Intracerebral hemorrhage

Intracerebral hemorrhage (ICH) is bleeding into the tissue of the brain following rupture of a blood vessel. Spontaneous, nontraumatic ICH is a form of hemorrhagic stroke. It differs from subarachnoid hemorrhage, which occurs in the space surrounding the brain, and from subdural and epidural hemorrhages, which occur outside brain tissue. Intracerebral bleeding may also follow trauma or accompany an underlying structural lesion. (ninds.nih.gov)

Spontaneous ICH is an important cause of death and long-term disability. Major underlying vessel disorders include cerebral arteriolosclerosis and cerebral amyloid angiopathy; uncontrolled hypertension is its principal modifiable risk factor. Injury results from the accumulating blood and from subsequent physiological and cellular responses. The location, size, and evolution of the hemorrhage influence its clinical effects and outcome. (professional.heart.org)

Clinical assessment combines neurological examination with brain imaging to establish the diagnosis and investigate its cause. Care encompasses acute medical management, selected surgical procedures, and multidisciplinary rehabilitation. Understanding the underlying vascular disease also informs the study of recurrence and prevention. (professional.heart.org)

Definition and anatomical classification

Intracerebral hemorrhage (ICH) is bleeding into the parenchyma—the functional tissue—of the brain, producing a localized collection of blood termed a hematoma. The 2022 American Heart Association/American Stroke Association (AHA/ASA) guideline defines ICH in terms of brain injury caused by acute escape of blood from a ruptured cerebral vessel into this tissue. It is one anatomical subtype of intracranial hemorrhage, the broader category encompassing bleeding anywhere within the skull. Other compartments include the ventricular system, the subarachnoid space surrounding the brain, and the subdural and epidural spaces associated with its coverings. Thus, “intracranial” and “intracerebral” are not interchangeable: the former specifies the skull as the outer boundary, whereas the latter usually identifies bleeding within brain tissue. (ahajournals.org)

In stroke medicine, spontaneous ICH denotes nontraumatic bleeding and is distinguished from hemorrhagic contusions or hematomas caused by head injury. “Spontaneous” does not imply that no underlying cause exists; nontraumatic hemorrhage may arise from vessel disease, a vascular malformation, or another identifiable disorder. The scope of the 2022 AHA/ASA guideline is narrower, excluding hemorrhages with a demonstrated structural cause such as a vascular malformation or bleeding tumor. A separate distinction concerns hemorrhagic transformation of an ischemic infarct: blood enters tissue already injured by interrupted blood supply, rather than constituting the initiating event. Transformation ranges from scattered petechial bleeding to a confluent parenchymal hematoma with mass effect; consequently, some stroke definitions include its hematoma forms under ICH while retaining the preceding infarction as a distinct mechanism. (ahajournals.org)

Parenchymal hemorrhages are commonly described in four anatomical groups. Lobar hemorrhage involves the cerebral cortex or underlying white matter of a cerebral lobe, such as a frontal or occipital hematoma. Deep hemispheric hemorrhage involves subcortical structures, notably the basal ganglia or thalamus; a putaminal hematoma is a basal-ganglia example. Brainstem hemorrhage is recorded separately, with the pons an important specific site. Cerebellar hemorrhage involves the cerebellum, rather than the cerebral hemispheres. Lobar and deep hemispheric lesions are supratentorial, whereas brainstem and cerebellar lesions are infratentorial. These labels describe location, not a definitive cause. Hematomas can cross anatomical boundaries, so extensive lesions may require a mixed-location description rather than assignment to a single compartment. (pmc.ncbi.nlm.nih.gov)

Ventricular extension means that a parenchymal hematoma also releases blood into the ventricular system, whose cavities normally contain cerebrospinal fluid. Such a lesion is described as ICH with intraventricular hemorrhage (IVH), preserving both its tissue location and its additional compartment of spread. Isolated IVH, by contrast, denotes ventricular bleeding without extension from a parenchymal or subarachnoid hemorrhage. Terminology varies: the 2013 AHA/ASA stroke definition includes blood collections in either brain parenchyma or the ventricular system within its definition of ICH, whereas anatomical descriptions frequently distinguish intraparenchymal hemorrhage from IVH explicitly. (pmc.ncbi.nlm.nih.gov)

Hemorrhagic stroke is not an exact synonym for ICH. It includes spontaneous ICH and subarachnoid hemorrhage, in which blood enters the space between the arachnoid and pia mater rather than primarily accumulating within brain tissue. Conversely, a traumatic intracerebral hematoma is an intracerebral hemorrhage anatomically but is not classified as a spontaneous hemorrhagic stroke. Anatomical compartment and clinical mechanism therefore provide complementary classifications. (pmc.ncbi.nlm.nih.gov)

Epidemiology and population burden

The Global Burden of Disease Study (GBD) estimated 3.44 million new intracerebral hemorrhages worldwide in 2021. The crude incidence rate was 43.65 per 100,000 people per year, whereas the age-standardized rate was 40.83 per 100,000, with a 95% uncertainty interval of 36.20–45.21. These measures answer different questions: absolute case numbers describe the volume of disease, crude rates relate cases to the actual population, and age-standardized rates permit comparisons after adjusting for differences in population age structure. The World Stroke Organization’s 2025 factsheet, using these 2021 estimates, reported that ICH accounted for approximately 29% of incident strokes globally. (journals.sagepub.com)

The United States has a substantially smaller proportion of strokes attributed to ICH. The American Heart Association/American Stroke Association’s 2022 guideline summary estimated that approximately 10% of the 795,000 strokes occurring annually in the country were intracerebral hemorrhages. This is a proportion of all stroke events, not an incidence rate per population, and it should not be interpreted as a universal estimate. (professional.heart.org) Geographical differences are considerable even after age adjustment: a published analysis of GBD 2021 estimated national ICH incidence rates ranging from 9.9 per 100,000 in Switzerland to 198.1 in the Solomon Islands. These are modeled population estimates rather than directly enumerated national case counts. The same analysis estimated a 31.4% decline in global age-standardized incidence between 1990 and 2021. (pmc.ncbi.nlm.nih.gov)

Incidence increases steeply with age, although older adults do not account for all, or necessarily most, cases. Charlotte van Asch and colleagues’ 2010 meta-analysis of population-based studies found that incidence among people older than 85 years was approximately 9.6 times that among those aged 45–54. (pubmed.ncbi.nlm.nih.gov) In the GBD 2021 estimates, 56% of new hemorrhages occurred before age 70, and 18% occurred at ages 15–49; men accounted for 56% of incident cases. (journals.sagepub.com) The apparent contrast between high incidence in advanced age and substantial numbers of younger patients reflects the much larger populations in younger age groups. Differences by age, sex, geographical region, or recorded ethnicity describe epidemiological patterns; these categories alone do not establish a biological cause.

