Subarachnoid hemorrhage

Subarachnoid hemorrhage (SAH) is bleeding into the subarachnoid space, the normally cerebrospinal-fluid-filled compartment between the arachnoid membrane and pia mater surrounding the brain. It can follow head trauma or occur spontaneously, notably through rupture of an intracranial aneurysm. Spontaneous SAH is a form of hemorrhagic stroke, anatomically distinct from intracerebral hemorrhage, in which bleeding occurs within brain tissue. (medlineplus.gov)

The condition is a neurological emergency because the initial hemorrhage and subsequent complications can cause death or lasting disability. A sudden, severe headache is characteristic, although presentation can include impaired consciousness, seizures, or focal neurological deficits. Investigation establishes both the presence of blood and its source; care depends on the cause and may involve securing a ruptured aneurysm, neurocritical care, and rehabilitation. Important distinctions include traumatic versus spontaneous hemorrhage, aneurysmal versus nonaneurysmal bleeding, and immediate injury versus delayed complications. (nice.org.uk)

Anatomical definition and classification

Subarachnoid hemorrhage (SAH) is bleeding into the space between the arachnoid mater and pia mater, two of the three meningeal membranes surrounding the brain and spinal cord. From the skull inward, the coverings are the dura mater, arachnoid mater, and pia mater. The dura forms the tough outer covering; the arachnoid lies beneath it; and the pia closely follows the brain’s surface, including its folds. The subarachnoid space therefore lies outside the brain tissue but inside the arachnoid, rather than directly against the skull. It normally contains cerebrospinal fluid (CSF), traversed by delicate arachnoid trabeculae. SAH denotes blood entering this normally fluid-filled compartment. (ninds.nih.gov)

The subarachnoid space is not uniformly wide. The pia follows the grooves of the cerebral cortex—the cortical sulci—whereas the arachnoid bridges over them, leaving CSF-filled spaces beneath it. Where these membranes are more widely separated, they form subarachnoid cisterns. Basal cisterns include spaces around the optic chiasm, between the cerebral peduncles, and in front of the pons; the cisterna magna lies between the cerebellum and medulla. These cisterns and sulcal spaces are subdivisions of the subarachnoid compartment, not separate kinds of hemorrhage: blood confined to cortical sulci and blood filling cisterns at the brain’s base both meet the anatomical definition of SAH. (oac22.hsc.uth.tmc.edu)

The location of blood distinguishes SAH from other intracranial hemorrhages. Intracerebral hemorrhage, also called intraparenchymal hemorrhage, occurs within brain tissue. Intraventricular hemorrhage occupies the brain’s internal, CSF-containing ventricular cavities. Subdural hemorrhage collects between the dura and arachnoid, while cranial epidural hemorrhage collects between the dura and skull. These labels are not mutually exclusive diagnoses in an individual patient: a single injury can produce hemorrhage in several compartments. Traumatic SAH commonly accompanies other intracranial bleeding, and aneurysmal SAH may coexist with intraparenchymal or intraventricular blood. The presence of subarachnoid blood describes its location without establishing the original bleeding vessel. (ninds.nih.gov)

Etiologically, SAH is divided into traumatic and spontaneous, or nontraumatic, hemorrhage. Traumatic SAH occurs in association with head injury, such as a fall or motor-vehicle collision, and belongs to the spectrum of traumatic brain injury. Spontaneous SAH is further divided into aneurysmal hemorrhage, caused by rupture of an intracranial aneurysm, and nonaneurysmal hemorrhage. The latter is a heterogeneous category that includes bleeding from other vascular abnormalities or disorders affecting coagulation, as well as cases in which no source is established. “Spontaneous” thus means nontraumatic, not necessarily unexplained; “nonaneurysmal” does not identify a single mechanism. (medlineplus.gov)

This etiological distinction also defines the applicability of research. Spontaneous SAH is included among hemorrhagic forms of stroke, whereas traumatic SAH is classified with traumatic injury. The 2023 American Heart Association/American Stroke Association guideline specifically addresses aneurysmal SAH. Its evidence concerning treatment of the ruptured aneurysm and prevention of subsequent complications does not automatically establish equivalent benefit in traumatic SAH, where the initiating injury and accompanying lesions differ. Anatomically identical placement of blood therefore does not make the underlying conditions interchangeable populations for clinical trials or treatment recommendations. (pmc.ncbi.nlm.nih.gov)

Causes and predisposing factors

Rupture of an intracranial aneurysm is a major cause of spontaneous subarachnoid hemorrhage. The usual lesion is a saccular, or “berry,” aneurysm: a localized outpouching of an arterial wall, commonly situated at a branching point in the circle of Willis at the base of the brain. When its wall ruptures, arterial blood escapes into the surrounding subarachnoid space. An aneurysm’s presence is not equivalent to hemorrhage: unruptured aneurysms may remain asymptomatic and be discovered incidentally during imaging for another condition. Formation of the lesion and subsequent failure of its wall are therefore distinct events. (meshb.nlm.nih.gov)

Aneurysm size, location, and shape influence rupture risk, but none provides an absolute boundary between lesions that will and will not bleed. In the 2012 UCAS Japan study, Akio Morita and colleagues found that aneurysms with a “daughter sac”—an irregular protrusion from the aneurysm wall—had a higher rupture hazard than those without one, with a hazard ratio of 1.63 (95% confidence interval, 1.08–2.48). A 2014 pooled analysis by Jacoba Greving and colleagues included 8,382 participants from six prospective cohorts and recorded 230 ruptures during 29,166 person-years. Predictors included aneurysm size and location, age, hypertension, previous subarachnoid hemorrhage, and geographical population. Its overall five-year rupture estimate of 3.4% described those study populations, not a uniform risk applicable to every incidentally detected aneurysm. (pubmed.ncbi.nlm.nih.gov)

