thin-walled capillaries. ICH due to CMs commonly occurs in the brainstem, juxtacortical regions, or intraventricular space. (See "Brain arteriovenous malformations", section on 'Pathogenesis and pathology' and "Vascular malformations of the central nervous system", section on 'Cavernous malformations'.)

●Cerebral venous thrombosis – Obstruction of the cerebral veins or venous sinuses results in increased venous pressure and leads to venous or capillary rupture with hemorrhage, often with simultaneous venous infarction (image 6). Cerebral venous thrombosis most often occurs in individuals with a prothrombotic state. (See "Cerebral venous thrombosis: Etiology, clinical features, and diagnosis", section on 'Pathogenesis'.)

●Hemorrhagic infarction – Breakdown of the structural integrity of the neurovascular unit accompanies tissue death during ischemic infarction. Blood extravasation during acute infarction may be visible on imaging as acute hemorrhagic infarction (or hemorrhagic transformation) (image 7). Hemorrhagic infarction is common in ischemic infarcts that are large or associated with cerebral edema and those from an embolic source. Reperfusion injury after thrombolytic therapy or thrombectomy may also be a source of hemorrhagic infarction. (See "Pathophysiology of ischemic stroke".)

●Reversible cerebral vasoconstriction syndrome (RCVS) – The transient multifocal cerebral arterial narrowing frequently seen with RCVS may also be associated with cerebral insults, including ischemic infarction or ICH (image 8). ICH may be accompanied by subarachnoid hemorrhage in patients with RCVS. The typical clinical presentation includes recurrent thunderclap headaches. (See "Reversible cerebral vasoconstriction syndrome".)

●Primary or metastatic tumor – Primary brain tumors such as high-grade gliomas and several metastatic brain tumors such as melanoma, choriocarcinoma, and lung, renal cell, and thyroid carcinomas may cause intratumoral hemorrhage, often associated with a large amount of surrounding edema (image 9). (See "Overview of the clinical features and diagnosis of brain tumors in adults" and "Epidemiology, clinical manifestations, and diagnosis of brain metastases", section on 'Pathogenesis'.)

●Central nervous system infection – Bacterial infections may cause brain abscess leading to ICH via mass effect or vessel wall erosion. Hemorrhagic necrosis of brain tissue may also occur with various viral infection (eg, herpes simplex encephalitis). (See "Pathogenesis, clinical manifestations, and diagnosis of brain abscess" and "Viral encephalitis in adults".)

●Mycotic intracranial aneurysm – Cerebral emboli occurring as a complication of infective endocarditis may infect the arterial wall causing weakening and dilation, termed a mycotic aneurysm. They typically occur in distal arterioles and may be more likely to rupture when associated with a low platelet count or acute ischemic embolic stroke. ICH from this source may be intraparenchymal as well as subarachnoid (image 10). (See "Complications and outcome of infective endocarditis", section on 'Neurologic complications'.)

●Moyamoya disease – Progressive arterial narrowing and prominent collateral vessel formation around the circle of Willis in patients with moyamoya disease (image 11 and image 12) can lead to ICH with or without subarachnoid hemorrhage. Hemorrhage frequently involves the basal ganglia, thalamus, and/or ventricular system (image 13). ICH from moyamoya disease is more common in adults than children and more frequent in the carotid than vertebrobasilar arterial territories. (See "Moyamoya disease and moyamoya syndrome: Etiology, clinical features, and diagnosis", section on 'Intracerebral, intraventricular, and subarachnoid hemorrhage'.)

●Cerebral vasculitis – Multiple abnormalities on brain MRI may be found in patients with primary central nervous system or systemic vasculitides including T2 hyperintensities, ischemic infarcts, and hemorrhage. Vascular imaging frequently features multifocal segmental narrowing of inflamed blood vessels. (See "Primary angiitis of the central nervous system in adults" and "Overview of and approach to the vasculitides in adults".)

●Cerebral hyperperfusion syndrome – After carotid revascularization procedures, the previously hypoperfused tissue may be unable to accommodate initially to the significant increase in cerebral perfusion pressure. Ipsilateral cerebral edema and hemorrhage typically occur within the first week after revascularization; elevated post-procedural blood pressure is an associated risk factor. (See "Complications of carotid endarterectomy", section on 'Hyperperfusion syndrome'.)

●Sickle cell disease – Arterial wall thickening and progressive fragility as well as impaired cerebrovascular autoregulation can lead to ICH in patients with sickle cell disease (SCD). The vasculopathy with SCD is also associated with the development of moyamoya syndrome or cerebral aneurysms. ICH in patients with SCD is more common in adults than children and more frequent in those with coexisting hypertension or coagulopathy. (See "Prevention of stroke (initial or recurrent) in sickle cell disease".)

●Bleeding disorders – Acquired or genetic bleeding disorders and conditions that impair hemostasis such as liver disease, antithrombotic therapy, or thrombolytic therapy are risk factors for ICH or systemic hemorrhage but may also be identified as the underlying cause when coagulopathy is severe and when evaluation excludes other sources. (See "Thrombotic and hemorrhagic disorders due to abnormal fibrinolysis" and "Hemostatic abnormalities in patients with liver disease" and "Intravenous thrombolytic therapy for acute ischemic stroke: Therapeutic use", section on 'Intracerebral hemorrhage'.)