Early death and persistent disability contribute substantially to the burden. A systematic review published in December 2024 synthesized 70 prospective population-based studies from 26 countries, including 19,470 patients. Pooled one-month case fatality was 35.5% (95% confidence interval 32.3–38.9), broadly consistent with the 2022 AHA/ASA estimate of 30–40% early mortality. (pmc.ncbi.nlm.nih.gov) In ten studies reporting functional outcomes, 31.2% of patients had a favorable outcome at 3–12 months. Definitions were not identical: nine studies used modified Rankin Scale scores of 0–2, whereas one used 0–3. Consequently, this proportion is not a uniform measure of complete recovery, and neither it nor case fatality describes every population equally. (pmc.ncbi.nlm.nih.gov)

GBD estimated 3.31 million deaths and 79.46 million disability-adjusted life-years attributable to ICH worldwide in 2021. Annual deaths and incident cases are separate population measures, not a cohort from which an individual case-fatality probability can be calculated. (journals.sagepub.com) The burden is unequally distributed: the AHA/ASA identifies disproportionate effects on lower-resourced populations within the United States and internationally. (professional.heart.org) The World Stroke Organization reported that age-standardized incidence declined more slowly in countries with low or low-middle sociodemographic index than in higher-index settings. Population burden therefore reflects not only demographic composition, but also differences in exposure to vascular risks and the resources available for prevention, acute care, and recovery. (journals.sagepub.com)

Causes and risk factors

Arteriolosclerosis and cerebral amyloid angiopathy (CAA) are the principal small-vessel diseases underlying spontaneous intracerebral hemorrhage without an identifiable structural lesion. Arteriolosclerosis involves concentric, hyaline thickening and degeneration of small arterial walls and is associated with chronic hypertension, aging, and diabetes. It preferentially affects penetrating arterioles supplying the basal ganglia, thalamus, brainstem, and deep nuclei of the cerebellum, producing the characteristic distribution of deep hemorrhages. Hypertension is therefore both a contributing exposure and a marker of an underlying vascular disorder, rather than an anatomical diagnosis. Arteriolosclerosis can also produce lobar bleeding; location suggests, but does not establish, the cause. (ahajournals.org)

CAA involves deposition of amyloid-β in vessels of the leptomeninges and cerebral cortex, weakening their walls and favoring lobar hemorrhage. Its distribution differs from the deep penetrating-vessel pattern of arteriolosclerosis, although both diseases may coexist. Increasing age and the apolipoprotein E gene alleles $\varepsilon2$ and $\varepsilon4$ are associated with CAA-related hemorrhage. These alleles confer susceptibility rather than determining that bleeding will occur. Rare familial forms of CAA, including forms caused by mutations in the amyloid precursor protein gene, can present earlier than sporadic disease. (ahajournals.org)

Structural causes are distinct from these diffuse small-vessel disorders. A cerebral arteriovenous malformation contains an abnormal vascular network connecting arteries directly to veins without an intervening capillary bed; high flow, elevated venous pressure, associated aneurysms, and impaired venous drainage can contribute to rupture. A cerebral cavernous malformation, by contrast, is a low-flow cluster of thin-walled vascular spaces that can bleed into adjacent brain tissue. Primary tumors and metastases can also hemorrhage. In a 2024 retrospective series of 116 patients with tumor-associated intracerebral hemorrhage, examples included glioblastoma, melanoma metastases, and lung-cancer metastases. Such a selected series describes tumor-related bleeding, not the proportion of all hemorrhages caused by tumors. (ahajournals.org)

Anticoagulants impair clot formation and can precipitate clinically significant bleeding in vessels already affected by small-vessel disease. Anticoagulant-associated hemorrhage therefore need not have a single cause. Warfarin reduces vitamin K–dependent coagulation-factor activity, whereas direct oral anticoagulants inhibit thrombin or factor Xa. In Christian Ruff and colleagues’ 2014 meta-analysis of four randomized trials involving 71,683 participants with atrial fibrillation, direct oral anticoagulants had a lower risk of intracranial hemorrhage than warfarin, with a relative risk of 0.48. This endpoint included intracranial bleeding beyond intracerebral hemorrhage and did not imply absence of risk. Congenital hemophilia, acquired coagulation-factor deficiencies associated with liver failure or disseminated intravascular coagulation, and low platelet counts can independently cause or aggravate bleeding. (ahajournals.org)

Heavy alcohol consumption and sympathomimetic drugs are established exposure-related risk factors. The 2017 Ethnic/Racial Variations of Intracerebral Hemorrhage study defined heavy drinking as at least five drinks daily and reported an adjusted odds ratio of 1.77 compared with no alcohol consumption. That observational association is not an individual probability of hemorrhage. Cocaine and amphetamines, including methamphetamine, are associated with intracerebral hemorrhage; marked sympathetic activation and acute blood-pressure elevation provide a plausible precipitating mechanism. Blood-pressure exposure, heavy drinking, and drug exposure are potentially modifiable, whereas age and inherited susceptibility are not. Etiological descriptions distinguish these exposures from the vessel pathology itself: a hemorrhage may arise from CAA or arteriolosclerosis while anticoagulation or an acute pressure surge contributes to its occurrence. (n.neurology.org)

Hematoma evolution and mechanisms of injury

Rupture of an intracerebral vessel releases blood into the brain parenchyma, forming a hematoma that disrupts tissue and compresses surrounding structures. The volume visible on the first scan is a measurement at that time, not necessarily the final volume: bleeding may continue or recur after presentation. Enlargement may reflect continued leakage from the original vessel and, potentially, mechanical tearing of neighboring small vessels as the collection expands. The 2022 American Heart Association/American Stroke Association guideline distinguishes these immediate pressure effects from secondary cellular injury caused by the hematoma and its breakdown products. (ahajournals.org)

Early hematoma expansion is concentrated in the first hours, although it can occur throughout the first 24 hours. In a prospective study published in 1997, Thomas Brott and colleagues evaluated 103 patients whose initial computed tomography examination occurred within three hours of symptom onset. Substantial hematoma growth occurred in 26% between the baseline scan and a scan one hour later; an additional 12% showed substantial growth between the one-hour and twenty-hour follow-up scans. Early enlargement was associated with neurological deterioration. These observations describe a selected, early-presenting cohort rather than a universal expansion rate: the apparent frequency depends partly on how soon the first scan is obtained and how growth is defined. (ahajournals.org)