Head trauma provides a separate mechanism, damaging blood vessels on or near the brain surface rather than necessarily rupturing a pre-existing aneurysm. Falls and motor-vehicle crashes are documented settings, and traumatic subarachnoid bleeding commonly accompanies other intracranial hemorrhages. Nontraumatic causes also include arteriovenous malformations, abnormal vascular tangles that may bleed, and cerebral venous thrombosis. Bleeding disorders and anticoagulant exposure, including warfarin, can contribute by impairing normal clot formation. These circumstances should not be conflated with aneurysm formation: a hemorrhage associated with disordered coagulation need not originate from an aneurysm. Less common documented causes include brain tumors and sickle cell disease; some hemorrhages remain without an identified cause. (medlineplus.gov)

Cigarette smoking and hypertension are established associations with subarachnoid hemorrhage and are important components of its vascular risk profile. Their effects may overlap rather than act independently. A 2012 prospective Norwegian study using the Nord-Trøndelag Health Study and Tromsø Study cohorts found evidence of interaction between current smoking and hypertension on an additive risk scale: the combined excess risk was greater than the sum of the separate excess risks. These population-level associations do not establish that a particular exposure caused an individual hemorrhage, nor does their absence exclude aneurysm rupture. Cocaine and methamphetamine use are also recognized predisposing exposures. (pubmed.ncbi.nlm.nih.gov)

Familial susceptibility is another component. A family history of intracranial aneurysms is associated with increased risk, and a family history of aneurysmal rupture may be relevant to rupture risk in someone who already has an aneurysm. Autosomal dominant polycystic kidney disease is an inherited condition associated with intracranial aneurysms; connective-tissue disorders and fibromuscular dysplasia are additional recognized associations. These relationships identify predisposition, not inevitable bleeding: inherited susceptibility, acquired vascular exposures, and the characteristics of the aneurysm contribute different information about risk. (ninds.nih.gov)

Epidemiology and population burden

Population estimates depend on which hemorrhages are counted. Traumatic SAH belongs to the epidemiology of traumatic brain injury and is excluded from nontraumatic stroke estimates. Nontraumatic SAH includes both bleeding from a ruptured intracranial aneurysm and nonaneurysmal hemorrhage. These categories are not interchangeable: the Global Burden of Disease (GBD) analysis encompasses nontraumatic SAH broadly, whereas studies of aneurysmal SAH address a narrower population. The GBD investigators specifically identified the inability to distinguish aneurysmal from nonaneurysmal cases consistently as a limitation of their estimates. (jamanetwork.com)

The GBD 2021 analysis, published in 2025, estimated worldwide age-standardized nontraumatic SAH incidence at 8.3 per 100,000 people in 2021 (95% uncertainty interval, 7.3–9.5), representing approximately 700,000 new cases. It also estimated approximately 350,000 deaths and 10.6 million disability-adjusted life-years, a measure combining premature death with years lived with disability. Age-standardized incidence declined by 28.8% between 1990 and 2021, although absolute case numbers increased, especially in lower-sociodemographic-index regions. Thus, declining standardized rates did not imply a declining number of people affected. (jamanetwork.com)

For aneurysmal SAH, Nima Etminan and colleagues’ 2019 meta-analysis included 75 population-based studies from 32 countries, comprising 8,176 cases over approximately 67.7 million person-years. Its estimated worldwide crude incidence fell from 10.2 per 100,000 person-years in 1980 to 6.1 in 2010. Regional estimates for 2010 differed substantially: 6.3 in Europe, 6.9 in North America, and 28.0 in Japan. These are dated, study-derived estimates rather than present-day national surveillance figures. The analysis found that declining incidence paralleled reductions in population blood pressure and smoking prevalence, but these population-level associations do not establish the cause of an individual hemorrhage. (jamanetwork.com)

Age and sex patterns also depend on the case definition and analytical method. In the 2019 meta-analysis, incidence generally increased with age, and the female-to-male risk ratio was 1.3 (95% confidence interval, 0.98–1.7); the female excess was most apparent at older ages, particularly in Japan. By contrast, GBD 2021 estimates for all nontraumatic SAH showed slightly higher age-standardized incidence in males than females, 8.51 versus 8.17 per 100,000. These findings should not be treated as equivalent measurements of aneurysm rupture risk. Age standardization applies a common population age distribution, allowing comparisons that are less affected by differences in population aging. (jamanetwork.com)

Ascertainment can materially alter measured incidence. Hospital-only series omit people who die before admission, while incomplete access to computed tomography or postmortem examination can leave hemorrhages unrecognized. A Finnish nationwide study therefore linked hospital-discharge and cause-of-death registers to identify both hospitalized and sudden-death events during 1998–2017. Such registries count identifiable events within a defined population; GBD instead combines available observations with statistical modeling to estimate burden across countries, including data-poor settings. Differences in diagnostic coverage, inclusion of prehospital deaths, age standards, and modeled versus directly observed data prevent a simple comparison of the 2010 meta-analysis estimate with the 2021 GBD estimate as evidence of a subsequent rise. (pubmed.ncbi.nlm.nih.gov)