EPIDEMIOLOGY — Spontaneous (atraumatic) ICH is the cause of 9 to 27 percent of all strokes globally [33,34]. The overall incidence of ICH ranges from 12 to 31 per 100,000 people and varies by race [35-40]. The incidence of ICH increases with age, doubling every 10 years after age 35 [41,42].

A 2013 systematic review found that the global burden of brain hemorrhage (mainly ICH but also subarachnoid hemorrhage) was greater than that of ischemic stroke in terms of death and disability, even though the incidence of ischemic stroke was twice as great [43]. The highest incidence of brain hemorrhage was found in Asia and southern sub-Saharan Africa, while the lowest incidence was found in North America, western Europe, Latin America, and Oceana.

In the United States, ICH incidence varies by race. The rate of occurrence is highest in Chinese and Japanese patients, intermediate in Black Americans, and lowest in White Americans. The higher rate of ICH in Black compared with White Americans is predominately attributable to excess ICH in deep cerebral and brainstem locations where hypertension is the major risk factor [37]. Mexican Americans also have a higher incidence of ICH than non-Hispanic White Americans [44,45].

Data about the association of sex with the incidence of ICH are inconclusive [46]. A 2010 systematic review and meta-analysis found no significant difference in incidence between male and female patients [39], while a 2009 systematic review found that the incidence of ICH was higher in males [47].

RISK FACTORS — Major risk factors for spontaneous ICH include older age, hypertension, and the use of antithrombotic (antiplatelet and anticoagulant) therapy.

Age — The risk of ICH increases with advancing age. A meta-analysis of more than 8100 patients with ICH assessed the age-related incidence of ICH over a 28-year period [39]. 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.

Hypertension — In addition to being the most common etiology of spontaneous ICH (see 'Hypertensive vasculopathy' above), hypertension is the most important risk factor for the development of ICH [40,48-51]. Hypertension more than doubles the risk of ICH [52-55].

The relative contribution of hypertension may be greater for deep than for lobar ICH [51,52,55,56]. In a meta-analysis that pooled data from 28 studies, hypertension was twice as common in patients with deep ICH as in patients with lobar ICH (odds ratio [OR] 2.1, 95% CI 1.82-2.42) [56]. In data from a subset of three studies in the meta-analysis meeting more rigorous methodologic criteria, the association of hypertension with deep ICH remained significant (OR 1.5, 95% CI 1.09-2.07).

In at least some studies, hypertension has also been shown to be a risk factor for ICH in the setting of other underlying etiologies for ICH (eg, cerebral amyloid angiopathy, antithrombotic-associated ICH) [57-59]. (See "Cerebral amyloid angiopathy".)

Antithrombotic medications — Anticoagulant therapy, particularly warfarin, is associated with an increased risk of ICH, whereas the risk of ICH with antiplatelet monotherapy appears to be minimal [60].

●Warfarin – Anticoagulation with warfarin increases the risk of ICH two- to five-fold, depending upon the intensity of anticoagulation. This is discussed separately. (See "Risks and prevention of bleeding with oral anticoagulants", section on 'Risk factors related to the anticoagulant'.)

In addition to an increased risk of ICH, retrospective evidence suggests that warfarin therapy with an international normalized ratio (INR) >3 is a risk factor for larger initial hemorrhage volume as well as poorer outcomes after ICH. (See "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis", section on 'Risk factors for poor outcomes'.)

●Direct oral anticoagulants – The risk of ICH is approximately 30 to 60 percent lower with the direct oral anticoagulants (DOACs; dabigatran, apixaban, edoxaban, rivaroxaban) than with warfarin in patients with nonvalvular atrial fibrillation, even when compared with well-controlled warfarin [61]. (See "Risks and prevention of bleeding with oral anticoagulants", section on 'Risk factors related to the anticoagulant'.)

●Parenteral anticoagulants – Heparin products, including unfractionated and low-molecular weight heparins, are associated with an elevated risk of ICH. In general, the risk of bleeding rises with the intensity of anticoagulation, duration of therapy, and associated patient-level risk factors. (See "Heparin and LMW heparin: Dosing and adverse effects", section on 'Bleeding'.)

●Thrombolytic agents – The risk of ICH associated with thrombolytic medications such as alteplase or tenecteplase varies by dosage and indications. As examples, the rate of ICH was 1.3 percent in a trial of patients receiving thrombolysis for myocardial infarction; whereas, for patients with acute brain ischemia receiving thrombolytic therapy for acute ischemic stroke, the rate of symptomatic ICH was 6 percent [62,63]. (See "Acute ST-elevation myocardial infarction: The use of fibrinolytic therapy", section on 'Stroke' and "Approach to reperfusion therapy for acute ischemic stroke", section on 'Risk of intracerebral hemorrhage'.)

●Antiplatelet agents – There is probably a small absolute increased risk of primary ICH associated with aspirin or antiplatelet agent monotherapy, based on meta-analyses of randomized trials [60,64,65], although other case-control studies have not found an increased risk [66,67]. A subsequent review estimated the absolute risk of ICH attributed to the use of aspirin for primary and secondary prevention of coronary heart disease to be 0.2 events per 1000 patient years [68].