Primary mechanical injury includes local compression and displacement of tissue. A sufficiently large hematoma produces mass effect, potentially causing midline shift, increased intracranial pressure, or herniation; ventricular blood or compression can also obstruct cerebrospinal-fluid circulation. Raised intracranial pressure can compromise cerebral perfusion. (ahajournals.org) However, reduced blood flow immediately around the clot does not necessarily indicate an ischemic penumbra analogous to that of an arterial-occlusion stroke. In a 2001 positron-emission tomography study of 19 patients examined 5–22 hours after onset, Allyson Zazulia and colleagues found that perihematomal oxygen metabolism fell more than blood flow, with reduced rather than increased oxygen extraction. This supported metabolic suppression, rather than ischemia, in the surrounding tissue studied. (pubmed.ncbi.nlm.nih.gov)

Secondary injury develops as extravasated blood interacts with brain tissue. Thrombin, which participates in clot formation, also has effects on barrier permeability and cell survival. In a 1997 rat experiment, intracerebral thrombin infusion disrupted the blood–brain barrier and caused parenchymal-cell death without a corresponding alteration in cerebral blood flow. Later, red-cell lysis releases hemoglobin, whose degradation generates heme and iron. Experimental work by Feng-Ping Huang, Guohua Xi, and colleagues, published in 2002, implicated hemoglobin degradation products in delayed brain edema. Iron-mediated oxidative reactions can damage cellular components; these experimental findings establish plausible injury pathways, but do not by themselves quantify their contribution in individual patients. (pubmed.ncbi.nlm.nih.gov)

The hematoma also initiates inflammatory responses, involving resident microglia and infiltrating leukocytes. A 2019 human postmortem study documented neutrophils in perihematomal tissue as early as one day after hemorrhage, with infiltration evident within two days in most cases examined. Activated microglia and macrophages participate in removing blood cells and damaged tissue, while inflammatory-cell products can contribute to oxidative injury and barrier dysfunction. Thus, immune activation includes both potentially injurious activity and clearance functions; it is not a single, uniformly destructive process. (pmc.ncbi.nlm.nih.gov)

Perihematomal edema is accumulation of water around the clot, distinct from enlargement of the blood collection itself. Barrier disruption permits fluid to enter surrounding tissue, so swelling may continue after bleeding has stopped. In a serial magnetic resonance imaging study published in 2011, Chitra Venkatasubramanian and colleagues found that edema grew fastest during the first 48 hours and reached its maximum volume at approximately $12 \pm 3$ days. This time course was an observed cohort average, not a fixed schedule for every hemorrhage. Edema adds to the space-occupying burden and can therefore intensify tissue displacement even when hematoma volume is stable. (pmc.ncbi.nlm.nih.gov)

Clinical presentation and neurological assessment

Focal neurological deficits usually begin suddenly, but their pattern depends on the structures affected. Deep hemispheric hemorrhage involving motor pathways can produce weakness of the opposite face, arm, and leg; involvement of sensory pathways can cause contralateral numbness or loss of sensation. Lobar hemorrhage may produce language disturbance, weakness, sensory loss, or visual-field abnormalities. Aphasia—impaired production or comprehension of language—is associated particularly with injury to the language-dominant hemisphere, usually the left; it differs from dysarthria, in which articulation is impaired. Occipital involvement can cause homonymous hemianopia, loss of the same half of the visual field in both eyes. These patterns overlap rather than identifying the bleeding site with certainty: an analysis of 225 patients in the NINDS Stroke Data Bank found that putaminal, thalamic, and lobar hemorrhages could all present with hemiparesis, sensory loss, and higher cortical deficits. (pubmed.ncbi.nlm.nih.gov)

Posterior-fossa hemorrhage may present differently. Bleeding in the cerebellum can produce marked imbalance, occipital headache, and vomiting without prominent hemiparesis; large pontine hemorrhage can cause early coma, extensive paralysis, and pinpoint pupils. Headache, nausea, vomiting, and impaired consciousness also occur with hemispheric hemorrhage, but none is obligatory. Seizures may accompany the initial presentation, particularly with lobar bleeding: a hospital-based registry study published in 2023, comparing 165 lobar with 133 deep subcortical hemorrhages, found early seizures associated with lobar location. Nevertheless, symptoms and examination findings alone cannot reliably distinguish hemorrhage from ischemic stroke; urgent brain imaging establishes that distinction. (michigan.gov)

The initial history records the exact time symptoms began or, when onset was unwitnessed, the last time the patient was known to be neurologically normal. It also documents the sequence of deficits, headache, seizures, and changes in alertness. The 2022 AHA/ASA guideline specifies medication names, doses, and time of last ingestion, especially anticoagulants such as warfarin, direct oral anticoagulants, antiplatelet agents such as aspirin, and thrombolytic drugs. Relevant history includes hypertension, previous stroke or hemorrhage, cognitive impairment, alcohol or sympathomimetic exposure, and liver, renal, malignant, or hematological disease that may affect coagulation. (ahajournals.org)

Assessment combines vital signs and airway, breathing, and circulation with a structured neurological examination. The National Institutes of Health Stroke Scale (NIHSS) documents consciousness, gaze, visual fields, facial movement, limb strength, coordination, sensation, language, dysarthria, and neglect. Testing naming and comprehension helps distinguish language impairment from poorly articulated speech; examining both sides records the distribution of motor and sensory deficits. A standardized baseline permits clearer communication between clinicians and comparison with subsequent examinations. (cde.nlm.nih.gov)

For impaired consciousness, the Glasgow Coma Scale records eye opening, verbal response, and motor response separately. When all components are testable, their sum ranges from 3 to 15, but the component findings provide information obscured by the total alone. Intubation, sedation, or communication difficulties are documented as assessment limitations; an untestable response is not equivalent to an absent response. Repeated examinations identify new weakness, reduced responsiveness, or other deterioration. The 2022 guideline describes the first 12–24 hours after onset as a period of particularly high risk for neurological decline. (glasgowcomascale.org)

Imaging and etiological diagnosis

Noncontrast computed tomography (CT) is the usual initial examination because it is rapid, widely available, and highly sensitive to acute intracerebral blood, which generally appears hyperdense relative to surrounding brain tissue. It establishes the hemorrhage’s location and extent and shows ventricular extension, ventricular enlargement from hydrocephalus, and mass effect, including midline shift. These findings distinguish hemorrhage from ischemic stroke and document complications that cannot be reliably inferred from symptoms alone. (ahajournals.org)

Hematoma volume can be estimated using $V \approx ABC/2$, a bedside method validated by Rashmikant Kothari and colleagues in 1996. Here, $A$ is the greatest hematoma diameter, $B$ its perpendicular diameter on the same CT slice, and $C$ the approximate craniocaudal extent, calculated from the number of involved slices and their thickness. Measurements in centimeters yield volume in milliliters; dimensions of $6 \times 3 \times 3$ cm, for example, give approximately 27 mL. The method approximates an ellipsoid; slice-by-slice segmentation provides a more direct measurement, particularly when the clot has an irregular configuration. Ventricular blood is assessed separately from the parenchymal hematoma. (pubmed.ncbi.nlm.nih.gov)