Mechanisms of brain injury

Major aneurysmal bleeding can abruptly increase intracranial pressure because blood enters a skull compartment with limited capacity to accommodate additional volume. Cerebral perfusion pressure is approximately the difference between mean arterial pressure and intracranial pressure, expressed as $CPP \approx MAP-ICP$. A sufficiently large pressure surge can therefore sharply reduce, or briefly interrupt, cerebral blood flow, depriving neurons of oxygen and glucose and causing sudden loss of consciousness. Impaired autoregulation—the ability of cerebral vessels to stabilize blood flow as pressure changes—can prolong vulnerability after the initial surge subsides. “Early brain injury” generally denotes the interacting disturbances occurring within the first 72 hours, rather than a single lesion. In a 2016 cohort study, 590 of 1,460 patients, or 40.4%, had lost consciousness at hemorrhage onset; this was associated with more extensive bleeding and more frequent global cerebral swelling. (pmc.ncbi.nlm.nih.gov)

Early edema has overlapping cellular and vascular mechanisms. Inadequate energy supply compromises membrane ion pumps, allowing sodium and water to accumulate inside cells: cytotoxic edema. Injury to the blood–brain barrier, formed principally by specialized cerebral vascular endothelium and its supporting structures, permits fluid and plasma constituents to enter extracellular tissue, producing vasogenic edema. Both processes can increase tissue volume and further raise intracranial pressure, reinforcing the reduction in perfusion. Human permeability-imaging studies show that barrier dysfunction is not confined to already obvious infarcts: a 2019 study detected increased permeability in apparently normal brain tissue as early as 24 hours after hemorrhage, with some affected regions subsequently developing imaging abnormalities. These observations associate barrier injury with evolving tissue damage without establishing it as the sole cause. (pmc.ncbi.nlm.nih.gov)

Blood within the cerebrospinal-fluid compartment also initiates chemical and inflammatory injury. Breakdown of erythrocytes releases hemoglobin and heme-containing products, which activate resident microglia and inflammatory signaling pathways. Experimental studies implicate Toll-like receptor 4 and interleukin-1 signaling in this response; cytokine release and oxidative injury can damage neurons and vascular barriers. These pathways are not uniformly destructive: microglia also remove blood products. A 2015 study combining patient cerebrospinal-fluid measurements with mouse experiments identified microglial heme oxygenase-1 as an important component of blood clearance and protection against injury. Thus, inflammation encompasses both damaging responses and clearance mechanisms, whose relative contributions change over time. (pmc.ncbi.nlm.nih.gov)

Accumulated blood can impede cerebrospinal-fluid circulation and clearance, producing hydrocephalus and an additional pressure load. Clots may obstruct ventricular pathways, while blood and subsequent inflammatory changes can interfere with outflow through the subarachnoid compartment. Altered secretion may contribute alongside impaired drainage: a 2022 study found inflammatory abnormalities in cerebrospinal fluid from patients with subarachnoid hemorrhage and demonstrated increased fluid secretion and altered choroid-plexus transporter activity in an experimental intraventricular-hemorrhage model. The secretion findings establish a plausible additional mechanism, not proof that hypersecretion explains hydrocephalus in every human case. (pubmed.ncbi.nlm.nih.gov)

Delayed cerebral ischemia develops after the immediate injury, usually during the following two weeks, when tissue oxygen delivery becomes insufficient and may progress to infarction. It is not synonymous with angiographic vasospasm, which describes narrowing of visible arteries. The 2023 American Heart Association/American Stroke Association guideline identifies several potentially interacting mechanisms: large- and small-vessel constriction, microvascular thrombosis, impaired autoregulation, barrier disruption, inflammation, and spreading depolarizations. The latter are waves of near-complete neuronal and glial depolarization that impose a substantial energy demand; in vulnerable tissue, an inadequate or constrictive vascular response can deepen the oxygen deficit. In Jens P. Dreier and colleagues’ 2006 multicenter study, cortical recordings detected 298 spreading depolarizations in 13 of 18 surgically treated patients, and delayed neurological deficits were temporally associated with recurrent clusters. This small, selected cohort demonstrated a human physiological association rather than a universal explanation for delayed ischemia. (pmc.ncbi.nlm.nih.gov)

Clinical presentation and differential diagnosis

The characteristic presentation is an abruptly developing headache that rapidly reaches maximum intensity—a thunderclap headache. The defining feature is its time course, not simply a description such as “the worst headache of my life.” NICE’s 2022 guideline describes a typical peak within 1–5 minutes; the International Headache Society’s criteria for primary thunderclap headache use a stricter threshold of less than 1 minute. Associated symptoms include neck pain or stiffness, photophobia, nausea, and vomiting. Limited or painful neck flexion may be evident on physical examination, but the complete symptom cluster is not necessary for suspicion. (nice.org.uk)

Presentation ranges from headache in an otherwise alert person to collapse, reduced consciousness, seizures, or focal neurological deficits. These manifestations create different recognition problems: preserved alertness and an apparently normal neurological examination can give false reassurance, whereas impaired consciousness or communication can prevent a reliable account of headache. A witnessed description of the onset can therefore be important. NICE specifically identifies nonspecific presentations as being at greater risk of missed diagnosis. Neither an absent headache report in an uncommunicative patient nor the absence of focal findings excludes hemorrhage. (nice.org.uk)

Migraine is an important competing explanation because it can also produce severe pain, nausea, vomiting, and light sensitivity. Cough, sexual activity, and exertion can accompany thunderclap presentations without subarachnoid bleeding, so the circumstances of onset do not establish the cause. Conversely, a previous primary headache disorder does not exclude a new vascular headache: the International Headache Society classifies a new headache attributable to a demonstrated vascular disorder as secondary even when it resembles migraine. Primary thunderclap headache is a diagnosis of exclusion rather than a label established by pain intensity or a normal initial examination. (nice.org.uk)