Dual antiplatelet therapy likely confers a higher risk compared with monotherapy. In a trial of over 7500 patients with atrial fibrillation for whom warfarin therapy was considered unsuitable, antiplatelet therapy with aspirin plus clopidogrel increased the risk of ICH twofold compared with aspirin alone (0.4 versus 0.2 percent) [69]. However, in another randomized trial of over 4800 adults with minor stroke or transient ischemic attack (TIA), short-term (90-day) treatment with clopidogrel plus aspirin did not increase the risk of ICH compared with aspirin alone [70].

Nonsteroidal anti-inflammatory drugs (NSAIDs) do not appear to increase the risk of ICH [71-74].

Prasugrel and ticagrelor have been associated with higher rates of major bleeding when used during percutaneous coronary intervention procedures in patients with acute coronary syndromes. (See "Periprocedural bleeding in patients undergoing percutaneous coronary intervention", section on 'Antithrombotic therapy'.)

Other risk factors

●Obesity and inactivity – Inactivity and obesity are comorbidities that can lead to increased risk for ICH. In one study including 777 ICH cases and 2083 control subjects, obesity had a small effect on the risk of ICH (OR 1.28, 95% CI 1.03-1.57), mostly through an indirect effect on hypertension [75]. Additionally, obesity is associated with the risk of obstructive sleep apnea, which is related to an increased prevalence of atrial fibrillation, higher use of anticoagulants, and a surge in nocturnal blood pressure [76].

●High alcohol intake – Heavy alcohol use is associated with an approximately threefold increased risk of ICH [53,55,77]. In a meta-analysis including 11 prospective studies in patients with ICH, heavy alcohol consumption (>4 drinks/day) was associated with risk for ICH (relative risk 1.67, 95% CI 1.25-2.23) [78]. The Ethnic/Racial Variations of Intracerebral Hemorrhage (ERICH) study found a similar association (OR 1.77, 95% CI 1.30-2.41) at the threshold of ≥5 drinks per day [79].

Heavy alcohol consumption can also contribute to the risk of ICH indirectly due to its contribution to hypertension [80].

●Race and ethnicity – Race appears to be associated with an increased risk of ICH that is age related. Findings from a surveillance study and a prospective cohort have found that the risk factors of race and age appear to interact, such that young (45- to 60-year-old) Black Americans have a higher risk of ICH than White Americans, but this increased risk declines with increasing age [54,81].

●Lower cholesterol and lower low-density lipoprotein cholesterol – A systematic review and meta-analysis of 23 prospective studies found low cholesterol was associated with an increased ICH risk [82].

However, the available data suggest that treatment with statins does not clearly increase the risk of ICH [83-88]. One meta-analysis of randomized trials did not find a higher risk of ICH in patients on statins (OR 1.08, 95% CI 0.88-1.32) [85]. However, another meta-analysis of randomized trials involving patients treated with high-dose statins did find an increased risk of ICH (risk ratio 1.53, 95% CI 1.16-2.01) [89].

●Genetic variation – Specific genetic features may contribute to as much as 44 percent of ICH risks [90]. However, most cases of ICH are not believed to have a monogenetic component. One exception is cerebral amyloid angiopathy-related ICH, which has been shown to have an association with apolipoprotein E (APOE) genotype. A large-scale genetic association study of 2189 ICH cases and 4041 controls revealed that the APOE allele epsilon 4 was also associated with deep ICH (OR 1.21), a location not typical for cerebral amyloid angiopathy [91]; however, in another large study, APOE epsilon 2 or epsilon 4 allele was specifically associated with lobar and not deep ICH [52]. (See "Cerebral amyloid angiopathy", section on 'Genetic susceptibilities'.)

●Small-vessel vascular disease – Evidence of small-vessel disease may be seen on brain magnetic resonance imaging studies as lacunar infarcts, white matter hyperintensities on T2 sequences, or microbleeds. These markers of blood vessel fragility or atherosclerosis within the small vessels of the brain are associated with an elevated risk of ICH. In an observational study involving nearly 1500 patients with prior ischemic stroke and atrial fibrillation taking anticoagulation, the rate of ICH was higher in those with evidence of small-vessel disease than those without (0.6 versus 0.1 percent/year) [92,93].

●Tobacco use – Tobacco use may be associated with the risk of ICH, presumably by contributing to elevated blood pressure secondary to an increase in cardiac output and total peripheral vascular resistance and arterial wall damage, which predispose to rupture of small vessels in the brain. In the Physicians Health Study, active smokers had a relative ICH risk of 2.06 percent (95% CI 1.08-3.96) compared with nonsmokers [94].

●Stimulant drug use – The use of stimulant medications has been associated with a risk of ICH due to possible spikes in blood pressure and vasospasm. Appetite suppressant or cough medications that contain sympathomimetic properties such as phenylpropanolamine are associated with an elevated risk of ICH [95]. Likewise, cocaine use has been associated with the risk of incidence ICH. Caffeine-containing medications have also been associated with ICH [96].