CT angiography (CTA) investigates arterial abnormalities, including arteriovenous malformations, aneurysms, and dural arteriovenous fistulas; CT or magnetic resonance venography investigates suspected cerebral venous thrombosis. Selection for further vascular investigation incorporates age, hemorrhage location, and imaging evidence of small-vessel disease rather than assuming that a history of hypertension establishes the cause. Catheter digital subtraction angiography remains the reference examination for vascular malformations, especially when noninvasive imaging is suspicious or clinical suspicion persists despite negative results. In Charlotte van Asch and colleagues’ 2015 DIAGRAM study of 298 selected adults aged 18–70 years, CTA identified a macrovascular cause in 17%; the combined strategy of CTA, MRI/MR angiography, and catheter angiography increased the yield to 23%. These figures describe that selected cohort, not all patients with intracerebral hemorrhage. (pmc.ncbi.nlm.nih.gov)

Magnetic resonance imaging (MRI) investigates causes poorly characterized by angiography, including cavernous malformations, tumors, and small-vessel disease. Gradient-echo and susceptibility-weighted sequences detect previous microbleeds and cortical superficial siderosis, which help characterize underlying disorders such as cerebral amyloid angiopathy. Contrast-enhanced T1-weighted imaging investigates an underlying mass; delayed MRI may reveal a lesion obscured by the acute hematoma. (heart.org)

Repeat CT documents hematoma expansion, new ventricular bleeding, hydrocephalus, and increasing edema. The 2022 AHA/ASA guideline describes scans at approximately 6 and 24 hours after onset as useful in neurologically stable patients; deterioration prompts earlier reassessment, while subsequent imaging depends on the clinical course. CTA also detects the spot sign: focal contrast pooling within the hematoma, indicating contrast extravasation and an increased likelihood of expansion. In the 2012 prospective PREDICT study, its sensitivity for expansion was 51% and specificity 85%, so absence does not exclude subsequent growth. Heterogeneous density and irregular margins on noncontrast CT are additional expansion-risk markers, not definitive predictions. (heart.org)

Laboratory assessment complements imaging. A complete blood count measures platelet number and may identify hematological abnormalities; prothrombin time, international normalized ratio, and activated partial thromboplastin time assess coagulation disturbances. Results are interpreted alongside anticoagulant exposure, including warfarin, dose timing, and renal function. Electrolytes, glucose, and kidney-function tests identify physiological abnormalities relevant to interpreting the presentation and planning further investigations. (heart.org)

Acute medical management

Organized stroke units provide specialist nursing, repeated neurological assessment, and coordinated medical care; patients with impaired airway protection, respiratory failure, or severe neurological deterioration require intensive or neurocritical care. Monitoring follows consciousness, pupils, focal deficits, blood pressure, oxygenation, temperature, and glucose, with deterioration prompting urgent reassessment. Endotracheal intubation supports patients unable to protect their airway or maintain adequate ventilation. The 2025 ESO/EANS guideline supports organized stroke-unit admission for patients not requiring intensive care. (ahajournals.org) Coordinated protocols have outcome evidence: Lu Ma and colleagues’ 2023 INTERACT3 cluster-randomized trial enrolled 7,036 patients at 121 hospitals. A bundle combining early blood-pressure reduction, glucose and fever management, and anticoagulation reversal improved the distribution of six-month functional outcomes compared with usual care (common odds ratio for worse outcome, 0.86; 95% confidence interval, 0.76–0.97). This result establishes the bundle’s effect, not the independent contribution of each component. (pmc.ncbi.nlm.nih.gov)

Controlled blood-pressure reduction aims to limit further bleeding while preserving cerebral perfusion. The 2022 AHA/ASA guideline describes targeting systolic pressure around 140 mmHg and maintaining 130–150 mmHg as reasonable in mild-to-moderate hemorrhage presenting at 150–220 mmHg; reduction below 130 mmHg may be harmful. Smooth, sustained control is emphasized over abrupt reductions or large fluctuations. INTERACT2 found modest benefits in secondary functional analyses, whereas ATACH-2 found no additional functional benefit from its more intensive target and more renal adverse events. Evidence is less secure for massive hemorrhage, compromised cerebral perfusion, and patients requiring decompression. (ahajournals.org)

Anticoagulant reversal addresses impaired clot formation and the risk of hematoma expansion. For vitamin K antagonists such as warfarin, four-factor prothrombin complex concentrate rapidly replaces depleted coagulation factors, while intravenous vitamin K restores their production and prevents rebound anticoagulation; concentrate is preferred to fresh-frozen plasma for rapid correction. Direct oral anticoagulants require different approaches: idarucizumab binds the thrombin inhibitor dabigatran, whereas prothrombin complex concentrates provide nonspecific hemostatic support for factor Xa inhibitor–associated bleeding. Drug identity, timing of the last dose, renal function, and evidence of residual anticoagulant activity influence selection. (ahajournals.org)

Andexanet alfa, a modified factor Xa decoy, illustrates the distinction between hemostatic efficacy and patient-centered benefit. In the 2024 ANNEXA-I trial, effective hemostasis occurred in 67.0% of evaluable andexanet recipients versus 53.1% receiving usual care, predominantly prothrombin complex concentrate. Thrombotic events occurred in 10.3% versus 5.6%, including ischemic stroke in 6.5% versus 1.5%; no appreciable difference in 30-day functional outcome or death was demonstrated. (nejm.org) On December 18, 2025, the United States Food and Drug Administration concluded that Andexxa’s serious risks outweighed its benefits and reported that U.S. commercial sales would end on December 22, 2025. Earlier guideline recommendations therefore do not establish its continuing U.S. availability. (fda.gov)

For raised intracranial pressure, bolus hyperosmolar therapy can transiently reduce pressure, but routine prophylactic administration has no established functional benefit; corticosteroids are not used to treat hemorrhage-associated pressure elevation. Hydrocephalus contributing to depressed consciousness may require ventricular drainage, a procedural rather than purely medical intervention. Clinical seizures and relevant electrographic seizures receive antiseizure treatment. Unexplained or fluctuating consciousness can justify continuous electroencephalography for at least 24 hours to detect otherwise unrecognized seizures. Routine antiseizure prophylaxis in patients without seizures has not demonstrated improved functional outcomes. (ahajournals.org)