Other vascular explanations include intracerebral hemorrhage, cerebral venous thrombosis, cervical or intracranial arterial dissection, and reversible cerebral vasoconstriction syndrome (RCVS); pituitary apoplexy is another important cause of abrupt severe headache. Repeated thunderclap attacks over 1–2 weeks are characteristic of RCVS, although this pattern alone is not diagnostic. Dissection may produce head or neck pain with ischemic or local neurological signs, such as a painful Horner syndrome. These overlapping presentations mean that thunderclap onset identifies a clinical syndrome, not a specific bleeding compartment or vascular lesion. (ichd-3.org)

Infectious meningitis overlaps through headache, photophobia, vomiting, neck stiffness, altered cognition, and sometimes seizures or focal deficits. NICE’s 2024 meningitis guideline identifies fever, headache, neck stiffness, and altered consciousness or cognition as a red-flag combination for bacterial meningitis, while noting that individual features may be absent or difficult to recognize. Thus, meningeal symptoms do not distinguish infection from hemorrhage by themselves. Clinical history and examination establish the differential diagnosis; confirmation of subarachnoid bleeding depends on diagnostic investigations rather than headache severity alone. (nice.org.uk)

Diagnostic investigations and identification of the source

Noncontrast head computed tomography (CT) is the initial imaging investigation for suspected acute subarachnoid hemorrhage. Fresh blood appears hyperdense—brighter than surrounding cerebrospinal fluid—and follows the contours of the basal cisterns, Sylvian fissures, interhemispheric fissure, or cortical sulci. Distribution and extent are recorded separately from the presence of bleeding: a localized sulcal deposit and extensive cisternal hemorrhage are both subarachnoid blood, but provide different clues to its source. Noncontrast CT establishes whether visible hemorrhage is present; it does not by itself demonstrate the responsible vascular lesion. (nice.org.uk)

CT sensitivity depends on the interval from symptom onset, acquisition quality, and interpretation. As blood disperses and its components break down, its contrast with cerebrospinal fluid decreases. In Jeffrey Perry and colleagues’ 2011 prospective study, CT performed within six hours identified all 121 hemorrhages among 953 patients scanned during that interval: sensitivity was 100%, with a 95% confidence interval of 97.0–100%. This finding concerned a selected population of alert patients with acute nontraumatic headache, modern scanners, and interpretation by qualified radiologists; it does not establish perfect sensitivity under all circumstances. The 2022 NICE guideline consequently distinguishes an early negative scan reported by a radiologist from a negative scan obtained more than six hours after onset, when false-negative results become more likely. (bmj.com)

Lumbar puncture can provide evidence of hemorrhage after selected negative CT examinations by sampling cerebrospinal fluid (CSF). Analysis includes red-cell counts and examination for xanthochromia, the discoloration associated with blood-breakdown products. Bilirubin forms after erythrocyte breakdown within the CSF compartment, so testing too early may miss it. NICE specifies considering lumbar puncture after a negative CT obtained more than six hours after symptom onset and allowing at least 12 hours before sampling for spectrophotometric bilirubin analysis; it does not routinely recommend puncture after an appropriately reported negative CT within six hours. These are timing-dependent recommendations rather than a universal requirement following every negative scan. (nice.org.uk)

Interpretation is complicated by a traumatic tap, in which the needle introduces blood during sampling. Red cells in CSF therefore do not necessarily indicate intracranial hemorrhage, and declining counts across successive collection tubes do not reliably exclude it. A traumatic puncture and genuine hemorrhage can also coexist. Perry and colleagues’ 2015 prospective study evaluated the combination of the final-tube red-cell count and xanthochromia rather than treating either visible blood or tube-to-tube clearing as decisive. Sampling time, laboratory technique, and the combined findings remain important when interpreting an abnormal specimen. (pmc.ncbi.nlm.nih.gov)

Once hemorrhage is established, CT angiography (CTA) addresses the separate question of its source. Intravenous iodinated contrast material outlines the cerebral arteries, allowing assessment of an aneurysm and its relationship to adjacent vessels. Finding an aneurysm does not alone prove that it ruptured: its location must be compatible with the blood distribution, especially when several aneurysms are present. Catheter digital-subtraction angiography (DSA) provides detailed vascular imaging but is invasive and carries procedural risks, including stroke and arterial-access hematoma. NICE identifies DSA as a further investigation when CTA does not reveal the cause and an aneurysm remains suspected; magnetic resonance angiography (MRA) is an alternative when DSA is contraindicated. (nice.org.uk)

Magnetic resonance imaging (MRI) has selected complementary uses, particularly in delayed presentations when CT blood conspicuity has diminished. Fluid-attenuated inversion recovery (FLAIR) suppresses normal CSF signal, making subarachnoid abnormalities more conspicuous, while susceptibility-sensitive sequences detect magnetic effects of blood products. A 2001 study of 41 patients investigated for suspected hemorrhage demonstrated the usefulness of FLAIR and gradient-echo T2*-weighted imaging, particularly in subacute cases. MRI assesses blood and associated brain abnormalities; MRA assesses vessels and should not be confused with direct confirmation that hemorrhage occurred. (jnnp.bmj.com)