●Infections – Several infectious pathogens have been associated with the risk of incident ICH. These include HIV [97], hepatitis C virus [98,99], and the varicella-zoster virus [100,101] as well as the spirochete from the Leptospira genus that causes leptospirosis [102].

●Less-certain risk factors – Other risk factors for ICH are supported by observational data, including chronic kidney disease [103-105], diabetes [106], use of selective serotonin reuptake inhibitors [107,108], migraine [109], and systemic amyloidosis [110].

CLINICAL PRESENTATION — The signs and symptoms of ICH vary according to the location and size of the hemorrhage (table 1).

Onset and progression — In most circumstances, ICH onset occurs during routine activity. However, some hypertensive hemorrhages occur with exertion or intense emotional activity [111].

The neurologic symptoms and signs may be progressive over minutes or a few hours (figure 1), in contrast with brain embolism and subarachnoid hemorrhage, where the neurologic symptoms and signs are often maximal at onset. However, some patients with large ICH are obtunded or comatose when first discovered or at first evaluation upon arrival to the emergency department.

Headache, vomiting, and a decreased level of consciousness may develop if the hemorrhage is large. This symptom complex is typically absent with small hemorrhages. However, headache and vomiting occur in approximately one-half of patients with ICH (figure 2). Headache may be due to traction on meningeal pain fibers, increased intracranial pressure (ICP), or blood in the cerebrospinal fluid; it is most common with cerebellar and lobar hemorrhages. Patients may complain of a stiff neck and have meningismus on physical examination if there is intraventricular blood.

Stupor or coma attributed to the ICH is often an ominous sign. Exceptions include patients with thalamic hemorrhage with involvement of the reticular activating system who may recover after acute blood is resorbed and those with acute hydrocephalus who might improve if treated with an external ventricular drain. Patients may also present with stupor or coma due to reversible causes such as acute metabolic derangements or seizure.

Neurologic signs and ICH location — Neurologic signs vary depending upon the location of the hemorrhage (image 1). In one study, bleeding involved the putamen in approximately 35 percent of cases, cerebral lobes in 30 percent, cerebellum in 16 percent, thalamus in 15 percent, and pons in 5 to 12 percent [112]. In the subsequent Intensive Blood Pressure Reduction in Acute Cerebral Hemorrhage Trial (INTERACT2) of over 2000 subjects with imaging-confirmed ICH and hypertension, the frequency of affected brain structures was as follows [113]:

●Putamen/globus pallidus – 56 percent

●Posterior limb of internal capsule – 46 percent

●Anterior limb of internal capsule – 5 percent

●Thalamus – 31 percent

●External capsule – 27 percent

●Lobar – 14 percent

●Cerebellum or brainstem – 7 percent

●Caudate head – 2 percent

Intraventricular extension of the ICH was identified in 29 percent in this study.

Neurologic exam deficits typically correspond to the location of the hemorrhage and associated edema. Patients with deficits not attributable to the hemorrhage should be evaluated for other causes, such as expansion of the hemorrhage, post-ictal symptoms after a seizure, or increased intracranial pressure.

The localization of ICH may be associated with typical neurologic exam findings:

●Putaminal hemorrhage – Spread of hemorrhage into the putamen most commonly occurs along white matter fiber tracts, causing hemiplegia, hemisensory loss, homonymous hemianopsia, and gaze palsy. Stupor and coma may develop if the hemorrhage is large.

●Caudate hemorrhage – Hemorrhage typically originating within the head of the caudate nucleus may cause acute-onset confusion, personality changes, or memory impairment as well as transient contralateral weakness or numbness [114]. Headache and drowsiness may also occur, especially if bleeding extends into the adjacent intraventricular space.

●Internal capsule hemorrhage – Small hemorrhages restricted to the internal capsule may cause mild dysarthria, contralateral hemiparesis, and sensory deficit [115].

●Cerebellar hemorrhage – Cerebellar hemorrhage usually originates in the dentate nucleus and may extend into the hemisphere and fourth ventricle and possibly into the pontine tegmentum. These bleeds typically cause an inability to walk due to imbalance, vomiting, and occipital headache. Some patients have referred pain to the neck or shoulder, neck stiffness, gaze palsy, and/or facial weakness. Notably, there is often no hemiparesis. The patient may become stuporous due to obstructive hydrocephalus or brainstem compression.

Patients with acute cerebellar hemorrhage may frequently deteriorate and require surgery. (See "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis", section on 'Surgical approaches for selected patients' and "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis", section on 'Cerebellar hemorrhage'.)

●Thalamic hemorrhage – Thalamic hemorrhages may extend in a transverse direction to the posterior limb of the internal capsule, downward to put pressure on the tectum of the midbrain, or medially to rupture into the third ventricle. Symptoms include hemiparesis, hemisensory loss, and occasionally transient homonymous hemianopsia. Pupils may be miotic and unreactive along with a gaze palsy (eg, peering at the tip of the nose, skewed, or "wrong-way eyes" toward the weak side [in contrast with hemispheric cortical injury in which the eyes are deviated away from the hemiparesis]). Aphasia may occur if the bleed affects the dominant hemisphere, while neglect may develop if the bleed affects the nondominant hemisphere. Patients with small anterior thalamic hemorrhages may present with drowsiness, acute confusion, or neuropsychiatric symptoms.