Temperature and glucose management distinguish correction of physiological disturbance from proven recovery benefits. Fever prompts assessment for infection and other causes, with antipyretic treatment incorporated into care bundles; isolated device-based temperature control has not demonstrated improved functional outcome or survival. Glucose monitoring identifies hypoglycemia and marked hyperglycemia, both clinically important disturbances. Intensive normalization risks hypoglycemia, and an optimal ICH-specific glucose target remains unestablished. The 2025 ESO/EANS guideline distinguishes support for bundled care from the insufficient evidence that temperature or glucose management alone reduces death or dependence. (pmc.ncbi.nlm.nih.gov)

Neurosurgical and ventricular procedures

Open hematoma evacuation uses a craniotomy—a temporarily removed skull flap—to expose and remove the clot, relieve local mass effect, and permit direct control of bleeding. Its potential benefit must be balanced against injury incurred while reaching the hematoma, particularly when blood lies deep within the brain. Routine evacuation of supratentorial hemorrhage has not demonstrated a consistent functional advantage: A. David Mendelow and colleagues’ STICH trial (2005; 1,033 participants) and STICH II (2013; 601 participants) found no significant improvement in their primary outcomes. STICH II specifically examined superficial lobar hematomas without ventricular extension. Nevertheless, the 2022 AHA/ASA guideline recognizes open evacuation as a possible lifesaving intervention in patients who are deteriorating neurologically. (pmc.ncbi.nlm.nih.gov)

Minimally invasive evacuation seeks to reduce clot volume while limiting disruption of surrounding tissue. Techniques include image-guided catheter aspiration, endoscopic removal, and trans-sulcal access through a tubular retractor. In the ENRICH randomized controlled trial, reported by Gustavo Pradilla and colleagues in 2024, 300 patients with lobar or anterior basal ganglia hematomas of 30–80 mL were assigned to early minimally invasive removal plus medical management or medical management alone; surgery began within 24 hours of onset. The mean utility-weighted modified Rankin Scale score at 180 days was 0.458 with surgery and 0.374 with medical management, a difference of 0.084 (95% Bayesian credible interval, 0.005–0.163). Benefit was attributable to the lobar group; enrollment of patients with anterior basal ganglia hemorrhage stopped under a prespecified futility rule. These results do not establish equivalent benefit for every location or evacuation technique. (pmc.ncbi.nlm.nih.gov)

Catheter evacuation combined with thrombolysis removes blood gradually rather than through a single operative extraction. MISTIE III, reported by Daniel Hanley and colleagues in 2019, randomized 506 patients with supratentorial hematomas of at least 30 mL. Image-guided aspiration was followed by repeated administration of alteplase through a catheter positioned within the remaining clot. The trial did not significantly improve its primary functional endpoint: the adjusted estimated proportion with a modified Rankin Scale score of 0–3 at one year was 45% with intervention and 41% with standard medical care (absolute difference, 4 percentage points; 95% confidence interval, −4 to 12). Thus, successful volume reduction and demonstrated functional benefit are distinct outcomes. (pmc.ncbi.nlm.nih.gov)

Decompressive craniectomy instead removes a substantial skull segment to provide space for swollen brain tissue; the hematoma may remain in place. Selection is influenced by coma, major midline shift, and intracranial pressure unresponsive to medical treatment. SWITCH (2024) examined decompression without clot evacuation in severe basal ganglia or thalamic hemorrhage. Its primary result provided weak evidence of benefit rather than definitive proof, with an adjusted risk ratio of 0.77 for death or very severe disability at 180 days (95% confidence interval, 0.59–1.01). (heart.org)

Hemorrhage in the cerebellum presents a different anatomical problem: the confined posterior fossa allows a hematoma to compress the brainstem or obstruct the fourth ventricle. The 2022 AHA/ASA guideline recommends immediate evacuation, with or without ventricular drainage, for neurological deterioration, brainstem compression, obstructive hydrocephalus, or a hematoma volume of at least 15 mL. The principal established objective is reduced mortality; improved functional outcome remains uncertain. Ventricular drainage alone does not remove the posterior-fossa mass and may be insufficient when brainstem compression persists. (heart.org)

External ventricular drainage places a catheter within a cerebral ventricle to divert cerebrospinal fluid and reduce pressure when hemorrhage obstructs its circulation. Hydrocephalus contributing to reduced consciousness is a major indication; ventricular blood alone is not an interchangeable indication. Unlike a catheter inserted into a parenchymal hematoma, a ventricular drain does not directly evacuate that hematoma. Intraventricular thrombolysis through the drain is a separate adjunct intended to accelerate ventricular clot clearance: CLEAR III did not demonstrate improved functional outcome despite lower mortality. (heart.org)

Complications, prognosis, and outcome measurement

Neurological impairment creates risks beyond the hemorrhage itself. Impaired swallowing and reduced consciousness allow food, liquids, or oral secretions to enter the airway, contributing to aspiration pneumonia; swallowing assessment before oral intake is therefore a component of organized inpatient care. Paralysis and immobility promote venous stasis, increasing the risk of deep-vein thrombosis and pulmonary embolism. Other medical complications include urinary tract infection, sepsis, and acute kidney injury. A United States administrative-data study covering 2004–2013 reported pneumonia in 7.8% of ICH hospitalizations, urinary tract infection in 14.8%, deep-vein thrombosis in 2.7%, and pulmonary embolism in 0.7%. These recorded hospital frequencies depend on ascertainment and case mix and do not represent universal patient-level risks. (heart.org)

Lower consciousness scores, larger hematoma volume, infratentorial location, and extension into the ventricles are associated with higher early mortality. Location matters because hemorrhage can disrupt critical pathways or compress structures within restricted anatomical spaces; ventricular blood can obstruct cerebrospinal-fluid circulation and produce hydrocephalus. These features support risk stratification, but an association across a cohort is not a certain prediction for an individual. In particular, an admission assessment does not incorporate the subsequent course of hematoma expansion, reversible complications, or response to treatment. Prognostic estimates also depend on the population and care practices from which they were derived. (pubmed.ncbi.nlm.nih.gov)

The ICH Score, developed by J. Claude Hemphill III and colleagues in 2001 from patients presenting in 1997–1998, standardizes initial severity using five variables. Glasgow Coma Scale scores of 3–4 contribute two points, scores of 5–12 one point, and scores of 13–15 none. Age of at least 80 years, hematoma volume of at least 30 mL, infratentorial origin, and intraventricular hemorrhage each contribute one additional point, giving a theoretical range of 0–6. The original endpoint was 30-day mortality, not later functional independence. All 26 patients scoring zero survived in the derivation cohort; in an independent validation cohort, four of 41 patients scoring zero died within 30 days. This difference illustrates why a score category cannot be equated with a fixed outcome probability in every setting. (pubmed.ncbi.nlm.nih.gov)