Severity grading and prognostic assessment

Clinical grading describes the neurological condition at a particular assessment, whereas computed-tomography grading describes the distribution and burden of blood. In aneurysmal hemorrhage, the 2023 American Heart Association/American Stroke Association guideline recommends the Hunt–Hess or World Federation of Neurosurgical Societies (WFNS) grade for documenting initial severity and informing outcome prediction. These grades standardize communication, but their association with outcomes in patient groups does not make them deterministic forecasts for individuals. (ahajournals.org)

The Hunt–Hess scale, introduced by William E. Hunt and Robert M. Hess in 1968, combines headache, neck stiffness, level of consciousness, and neurological deficits. Grade I denotes absent or minimal symptoms, with at most mild headache and slight neck stiffness; grade II denotes moderate-to-severe headache and neck stiffness, without a neurological deficit other than a possible cranial-nerve palsy. Grade III indicates drowsiness, confusion, or a mild focal deficit; grade IV indicates stupor with moderate-to-severe weakness on one side, sometimes accompanied by abnormal extension responses; and grade V indicates deep coma with decerebrate rigidity or a moribund state. Its categories therefore incorporate both subjective symptoms and examination findings, rather than a numerical consciousness score alone. (nice.org.uk)

The WFNS scale uses the Glasgow Coma Scale (GCS) and the presence of a motor deficit. Grade I corresponds to GCS 15 without a motor deficit; grade II to GCS 13–14 without a deficit; and grade III to GCS 13–14 with a deficit. Grades IV and V correspond respectively to GCS 7–12 and 3–6, regardless of whether a motor deficit is present. For example, two patients with GCS 14 receive different grades if one has limb weakness and the other does not. Hunt–Hess or WFNS grades IV–V are commonly described as “poor-grade” or “high-grade” hemorrhage. Grading is time-dependent: early clinical status can change, and sedating medication can alter the examination. NICE specifically notes the sedating and pupillary effects of opioid analgesia when interpreting neurological findings. (build.fhir.org)

The Fisher scale classifies blood on computed tomography (CT), principally to stratify the risk of subsequent cerebral vasospasm. Developed by C. Miller Fisher, J. Philip Kistler, and J. Michael Davis in 1980, it assigns grade 1 when no blood is visible, grade 2 for diffuse thin subarachnoid blood, grade 3 for localized clots or blood layers at least 1 mm thick, and grade 4 for intraventricular or intracerebral blood with absent or only diffuse subarachnoid blood. It is not a simple ascending measure of vasospasm risk: grade 3 can carry greater risk than grade 4. The modified Fisher scale separates subarachnoid thickness from ventricular involvement: grade 0 denotes neither subarachnoid nor intraventricular blood; grades 1 and 2 denote thin subarachnoid blood without and with intraventricular blood, respectively; grades 3 and 4 denote thick subarachnoid blood without and with intraventricular blood. (pubmed.ncbi.nlm.nih.gov)

In Jennifer Frontera and colleagues’ 2006 analysis of 1,355 patients, 451 (33%) developed symptomatic vasospasm. Modified Fisher grade 4 was associated with a crude odds ratio of 2.2 compared with grades 0–1 (95% confidence interval, 1.6–3.1). This illustrates complication-risk stratification, not a percentage probability of death or permanent disability for a particular patient. NICE’s 2022 assessment found insufficiently consistent prognostic accuracy to support using any severity score alone to select management. Treatment assessment also incorporates neurological evolution, premorbid function, comorbidities, and other clinical findings; a high score alone does not establish treatment futility or exclude specialist care. (pubmed.ncbi.nlm.nih.gov)

Acute care and treatment of a ruptured aneurysm

Acute aneurysmal subarachnoid hemorrhage is managed in specialist centers with neurocritical care, vascular neurosurgery, and endovascular services. The 2023 American Heart Association/American Stroke Association (AHA/ASA) guideline associates experienced multidisciplinary teams and dedicated neurocritical-care units with lower mortality and better functional outcomes, and recommends timely transfer to such centers. Initial stabilization addresses airway protection, oxygenation, ventilation, and circulation; impaired consciousness or respiratory failure may require mechanical ventilation. Frequent neurological examinations establish whether consciousness or focal findings are changing, alongside close cardiovascular monitoring. These measures support the patient while the ruptured aneurysm is being assessed and secured. (professional.heart.org)

Before aneurysm occlusion, blood-pressure management balances the risk of further bleeding against inadequate cerebral perfusion. Frequent measurements and short-acting, titratable medication allow clinicians to limit severe hypertension, hypotension, and large pressure fluctuations. The 2023 AHA/ASA guideline does not establish a single blood-pressure target suitable for every patient. Anticoagulant-associated bleeding requires emergency reversal of anticoagulation, with the reversal strategy determined by the drug involved. These interventions accompany, rather than replace, definitive aneurysm treatment. Routine antifibrinolytic treatment has not improved functional outcomes and is not an alternative to securing the bleeding source. (uptodate.com)

Definitive treatment excludes the aneurysm from the circulation to limit rebleeding. The 2023 AHA/ASA guideline favors treatment as early as feasible, preferably within 24 hours of onset, with complete obliteration whenever possible. In microsurgical clipping, the surgeon exposes the aneurysm through an opening in the skull and places a small metal clip across its neck. The clip closes the connection between the aneurysm sac and the parent artery. In endovascular coiling, a catheter is guided through the arterial circulation into the aneurysm, where detachable platinum coils are deployed to obstruct blood flow into the sac. Clipping therefore closes the aneurysm from outside the vessel, whereas coiling treats it from within the vascular lumen. (professional.heart.org)