●Lobar hemorrhage – Lobar hemorrhages vary in their neurologic signs depending upon location. They most often affect the parietal and occipital lobes. These bleeds are associated with a higher incidence of seizures. Occipital hemorrhages frequently present with a very dense contralateral homonymous hemianopsia. Hemorrhages in the frontoparietal region will produce a contralateral plegia or paresis of the leg with relative sparing of the arm.

●Pontine hemorrhage – Pontine hemorrhages typically originate in brainstem nuclei and may extend into the base of the pons. These often lead to deep coma over the first few minutes following the hemorrhage, probably due to disruption of the reticular activating system. The motor examination may be marked by bilateral paralysis. The pupils are pinpoint and react to a strong light source. Horizontal eye movements are often absent, and there may be ocular bobbing, facial palsy, deafness, and dysarthria if the patient is awake.

Seizures — Seizures may accompany acute ICH. Seizures in the first days after ICH occur approximately 15 percent of patients [116,117]; they are more common in lobar hemorrhages (affecting cortical tissue) than in deep or cerebellar ICH [118-121]. (See "Overview of the management of epilepsy in adults", section on 'Poststroke seizures'.)

Cardiac abnormalities — Cardiac abnormalities are commonly associated with spontaneous ICH [122]. The most frequently associated electrocardiographic (ECG) changes are prolonged QT interval and ST-T wave changes. These changes may reflect catecholamine-induced cardiac injury, which is most likely due to a centrally mediated release of excess catecholamines caused by increased intracranial pressure or autonomic disturbance [123]. (See "Clinical manifestations and diagnosis of stress (takotsubo) cardiomyopathy" and "Complications of stroke: An overview", section on 'Neurogenic cardiac damage'.)

Mild elevations in serum myocardial enzymes often accompany the ECG changes, including cardiac troponin and beta natriuretic peptide [122]. Echocardiographic abnormalities may involve global or regional wall motion abnormalities and reduced ejection fraction. Ventricular arrhythmias may occur with brainstem compression [124].

BRAIN IMAGING — Both computed tomography (CT) or magnetic resonance imaging (MRI) are considered first-choice imaging options for the emergency diagnosis and assessment of ICH (image 14) [125].

Clinical deficits on neurologic examination can be correlated with impacted brain regions apparent on CT or MRI. Acute imaging also provides information about extension into the ventricular system, the presence of surrounding edema, shifts in brain contents (herniation), impending expansion, and underlying etiology. ICH severity can be assessed by calculating the volume of the hemorrhage. (See 'Estimating hemorrhage volume' below.)

Head CT — Noncontrast head computed tomography (CT) accurately identifies the presence of acute ICH, distinguishing it from ischemic stroke. Hyperacute blood will appear hyperdense except in rare cases of severe anemia when it might appear isodense.

Over weeks, the blood from an acute hemorrhage will typically become isodense and may have a ring-enhancement appearance. Chronically, the blood is hypodense (image 15).

CT angiography that may be performed along with a noncontrast head CT may identify an underlying vascular cause to the ICH [29,126].

Noncontrast head CT and CT angiography may also provide additional information regarding propensity for hemorrhagic expansion [127]. (See 'Predicting hemorrhage expansion' below.)

Brain MRI — Brain magnetic resonance imaging (MRI) is similarly sensitive as CT for detecting acute ICH. Acute ICH can be diagnosed by MRI with up to 100 percent sensitivity and accuracy by experienced readers [128]. In the Hemorrhage and Early MRI Evaluation (HEME) study, 200 patients presenting with focal stroke symptoms were evaluated with MRI including gradient recall echo (GRE) and diffusion-weighted imaging (DWI) sequences within six hours of symptom onset followed by CT [129]. MRI and CT were equivalent for the detection of acute ICH, and MRI was significantly more accurate than CT for the detection of chronic ICH.

However, MRI is contraindicated for patients with some metallic implants [130]. Additionally, MRIs are not as readily available on an emergent basis as CT and conventional protocols take longer than CT protocols.

●Age of hemorrhagic findings – MRI may also provide information about the age of the ICH (table 2). The appearance of intracerebral blood on MRI images varies with time depending on the paramagnetic properties of the various stages of hemoglobin and on the mode of imaging acquisition [131,132]. Of note, these imaging features do not apply to extra-axial or intraventricular hemorrhage.

•Hyperacute hemorrhage (0 to 3 hours) − A hyperacute parenchymal hematoma contains oxyhemoglobin and appears hypo- to isointense on T1-weighted images, hyperintense on T2-weighted images, and hypointense on T2*-weighted (GRE or susceptibility-weighted imaging) sequences (image 14) [133].

•Acute hemorrhage (3 hours to 3 days) – Acute ICH can be accurately detected due to the magnetic susceptibility of deoxygenated hemoglobin (deoxyhemoglobin) [128,129,132-136]. This property of superparamagnetic deoxyhemoglobin results in rapid dephasing of proton spins causing signal loss (darkening or hypointensity) that is best seen in T2*-weighted images.