The modified Rankin Scale (mRS), by contrast, records global disability and dependence rather than admission hemorrhage severity. Its categories are 0, no symptoms; 1, symptoms without significant disability; 2, slight disability but independence in personal affairs; 3, need for some help while retaining unassisted walking; 4, inability to walk or attend to bodily needs without assistance; 5, severe dependence requiring constant care; and 6, death. Assessment timing must accompany the result: discharge, 30-day, and 90-day scores are not interchangeable. Ninety-day mRS is frequently used in ICH studies, and scores of 0–2 are commonly grouped as functional independence. Because the scale is ordinal, a one-category difference does not represent an equal-sized change at every level. (cde.nlm.nih.gov)

Severity scores alone do not establish treatment futility. Early withdrawal of life-sustaining treatment can influence the mortality that prognostic models subsequently predict, creating a potential self-fulfilling association between an unfavorable prediction and death. The 2022 AHA/ASA guideline states that baseline severity scales should not be the sole basis for limiting life-sustaining treatment. For patients without previously documented treatment limitations, it considers postponement of new do-not-attempt-resuscitation (DNAR) orders or withdrawal of support until at least the second full hospital day reasonable. A DNAR order concerns resuscitation following cardiopulmonary arrest; it does not, by itself, authorize withholding other medical or surgical interventions. Those decisions require separate consideration of the patient’s expressed wishes and shared discussion with the patient or surrogate. (pubmed.ncbi.nlm.nih.gov)

Rehabilitation and long-term consequences

Multidisciplinary rehabilitation assesses both neurological impairments and their consequences for everyday life. The team typically includes rehabilitation physicians, nurses, physical and occupational therapists, speech-language therapists, and professionals addressing psychological and social needs. Mobility assessment covers strength, coordination, balance, transfers, and walking; physical therapy links these findings to functional movement tasks. Occupational assessment examines dressing, bathing, eating, and more complex activities such as preparing meals and managing medications. Speech-language assessment distinguishes impaired language from impaired speech production and evaluates comprehension, reading, writing, and practical communication. Swallowing ability, attention, memory, executive function, and mood are assessed separately because each can affect participation and the safety of discharge. Findings are interpreted alongside pre-hemorrhage function, medical conditions, the home environment, and available caregiver support. (ahajournals.org)

Persistent consequences can include limb weakness, impaired coordination, abnormal muscle tone, language impairment, cognitive difficulties, and dependence in personal care. Their functional significance extends beyond walking or performing isolated movements: communication difficulties can restrict relationships and social participation, while cognitive impairment can interfere with independent daily activities. Depression and anxiety are additional targets of follow-up rather than inevitable consequences of hemorrhage. Assessment of quality of life therefore includes the survivor’s priorities and participation in family, community, and occupational roles, not simply physical independence. The 2022 American Heart Association/American Stroke Association guideline distinguishes treatment of depression from pharmacological enhancement of recovery: trials did not establish a functional benefit from fluoxetine in patients without depression, although they found less new depression and more fractures. (ahajournals.org)

Rehabilitation combines recovery associated with reorganization of brain function with compensatory strategies that enable activities despite residual impairment. Examples include repeated practice of grasping, transfers, and self-care, alongside adaptations to the way a task is performed. Timing and intensity depend on medical stability, impairment severity, and ability to participate. The 2022 guideline identifies stretching and functional task training beginning 24–48 hours after moderate hemorrhage as potentially beneficial, but distinguishes these activities from intense, frequent out-of-bed mobilization during the first 24 hours, which was associated with worse recovery. These findings do not establish a universal rehabilitation timetable or a fixed duration of treatment. (professional.heart.org)

Recovery can continue beyond the three-month endpoint commonly used in hemorrhage trials. In a 2024 study, Andrea Morotti and colleagues retrospectively examined 703 patients with supratentorial hemorrhage treated at six European stroke centers. Among the 458 patients alive at three months, 131 (28.6%) improved on the modified Rankin Scale by 12 months, 249 (54.4%) remained stable, and 78 (17.0%) worsened. Older age and pre-existing disability were associated with lower odds of improvement. Missing follow-up data and the absence of detailed information on rehabilitation intensity limit interpretation: the results describe heterogeneous recovery trajectories, not the effect of a particular rehabilitation program or an individual deadline for recovery. (pmc.ncbi.nlm.nih.gov)

Discharge planning begins during inpatient care and connects functional assessment with the next care setting. For selected patients with mild-to-moderate impairment, early supported discharge means continuing coordinated rehabilitation at home with appropriate professional expertise; it is not simply shortening hospitalization without replacement services. Other patients require inpatient rehabilitation or continuing assistance. Caregiver preparation includes practical training in transfers, assisted walking, daily activities, and assistive-device use, together with education about cognitive, emotional, and behavioral changes. Plans also identify follow-up services, community resources, and whom to contact when difficulties arise after returning home. The caregiver’s capacity and support needs form part of this assessment, because family members often assume substantial responsibilities after discharge. (professional.heart.org)

Recurrence and prevention

Recurrent hemorrhage risk depends on the vascular disease that remains after the initial hematoma resolves. The 2022 American Heart Association/American Stroke Association guideline estimated recurrence at approximately 1.2–3% per year across unselected survivors, with the highest event rate in the first year. A pooled analysis cited in the guideline found annual recurrence of 7.4% after hemorrhage attributed to cerebral amyloid angiopathy (CAA), compared with 1.1% after non-CAA hemorrhage. CAA affects cortical and leptomeningeal vessels and characteristically produces lobar hemorrhages; hypertension-associated arteriolosclerosis more often produces deep hemorrhages. These distributions identify disease patterns rather than determine an individual outcome. Multiple strictly lobar microbleeds and disseminated cortical superficial siderosis—residual blood products along the cortical surface—on magnetic resonance imaging indicate a hemorrhage-prone vascular substrate and are associated with increased recurrence risk. (ahajournals.org)

Sustained control of blood pressure is central to both prevention of a first hemorrhage and secondary prevention after ICH. In a cohort of 1,145 survivors followed for a median of 36.8 months, inadequate blood-pressure control was associated with higher recurrence in both lobar and nonlobar disease, with hazard ratios of 3.53 and 4.23, respectively. Randomized evidence also supports blood-pressure lowering: in the prior-ICH subgroup of PROGRESS, published initially in 2001, recurrent stroke or transient ischemic attack occurred in 10.5% receiving a perindopril-based regimen, compared with 17.7% receiving placebo over approximately 3.9 years. The 2022 American guideline considered a long-term pressure of 130/80 mm Hg reasonable, while the 2025 European Stroke Organisation/European Association of Neurosurgical Societies guideline suggested a target of ≤130/80 mm Hg. These are post-acute prevention targets, not emergency treatment thresholds; the optimal target and drug regimen have not been established by dedicated post-acute ICH trials. (ahajournals.org)