Selection is made jointly by specialists in endovascular treatment and open surgery, rather than by assuming that either technique is universally preferable. NICE’s 2022 guideline specifies consideration of the patient’s clinical condition, aneurysm characteristics, and the amount and distribution of subarachnoid blood. Its treatment pathway offers endovascular coiling when intervention is appropriate, with neurosurgical clipping when coiling is unsuitable. Relevant anatomical considerations include the aneurysm’s location, size, and shape, which affect access and the feasibility of closing it safely. The proposed treatment plan also accounts for situations in which an immediate procedure is inappropriate and reassessment is needed. (nice.org.uk)

Medical treatment continues alongside aneurysm occlusion. Early enteral nimodipine reduces delayed cerebral ischemia and improves functional outcomes after aneurysmal hemorrhage; it does not mechanically seal the aneurysm. Enteral administration includes oral treatment or delivery through a feeding tube when swallowing is unsuitable. Fluid management aims for euvolemia, an appropriate circulating volume, using assessment of intravascular volume and hemodynamic status rather than routinely administering excess fluid. Prophylactic hypervolemia and routine hemodynamic augmentation are avoided because they expose patients to treatment-related harm without established preventive benefit. This preventive approach is distinct from treatment of established delayed cerebral ischemia, for which blood-pressure augmentation may be used under specialist monitoring. (professional.heart.org)

Complications and neurocritical care

Recurrent hemorrhage is particularly dangerous before the ruptured aneurysm has been secured; the risk is highest during the first 24 hours after symptom onset. New neurological deterioration therefore prompts reassessment rather than automatic attribution to the original injury. Noncontrast computed tomography can identify fresh bleeding, ventricular enlargement, or another structural cause. Acute hydrocephalus develops when blood obstructs cerebrospinal-fluid circulation or impairs its absorption, producing ventricular enlargement and sometimes declining consciousness. Diagnosis combines clinical findings with comparison of current and previous imaging. Symptomatic deterioration may require urgent cerebrospinal-fluid drainage, commonly through an external ventricular drain; selected circumstances permit lumbar drainage. Persistent symptomatic hydrocephalus may require permanent diversion through a shunt, sometimes after temporary drainage has demonstrated likely benefit. These interventions relieve fluid accumulation and pressure; they do not treat seizures or exclude recurrent bleeding. (nice.org.uk)

Seizures require a separate diagnostic and treatment pathway. Continuous electroencephalography can detect nonconvulsive seizures when consciousness remains depressed or the neurological examination fluctuates without an obvious explanation. The 2023 American Heart Association/American Stroke Association guideline supports short-term antiseizure treatment for new-onset seizures, generally for seven days, rather than automatic indefinite therapy. Routine prophylactic medication is not recommended for every patient, although it may be considered with high-risk features such as associated intracerebral hemorrhage, cortical infarction, hydrocephalus, or severe aneurysmal hemorrhage. An acute symptomatic seizure is therefore managed differently from established epilepsy. (professional.heart.org)

Delayed cerebral ischemia affects approximately 30% of patients with aneurysmal hemorrhage, most often between days 4 and 14. It is a clinical syndrome of new focal impairment or sustained deterioration in consciousness not explained by another cause, rather than a synonym for arterial narrowing on angiography. Surveillance combines repeated neurological examinations with selected investigations. The 2023 American guideline considers transcranial Doppler ultrasonography, CT angiography, and CT perfusion useful in appropriate settings; NICE’s 2022 guideline restricts Doppler-guided management to research. Enteral nimodipine and maintenance of normal circulating volume are preventive measures. Established symptomatic ischemia may instead prompt carefully monitored blood-pressure elevation; severe vasospasm may warrant intra-arterial vasodilators or angioplasty. Prophylactic hypertension and hypervolemia are not equivalent to this rescue treatment and are not recommended. (mcgill.ca)

Systemic complications can complicate neurological assessment and physiological support. Low sodium may reflect inappropriate antidiuretic-hormone secretion or renal salt loss; accompanying natriuresis can reduce circulating volume. Monitoring therefore considers electrolytes alongside fluid balance, rather than sodium concentration alone. Cardiac abnormalities include arrhythmias and neurogenic myocardial dysfunction; prospective multicenter research has evaluated these using electrocardiography, echocardiography, and myocardial-injury markers. Pulmonary edema, pneumonia, and acute respiratory distress syndrome can impair oxygenation and necessitate ventilation. Fever may be infectious or noninfectious. Standardized ventilator-care bundles address respiratory complications, while prevention of venous thromboembolism must account for bleeding risk, aneurysm occlusion, and planned neurosurgical procedures. The American guideline recommends mechanical or pharmacological prophylaxis once the ruptured aneurysm has been secured. (mcgill.ca)

Traumatic and nonaneurysmal hemorrhage

Traumatic subarachnoid hemorrhage (tSAH) forms part of the spectrum of traumatic brain injury, rather than a single, uniform clinical syndrome. Mechanical injury can damage vessels at the brain surface, producing subarachnoid blood alongside cerebral contusions, subdural or epidural hematomas, or intracerebral hemorrhage. Contusions involve bruising and bleeding within brain tissue and may occur beneath the impact site or opposite it. The significance of associated subarachnoid blood therefore depends on the overall injury: a small isolated hemorrhage in an alert patient differs substantially from hemorrhage accompanying extensive contusions, swelling, and impaired consciousness. Assessment and treatment are organized around the complete traumatic injury, not simply the presence of blood in one anatomical compartment. (ninds.nih.gov)