•Subacute hemorrhage (3 days to 3 weeks) – Subacute ICH is recognized by the presence of diamagnetic methemoglobin, which appears as a T1-hyperintense signal, often first along the periphery of the hemorrhage. The appearance on T2-weighted imaging is hypointense in the first week due to magnetic susceptibility but becomes T2 hyperintense in the late subacute stage (one to three weeks) as red blood cells lyse and methemoglobin becomes extracellular.

•Chronic hemorrhage (>3 weeks) – The presence of superparamagnetic hemosiderin and ferritin in chronic ICH causes marked hypointensity on T2- and T2*-weighted images. On T1-weighted images, chronic hemorrhage is isointense compared with brain tissue.

●Nonhemorrhagic findings in acute ICH – Brain MRI may show other findings associated with acute ICH.

•Cerebral edema – Tissue damage and subsequent inflammatory response to the presence of intracranial blood can lead to perilesional cerebral edema in ICH [137]. Edema can compress adjacent intracranial structures, causing elevated intracranial pressure (ICP) and leading to hydrocephalus. Cerebral edema may be seen in acute and subacute ICH and typically appears hypointense on T1-weighted images and hyperintense on T2-weighted images.

ICH-related cerebral edema often peaks two to three weeks after hemorrhage [7,138]. Acute imaging showing prominent edema along with ICH may indicate subacute age of bleeding or suggest ICH may be due to a specific underlying cause such as cerebral sinus thrombosis (image 6) or tumor (image 9). (See "Spontaneous intracerebral hemorrhage: Secondary prevention and long-term prognosis", section on 'Follow-up neuroimaging'.)

•Diffusion-weighted imaging lesions – Brain MRI may show hyperintense diffusion-weighted imaging (DWI) lesions associated with acute ICH. These may indicate that the underlying event was a primary ischemic infarct with subsequent hemorrhagic transformation (image 7) or an event that can produce simultaneous ischemic and hemorrhagic injury, such as an embolism from infective endocarditis (table 3). (See "Spontaneous intracerebral hemorrhage: Secondary prevention and long-term prognosis", section on 'Follow-up neuroimaging'.)

Small DWI hyperintense lesions may also be found in patients with primary acute ICH. These lesions are often punctate in size, multifocal in number, and remote from and contralateral to the site of hemorrhage in location (image 16). They appear to occur most frequently during the first several days after ICH but can also be found on subsequent nonacute imaging [19]. In a systematic review of observational studies that included 5211 patients with acute ICH, the pooled prevalence of DWI lesions on acute brain MRI was 24 percent [139]. DWI lesions are associated with larger ICH volume, subarachnoid bleeding, and elevated blood pressure, suggesting acute changes in cerebral perfusion and/or ICP may contribute to their development [139,140].

MRI may also demonstrate imaging findings that may suggest a specific underlying cause. (See 'Subsequent imaging' below.)

Estimating hemorrhage volume — An estimate of ICH volume is useful to communicate hemorrhage severity and make early assessments of prognosis. (See "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis", section on 'Clinical prediction scores'.)

The formula is calculated using the centimeter scale on the CT (or MRI) images as follows [141]:

●A is the greatest hemorrhage diameter on the CT slice with the largest area of hemorrhage.

●B is the largest diameter 90 degrees to A on the same (index) CT slice (image 17).

●C is the approximate number of CT slices with hemorrhage multiplied by the slice thickness in centimeters. To calculate C, each CT slice with hemorrhage is visually compared with the index CT slice [141]. An individual hemorrhage slice is counted as one full slice for determining C if the hemorrhage area is >75 percent of the area on the slice with the largest hemorrhage. A slice is counted as one-half if the hemorrhage area is approximately 25 to 75 percent of the area on the largest hemorrhage slice. The slice is not counted if the area is <25 percent of the largest hemorrhage slice.

●ABC/2 gives the ICH volume in cubic centimeters.

In children, ICH volume may instead be measured as a percent of total brain volume as ABC/XYZ, where X is the largest midline axial diameter of supratentorial brain, Y is the largest axial diameter perpendicular to X, and Z is the brain vertical diameter [142].

Predicting hemorrhage expansion — Enlargement of the hemorrhage is associated with neurologic deterioration and worse outcomes. These observations indicate that significant improvements in patient outcome from ICH may be achieved by minimizing both secondary brain ischemia and hemorrhage enlargement in the early hours following the onset of bleeding. (See "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis", section on 'Managing hemorrhagic expansion'.)

Serial CT scans in patients with hypertensive hemorrhage have shown that the hemorrhage enlarges in the first few hours after presentation in a subset of patients [2,6,143]. In most but not all cases, the bulk of hemorrhage expansion occurs in the first three hours after onset of ICH [2-5,144]. In a prospective series of 103 patients with ICH, significant hemorrhage growth (a >33 percent volume increase) occurred in 38 percent of patients over the first 24 hours [2].

In a patient-level meta-analysis of studies reporting ICH growth, with data on over 5400 subjects, several independent predictors of hemorrhage growth identified [145]:

●Shorter time from symptom onset to initial imaging

●ICH volume

●Antiplatelet or anticoagulant use

●Contrast extravasation (eg, "spot sign") on initial CT angiography (image 18)

While all patients with ICH should be treated to limit risk for hemorrhage expansion, the presence of specific radiologic findings may trigger a somewhat more aggressive approach to monitoring and treatment. (See 'Subsequent imaging' below and "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis".)