Exposure-related prevention addresses heavy alcohol consumption, tobacco smoking, and sympathomimetic drugs such as cocaine and methamphetamine. Heavy drinking can cause intermittent blood-pressure elevation, while observational studies associate it with ICH risk, although confounding by other exposures remains possible. The 2022 guideline includes smoking and recreational-drug cessation, reduced alcohol consumption, physical activity, and dietary measures in primary and secondary prevention. Evidence that these measures improve blood pressure and broader vascular health is stronger than direct randomized evidence that each independently prevents recurrent ICH: the lifestyle-intervention studies reviewed by the guideline largely involved other stroke populations rather than ICH survivors specifically. (ahajournals.org)

Antithrombotic decisions distinguish antiplatelet treatment for occlusive arterial disease from anticoagulation for cardioembolic or other thromboembolic indications. The RESTART randomized trial, reported in 2019, enrolled 537 survivors whose hemorrhage had occurred while taking antithrombotic medication. Randomization occurred a median of 76 days after onset. Over a median two years, recurrent ICH occurred in 12 participants assigned to start antiplatelet therapy and 23 assigned to avoid it, approximately 4% versus 9%; the adjusted hazard ratio was 0.51, with a 95% confidence interval of 0.25–1.03. This did not establish that antiplatelets prevent hemorrhage, but provided reassurance about restarting them for an established vascular indication in the population studied. The 2025 European guideline does not support initiating antiplatelet treatment without such an indication. (pubmed.ncbi.nlm.nih.gov)

Anticoagulation presents a different balance, particularly in atrial fibrillation. PRESTIGE-AF, reported in 2025 by Roland Veltkamp and colleagues, randomized 319 ICH survivors to a direct oral anticoagulant or no anticoagulation. Over a median 1.4 years, ischemic-stroke event rates were 0.83 versus 8.60 per 100 patient-years, but intracerebral-hemorrhage rates were 5.00 versus 0.82 per 100 patient-years. Thus, fewer ischemic strokes accompanied more recurrent hemorrhages. Clinical assessment weighs the thromboembolic indication and previous embolic events against hemorrhage location, CAA markers, blood-pressure control, and other bleeding risks. Mechanical heart valves represent a different thromboembolic context from nonvalvular atrial fibrillation; findings from one population cannot simply be transferred to the other. (sciencedirect.com)

Clinical research and standards of care

The 2022 American Heart Association/American Stroke Association (AHA/ASA) guideline, chaired by Steven M. Greenberg, replaced the association’s 2015 guideline and applied evidence-based medicine across diagnosis, emergency care, inpatient treatment, recovery, and secondary prevention. Its scope was adult spontaneous intracerebral hemorrhage without head trauma or a demonstrated structural cause, such as a vascular malformation or hemorrhage-prone tumor; investigation for these causes was nevertheless included. The principal literature search covered October 2020–March 2021, with selected consequential studies incorporated through November 2021. Recommendations received separate classifications for strength of recommendation and level of supporting evidence, distinguishing established practices from interventions supported by limited data. (ahajournals.org)

The 2025 European Stroke Organisation/European Association of Neurosurgical Societies (ESO/EANS) guideline updated the ESO’s 2014 guidance through a joint neurological and neurosurgical assessment of spontaneous ICH. The working group, coordinated by Thorsten Steiner and Rustam Al-Shahi Salman, addressed 37 Population, Intervention, Comparator, Outcome questions using the GRADE framework. It prioritized randomized trials and distinguished graded evidence-based recommendations from expert-consensus statements where evidence was insufficient. Its coverage included diagnostic investigation, acute medical and surgical interventions, complications, and secondary prevention. The two guidelines are not interchangeable: their evidence-review dates, question structures, and grading systems differ, and the European document incorporates trials published after the American guideline’s evidence cutoff. (pmc.ncbi.nlm.nih.gov)

ICH randomized controlled trials define eligibility by factors such as hemorrhage location, volume, neurological severity, and treatment window. Functional endpoints may measure the distribution of modified Rankin Scale scores or a utility-weighted score reflecting patients’ valuation of different outcomes. In the 2024 ENRICH trial, Gustavo Pradilla and colleagues randomized 300 patients with lobar or anterior basal ganglia hemorrhages of 30–80 mL to early minimally invasive evacuation plus medical management or medical management alone. The primary endpoint was the utility-weighted Rankin score at 180 days: mean scores were 0.458 and 0.374, respectively, a difference of 0.084 with a 95% Bayesian credible interval of 0.005–0.163. The prespecified adaptive design stopped anterior basal ganglia recruitment for futility; the observed benefit principally reflected the lobar population. These selection rules limit extrapolation to other hemorrhage locations or severities. (nejm.org)

Mortality and adverse-event reporting provide information distinct from functional or imaging endpoints. The 2024 ANNEXA-I trial randomized 530 patients with factor Xa inhibitor–associated ICH to andexanet or usual care. Hemostatic efficacy—a composite incorporating hematoma expansion, neurological deterioration, and rescue treatment—was assessed in 452 patients and occurred in 67.0% versus 53.1%. Safety analysis included all 530 participants: thrombotic events occurred in 10.3% versus 5.6%, including ischemic stroke in 6.5% versus 1.5%. There were no appreciable between-group differences in Rankin scores or death within 30 days. Reporting these outcomes separately prevents improved bleeding control from being treated as equivalent to demonstrated functional benefit or improved survival. (nejm.org)

The 2024 AHA/ASA performance and quality measures, developed by a writing group led by Ilana M. Ruff, translated selected 2022 recommendations into measurable care processes. They updated the 2018 measures and comprised 15 performance measures and five quality measures spanning prehospital, hospital, and posthospital care. Eligible cases included acute spontaneous, nontraumatic ICH without a known structural or macrovascular cause, whether present on admission or developing during hospitalization. Chronic or resolving hemorrhage, isolated MRI microhemorrhages, and hemorrhagic transformation of ischemic stroke were excluded. These measures assess a defined subset of care rather than every appropriate intervention. (professional.heart.org)

Process measurement specifies an eligible denominator, a numerator meeting the criterion, a measurement interval, and documented exclusions or exceptions. The radiology-reported diagnosis anchors timing for processes including anticoagulant reversal, specialized-unit admission, swallowing assessment, and transfer for ventricular drainage. For example, the specialized-unit measure uses a 24-hour interval after imaging diagnosis, with documented exceptions for medical circumstances, patient preferences, or system barriers. Attribution to particular facilities and care settings supports coordinated health-system evaluation across emergency services, admitting hospitals, referral centers, and outpatient care. Quality measures additionally permit piloting where data collection or attribution remains difficult; the blood-pressure-target measure was included partly to support such evaluation before possible conversion into a performance measure. (ahajournals.org)