Studies of isolated tSAH illustrate this distinction but do not describe all traumatic cases. A retrospective study published in 2019 examined 299 patients admitted during 2010–2014 with isolated tSAH and Glasgow Coma Scale scores of 13–15; patients with other traumatic brain lesions on the initial scan were excluded. Repeat CT showed worsening or new findings in 26 patients, or 8.7%, but none required neurosurgical intervention or deteriorated below the mild-injury Glasgow Coma Scale range. These findings characterize a selected, relatively low-severity subgroup; they cannot establish the risks associated with tSAH in patients with more severe neurological impairment or additional intracranial injuries. (pmc.ncbi.nlm.nih.gov)

Among spontaneous hemorrhages without an identified aneurysm, perimesencephalic nonaneurysmal hemorrhage is distinguished by its early CT distribution. Jan van Gijn and colleagues described the syndrome in 1985: blood was concentrated in the cisterns around the midbrain rather than distributed widely through the subarachnoid space. A 1991 study by Gabriël Rinkel and colleagues further characterized bleeding centered immediately anterior to the brainstem, sometimes extending into the ambient cisterns or basal Sylvian fissures, without complete filling of the lateral Sylvian or anterior interhemispheric fissures. This pattern is associated with a more favorable clinical course, but it is not itself proof that no aneurysm exists: in that study, both observers misclassified one basilar-artery aneurysm as having a nonaneurysmal perimesencephalic pattern. (pubmed.ncbi.nlm.nih.gov)

“Angiography-negative” describes an investigation result, not a demonstrated mechanism. Diffuse nonperimesencephalic bleeding with a negative initial angiogram remains distinct from the restricted perimesencephalic pattern. A ruptured lesion may initially be obscured by vasospasm, thrombosis, or an adjacent hematoma; inadequate projections or interpretation can also produce false-negative findings. In a 2022 referral-center study of 242 patients with spontaneous SAH and negative initial digital-subtraction angiography (DSA), repeat DSA identified a source in 10 of the 135 patients re-examined. The diagnostic yield was 9 of 80, or 11.3%, for nonperimesencephalic hemorrhage and 1 of 46, or 2.2%, for perimesencephalic hemorrhage. Detected sources included small or atypical aneurysms and arteriovenous shunts. Because repeat testing was selective, these percentages are not universal probabilities for every patient with a negative study. (pubmed.ncbi.nlm.nih.gov)

Further investigation consequently depends on the blood distribution, the quality and extent of the initial vascular examination, and continuing clinical suspicion. The 2022 NICE guideline distinguishes failure to detect an aneurysm from exclusion of an aneurysmal source: when CT angiography is negative but an aneurysm remains suspected, it identifies DSA as a further investigation, with magnetic-resonance angiography considered when DSA is contraindicated. Whether subsequent angiography is repeated is a separate question, particularly relevant to diffuse bleeding or an initially incomplete examination; an initially negative vascular study alone does not establish a benign cause. (nice.org.uk)

Outcomes, rehabilitation, and follow-up

Early case fatality, longer-term survival, functional independence, and health-related quality of life measure different outcomes. Hospital case-fatality estimates concern admitted patients and exclude people who die before admission; population-based estimates seek to include these deaths. Discharge outcomes also depend on length of stay and are not interchangeable with fixed 30-day outcomes. In the L’Aquila population registry in Italy, 194 people with nontraumatic subarachnoid hemorrhage were identified during 2011–2020: 28.4% died within 30 days and 37.1% within one year. These figures included aneurysmal and nonaneurysmal hemorrhage and are not estimates specifically for treated aneurysms. Longer-term survival additionally captures deaths after the acute neurological episode, including deaths from other causes. (pmc.ncbi.nlm.nih.gov)

Functional independence does not necessarily mean restoration of previous cognitive abilities, employment, or well-being. The 2023 American Heart Association/American Stroke Association guideline notes that conventional functional assessments may miss cognitive impairment after aneurysmal hemorrhage. Persistent difficulties include impaired memory, attention, processing speed, and executive functions such as planning and organizing tasks. A person who walks independently and manages basic self-care may therefore remain unable to perform complex work reliably. Fatigue, sleep disturbance, depressive symptoms, anxiety, and irritability can add to these limitations despite apparent physical recovery. The guideline supports validated cognitive and behavioral screening and considers the Montreal Cognitive Assessment more suitable than the Mini-Mental State Examination for identifying cognitive impairment in this setting. (ahajournals.org)

Return to employment is a separate participation outcome, and studies distinguish any paid work from resumption of the previous workload. In a 2024 study by Angelika Sorteberg, Aslan Lashkarivand, and Eva Western, 299 survivors of aneurysmal hemorrhage had been employed at onset. At a median follow-up of 3.3 years, approximately 51% had returned to some gainful employment, but only 32.4% had completely resumed their previous working capacity. Clinically significant fatigue was strongly associated with failure to return. These observational findings describe a particular survivor cohort rather than an individual prognosis; retirement opportunities, occupational demands, and social-support arrangements also affect employment outcomes. (frontiersin.org)

Rehabilitation addresses physical, cognitive, communication, and psychosocial needs rather than motor deficits alone. The 2023 AHA/ASA guideline supports multidisciplinary assessment of discharge needs and rehabilitation planning, involving rehabilitation physicians, nurses, physical therapists, occupational therapists, and speech and language therapists as appropriate. Cognitive and mood screening helps identify problems requiring more detailed assessment or treatment. NICE’s 2022 guideline connects post-hemorrhage rehabilitation with stroke and critical-illness rehabilitation pathways. Continuing support includes a documented follow-up plan, information for relatives or carers, specialist contact arrangements, and reassessment of return to work and other usual activities. (professional.heart.org)