●Spot sign on CT angiography – The spot sign (image 18) describes the appearance of small focal or multifocal areas of contrast enhancement within a hemorrhage on CT angiography source images [146]. The spot sign has been linked to hemorrhage expansion and poor outcomes, including mortality [127,147]. In a retrospective analysis of 367 patients with acute ICH, a spot sign was found in 19 percent of patients and was independently associated with hemorrhage expansion [148].

Accumulation of contrast extravasation within the hemorrhage on postcontrast CT also predicts subsequent hemorrhage expansion [149].

Although less well-studied, tiny spots of contrast extravasation within the hemorrhage on MRI (the MRI spot sign) have also been detected on postcontrast T1-weighted and dynamic T1-weighted images, particularly when performed in the first six hours after ICH onset [150]. This "MRI spot sign" may be associated with hematoma growth and worse clinical outcomes, but definitive data are lacking [151].

A spot sign score that grades the number of spot signs, their maximum dimension, and attenuation has been found to be a strong predictor of hemorrhage expansion; of these features, the number of spot signs appears to be most predictive [148,152,153].

●Signs on noncontrast head CT – Irregularity or heterogeneity of the hematoma on noncontrast head CT also suggests ongoing bleeding or risk of hematoma expansion [154,155]. In a systematic review including 25 studies and 10,650 patients with ICH, several imaging features visible on CT were associated with subsequent risk of poor functional outcome [156]. These include (image 19):

•Irregular hematoma shape

•Island sign (≥3 scattered hematomas separate from the main hematoma)

•Hypodensity (any hypodense focus within the hematoma)

•Heterogeneous density (≥3 foci of hypodensity within the hematoma)

•Black-hole sign (focus of hypodensity with a difference of >28 Hounsfield units [HU] from the hematoma)

•Swirl sign (hypodense region within the hematoma)

•Blend sign (focus of hypodensity with a difference of >18 HU adjacent to the hematoma)

●Subarachnoid extension – In one study, extension of hemorrhage into the subarachnoid space was associated with subsequent hemorrhagic expansion in patients with acute lobar ICH but not in patients with nonlobar ICH [157].

●Intraventricular extension – Expansion of ICH into the intraventricular space has been reported in up to 40 to 60 percent of patients in various studies and is associated with neurologic complications and worse outcomes [158-160].

EVALUATION AND DIAGNOSIS — ICH is a neurologic and medical emergency because it is associated with a high risk of ongoing bleeding, progressive neurologic deterioration, permanent disability, and death [161,162].

As the diagnostic evaluation is proceeding, patients with ICH may also need acute interventions, possibly including intubation and mechanical ventilation, anticoagulation reversal, blood pressure control, intracranial pressure monitoring and treatment, and consideration of the need for ventriculostomy or surgical hematoma evacuation (table 1). These issues are discussed in detail separately. (See "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis".)

Initial evaluation — Clinical suspicion for ICH is based upon features such as acute onset of gradually worsening symptoms and increasing neurologic deficit, particularly if accompanied by severe headache, vomiting, severe hypertension, and decreased level of consciousness or coma. However, the distinction between brain hemorrhage and ischemia cannot be made on the basis of clinical characteristics alone [162]. Importantly, headache may be absent in some cases of ICH. (See 'Clinical presentation' above.)

Neuroimaging with brain computed tomography (CT) or magnetic resonance imaging (MRI) is mandatory to confirm the diagnosis of ICH and to exclude ischemic stroke and stroke mimics as possible causes [162].

Laboratory and other dignostic testing — Routine laboratory evaluation to evaluate for underlying causes or associated risks in patients with ICH includes [162]:

●Complete blood count, electrolytes, blood urea nitrogen, creatinine, and glucose

●Prothrombin time (with international normalized ratio [INR]) and activated partial thromboplastin time for all patients; thrombin clotting time for patients taking direct oral anticoagulants (and/or ecarin clotting time where available for patients known or suspected to be taking direct thrombin inhibitors)

●Cardiac-specific troponin

●Toxicology screen to detect cocaine and other sympathomimetic drugs

●Urinalysis

●Pregnancy test in a female of childbearing age

An electrocardiogram (ECG) obtained at baseline may help identify patients with ICH and coexistent cardiac dysfunction [163,164]. A repeat ECG is also typically obtained along with an echocardiogram for those with heart failure or hemodynamic instability. (See "Electrocardiogram in the diagnosis of myocardial ischemia and infarction" and "Clinical manifestations and diagnosis of stress (takotsubo) cardiomyopathy".)

An electroencephalogram is reserved for patients with seizures or those with encephalopathy not explained by the location or size of the ICH. (See "Nonconvulsive status epilepticus: Classification, clinical features, and diagnosis", section on 'When to suspect NCSE' and 'Neurologic signs and ICH location' above.)

Subsequent imaging — Once ICH is confirmed by head CT or MRI, additional imaging is warranted in the event of clinical deterioration to evaluate for ICH expansion or rebleeding. A follow-up imaging study may also be performed to document that the bleeding has stabilized.