References

1. An Updated Definition of Stroke for the 21st Century: A Statement for Healthcare Professionals From the American Heart Association/American Stroke Associationpmc.ncbi.nlm.nih.gov
2. A Causal Classification System for Intracerebral Hemorrhage Subtypespmc.ncbi.nlm.nih.gov
3. A Standardized Anatomical Classification of Intracranial Hemorrhagepmc.ncbi.nlm.nih.gov
4. Global, regional, and national burden of intracerebral hemorrhage and its attributable risk factors from 1990 to 2021: results from the 2021 Global Burden of Disease Studypmc.ncbi.nlm.nih.gov
5. Incidence, case fatality, and functional outcome of intracerebral haemorrhage over time, according to age, sex, and ethnic origin: a systematic review and meta-analysispubmed.ncbi.nlm.nih.gov
6. Incidence, case fatality, and functional outcome of intracerebral haemorrhage, according to age, sex, and country income level: a systematic review and meta-analysispmc.ncbi.nlm.nih.gov
7. Cavernous malformations of the central nervous system: An international consensus statementpmc.ncbi.nlm.nih.gov
8. Clinical and imaging manifestations of intracerebral hemorrhage in brain tumors and metastatic lesions: a comprehensive overviewpmc.ncbi.nlm.nih.gov
9. Comparison of the efficacy and safety of new oral anticoagulants with warfarin in patients with atrial fibrillation: a meta-analysis of randomised trialspubmed.ncbi.nlm.nih.gov
10. Clinical characteristics and outcomes of methamphetamine-associated intracerebral hemorrhagepmc.ncbi.nlm.nih.gov
11. Early hemorrhage growth in patients with intracerebral hemorrhagemayoclinic.elsevierpure.com
12. Hypoperfusion without ischemia surrounding acute intracerebral hemorrhagepubmed.ncbi.nlm.nih.gov
13. Mechanisms of edema formation after intracerebral hemorrhage: effects of thrombin on cerebral blood flow, blood-brain barrier permeability, and cell survival in a rat modelpubmed.ncbi.nlm.nih.gov
14. Brain edema after experimental intracerebral hemorrhage: role of hemoglobin degradation productspubmed.ncbi.nlm.nih.gov
15. Rapid neuroinflammatory changes in human acute intracerebral hemorrhagepmc.ncbi.nlm.nih.gov
16. Clearance of Neutrophils From ICH-Affected Brain by Macrophages Is Beneficial and Is Assisted by Lactoferrin and CD91pubmed.ncbi.nlm.nih.gov
17. Natural history of perihematomal edema following intracerebral hemorrhage measured by serial magnetic resonance imagingpmc.ncbi.nlm.nih.gov
18. Natural history of perihematomal edema after intracerebral hemorrhage measured by serial magnetic resonance imagingpubmed.ncbi.nlm.nih.gov
19. Influence of site on course of intracerebral hemorrhagepubmed.ncbi.nlm.nih.gov
20. Clinical Presentation of Intracerebral Hemorrhagemichigan.gov
21. Acute Spontaneous Lobar Cerebral Hemorrhages Present a Different Clinical Profile and a More Severe Early Prognosis than Deep Subcortical Intracerebral Hemorrhages—A Hospital-Based Stroke Registry Studypmc.ncbi.nlm.nih.gov
22. NIH Common Data Elements (CDE) Repositorycde.nlm.nih.gov
23. NIH Stroke Scaleninds.nih.gov
24. FAQ - Glasgow Coma Scaleglasgowcomascale.org
25. Intracerebral Hemorrhage Protocolneurocriticalcare.org
26. The ABCs of measuring intracerebral hemorrhage volumespubmed.ncbi.nlm.nih.gov
27. Reliability of hematoma volume measurement at local sites in a multicenter acute intracerebral hemorrhage clinical trialpmc.ncbi.nlm.nih.gov
28. European Stroke Organisation (ESO) and European Association of Neurosurgical Societies (EANS) guideline on stroke due to spontaneous intracerebral haemorrhagepmc.ncbi.nlm.nih.gov
29. The third Intensive Care Bundle with Blood Pressure Reduction in Acute Cerebral Haemorrhage Trial (INTERACT3): an international, stepped wedge cluster randomised controlled trialpmc.ncbi.nlm.nih.gov
30. Update on the Safety of Andexxafda.gov
31. Minimally invasive surgery with thrombolysis in intracerebral haemorrhage evacuation (MISTIE III): a randomised, controlled, open-label phase 3 trial with blinded endpointpmc.ncbi.nlm.nih.gov
32. Minimally invasive surgery with thrombolysis in intracerebral haemorrhage evacuation (MISTIE III): a randomised, controlled, open-label phase 3 trial with blinded endpointdiposit.ub.edu
33. Ten-Year Temporal Trends in Medical Complications After Acute Intracerebral Hemorrhage in the United Statespubmed.ncbi.nlm.nih.gov
34. The ICH score: a simple, reliable grading scale for intracerebral hemorrhage.pubmed.ncbi.nlm.nih.gov
35. External validation of the ICH scorepubmed.ncbi.nlm.nih.gov
36. Modified Rankin Scale (mRS) Scorecde.nlm.nih.gov
37. Functional Outcomes Expressed as the Distribution of mRS Scoresncbi.nlm.nih.gov
38. Using 30-day modified rankin scale score to predict 90-day score in patients with intracranial hemorrhage: Derivation and validation of prediction model.pubmed.ncbi.nlm.nih.gov
39. Modified ICH score was superior to original ICH score for assessment of 30-day mortality and good outcome of non-traumatic intracerebral hemorrhage.pubmed.ncbi.nlm.nih.gov
40. Quantifying the Value of Stroke Disability Outcomes: WHO Global Burden of Disease Project Disability Weights for Each Level of the Modified Rankin Scalepmc.ncbi.nlm.nih.gov
41. Withdrawal of support in intracerebral hemorrhage may lead to self-fulfilling prophecies.pubmed.ncbi.nlm.nih.gov
42. Functional outcome improvement from 3 to 12 months after intracerebral hemorrhagepmc.ncbi.nlm.nih.gov
43. Effects of antiplatelet therapy after stroke due to intracerebral haemorrhage (RESTART): a randomised, open-label trialpubmed.ncbi.nlm.nih.gov
44. Direct oral anticoagulants versus no anticoagulation for the prevention of stroke in survivors of intracerebral haemorrhage with atrial fibrillation (PRESTIGE-AF): a multicentre, open-label, randomised, phase 3 trialpubmed.ncbi.nlm.nih.gov
45. Andexanet for Factor Xa Inhibitor–Associated Acute Intracerebral Hemorrhagebroomedocs.com