Subsequent imaging is also typically performed to identify an underlying cause of the ICH to help guide preventive measures to reduce the risk of recurrent hemorrhage. The necessity for further evaluation to determine the cause of ICH varies with the clinical setting. (See "Spontaneous intracerebral hemorrhage: Secondary prevention and long-term prognosis", section on 'Follow-up neuroimaging'.)

For some patients, the etiology of the ICH is identified with initial neuroimaging studies, and further imaging studies are not required. An example may include a stable patient with an acute ICH in the putamen suggestive of hypertensive vasculopathy who has a history of longstanding hypertension.

For other patients, initial imaging studies do not sufficiently exclude other causes of ICH and follow-up evaluation is required. We obtain follow-up imaging when an underlying cause is suspected by clinical features or initial imaging findings (table 3) [162]. Initial imaging findings that may suggest a specific underlying cause include:

•Early perihematomal edema out of proportion to the underlying ICH (image 9) [165]

•ICH within arterial vascular territory suggesting primary ischemic infarction (image 7)

•Multifocal hemorrhage (image 20)

•Isolated intraventricular hemorrhage (image 21)

The assessment of the possible underlying structural pathology may be shrouded and distorted by the hematoma or surrounding edema. In these cases, delayed imaging performed after bleeding and edema have resolved may identify patients with underlying structural abnormalities at high risk for ICH recurrence (algorithm 1). (See "Spontaneous intracerebral hemorrhage: Secondary prevention and long-term prognosis", section on 'Follow-up neuroimaging'.)

Brain MRI with contrast is the preferred modality to help identify the underlying cause of ICH for most patients. For patients who are unable to undergo MRI, head CT with contrast is a less sensitive alternative option. Vascular imaging (eg, CT or MR angiography or digital subtraction angiography) is performed when an underlying vascular lesion is suspected [29,30,126,166]. (See "Spontaneous intracerebral hemorrhage: Secondary prevention and long-term prognosis", section on 'Follow-up neuroimaging'.)

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 topics (see "Patient education: Intracerebral hemorrhage (The Basics)" and "Patient education: Arteriovenous malformations in the brain (The Basics)")

●Beyond the Basics topics (see "Patient education: Stroke symptoms and diagnosis (Beyond the Basics)" and "Patient education: Hemorrhagic stroke treatment (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Underlying causes – Injury to brain parenchyma occurs via hematoma expansion and perilesional edema as well as secondary excitotoxic and inflammatory injury from the breakdown of the blood-brain barrier. Common conditions associated with intracerebral hemorrhage (ICH) include hypertension, cerebral amyloid angiopathy, and ruptured vascular malformation. Other etiologies include cerebral venous thrombosis, vasculopathies, primary or metastatic tumors, and coagulopathies. (See 'Pathogenesis and etiologies' above.)

●Risk factors – Major risk factors for spontaneous ICH include older age, hypertension, and the use of antithrombotic (antiplatelet and anticoagulant) therapy. (See 'Risk factors' above.)

●Presenting signs and symptoms – The signs and symptoms of ICH vary according to the location and size of the hemorrhage (table 1). Patients typically present with an acute onset of a focal neurologic deficit such as hemiparesis, aphasia, or visual impairment corresponding to the part of the brain affected.

The neurologic symptoms and signs may be progressive over minutes or a few hours (figure 1). Headache, vomiting, and a decreased level of consciousness develop if the hemorrhage is large. Patients may also present with stupor or coma due to associated reversible causes such as acute metabolic derangements or seizure. (See 'Clinical presentation' above.)

●Initial diagnostic imaging – Neuroimaging with head computed tomography (CT) or magnetic resonance imaging (MRI) is mandatory to confirm the diagnosis of ICH and to exclude ischemic stroke and stroke mimics as alternative causes to symptoms. Acute imaging also provides information about the severity of the hemorrhage, the risk of expansion of bleeding, and the underlying cause of the ICH (image 19 and table 3). (See 'Evaluation and diagnosis' above and 'Brain imaging' above.)

•Both CT or MRI are considered first-choice imaging options for the emergency diagnosis and assessment of ICH. CT angiography may be performed along with a noncontrast head CT to help identify an underlying vascular cause to the ICH. (See 'Brain imaging' above.)

•Predictors of hemorrhage growth associated with neurologic deterioration include a shorter time from symptom onset to initial imaging, initial ICH volume, antithrombotic medication use, and imaging signs of ICH heterogeneity on noncontrast CT or focal contrast extravasation on CT angiography (image 18 and image 19). (See 'Predicting hemorrhage expansion' above.)

●Subsequent imaging evaluation – Additional imaging may be warranted after the diagnosis of ICH in the event of clinical deterioration to evaluate for ICH expansion or rebleeding and to document that the bleeding has stabilized. (See 'Subsequent imaging' above.)

We also obtain follow-up imaging when an underlying cause is suspected by clinical features or initial imaging findings to help guide preventive measures to reduce the risk of recurrent hemorrhage (table 3). Brain MRI with contrast is the preferred modality to help identify the underlying cause of ICH for most patients. Vascular imaging (eg, CT or MR angiography or digital subtraction angiography) is performed when an underlying vascular lesion is suspected. (See 'Subsequent imaging' above.)