Vascular follow-up has a different purpose from rehabilitation: it assesses the durability of aneurysm treatment and monitors additional aneurysms. The AHA/ASA guideline supports follow-up cerebrovascular imaging to identify residual filling, recurrence, regrowth, or changes in other aneurysms that could require further evaluation. NICE does not prescribe one universal imaging schedule. Modality, frequency, and duration depend on the original treatment and its result, additional aneurysms, estimated bleeding risk, recovery, investigation and intervention risks, and patient preferences. Additional unruptured aneurysms are assessed by a multidisciplinary team that includes an interventional neuroradiologist and a neurosurgeon; observation with imaging remains distinct from a decision to undertake preventive treatment. (ahajournals.org)

Development of the evidence base

The International Subarachnoid Aneurysm Trial (ISAT), reported in 2002 by Andrew Molyneux, Richard Kerr, and collaborators, established a large randomized comparison of neurosurgical clipping and endovascular coiling for ruptured intracranial aneurysms. It allocated 2,143 patients to clipping (1,070) or coiling (1,073), provided that their aneurysms were considered suitable for either approach. In the initial report, among patients with one-year outcomes available, death or dependency occurred in 190 of 801 patients allocated to coiling (23.7%) and 243 of 793 allocated to clipping (30.6%). The absolute risk reduction was 6.9 percentage points, with a 95% confidence interval of 2.5–11.3 percentage points. This was a combined survival-and-functional outcome, not a measure of mortality alone. Recruitment was stopped early following an interim analysis. (pubmed.ncbi.nlm.nih.gov)

ISAT’s eligibility requirement was central to its interpretation: it addressed treatment choice when both procedures were feasible, rather than comparing them across all ruptured aneurysms. Its findings therefore supported coiling in appropriately selected patients, not the abandonment of clipping. The original investigators also identified long-term durability and prevention of subsequent bleeding as questions requiring further follow-up; a one-year advantage could not by itself establish lifetime equivalence in aneurysm occlusion. (pubmed.ncbi.nlm.nih.gov)

The National Institute for Health and Care Excellence (NICE) published guideline NG228 on November 23, 2022, covering diagnosis and management of subarachnoid hemorrhage caused by a ruptured aneurysm. Its development framework combined clinical evidence reviews, assessment of evidence quality, health-economic analysis, and committee deliberation. Recommendations consequently represent an evidence-based synthesis rather than a collection of conclusions supported uniformly by randomized trials. An explicit example of an unresolved question concerned long-term blood-pressure control: the accompanying evidence review found no evidence establishing whether tighter control after aneurysmal hemorrhage improved outcomes compared with standard management and proposed a prospective randomized trial. This distinguished an identified research need from an established treatment effect. (nice.org.uk)

The American Heart Association/American Stroke Association guideline, published online on May 22, 2023, replaced its 2012 predecessor. Its scope was explicitly aneurysmal hemorrhage, excluding hemorrhage caused by trauma, vascular malformations, or hemorrhage-prone tumors. It integrated evidence across diagnosis, aneurysm treatment, neurocritical care, and recovery, while separately identifying knowledge gaps. Examples of evidence-supported recommendations included early enteral nimodipine and avoidance of routine prophylactic hypervolemia; the effectiveness of some monitoring approaches for improving clinical outcomes remained less certain. The guideline therefore distinguished physiological information supplied by a test from evidence that using that test changes disability or survival. (pubmed.ncbi.nlm.nih.gov)

Standardization of research measurements developed alongside treatment trials. The National Institute of Neurological Disorders and Stroke (NINDS), working with the National Library of Medicine and an expert working group, released version 1.0 of its subarachnoid hemorrhage Common Data Elements in 2017. Eight domains included subject characteristics, examinations, acute therapies and hospital course, imaging, biospecimens and biomarkers, long-term therapies, unruptured aneurysms, and outcomes. Defined variables, template case-report forms, and guidance documents provide a shared structure for data collection. Elements are classified as Core, Supplemental–Highly Recommended, Supplemental, or Exploratory, allowing studies to distinguish a common minimum from question-specific measurements. This framework supports comparison of patient populations, interventions, and endpoints across studies without making their designs or populations identical. (commondataelements.ninds.nih.gov)

References

1. Traumatic Brain Injury (TBI)ninds.nih.gov
2. Neuroanatomy Online: Lab 1 - Overview of the Nervous System - Meningeal Coveringsnba.uth.tmc.edu
3. Neuroanatomy Lab Online, Lab 1 Page 3oac22.hsc.uth.tmc.edu
4. Medical Neurosciencelumen.luc.edu
5. Terminologia Neuroanatomica: International Neuroanatomical Terminologycdn.dal.ca
6. NINDS ImagingRead Data Structurepdbp.ninds.nih.gov
7. Subarachnoid hemorrhage: MedlinePlus Medical Encyclopediamedlineplus.gov
8. 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
9. MeSH Browser: Intracranial Aneurysmmeshb.nlm.nih.gov
10. Preventionninds.nih.gov
11. Cerebral Aneurysmsninds.nih.gov
12. The natural course of unruptured cerebral aneurysms in a Japanese cohortpubmed.ncbi.nlm.nih.gov
13. Development of the PHASES score for prediction of risk of rupture of intracranial aneurysms: a pooled analysis of six prospective cohort studiespubmed.ncbi.nlm.nih.gov
14. Blood Thinnersmedlineplus.gov
15. Joint effect of modifiable risk factors on the risk of aneurysmal subarachnoid hemorrhage: a cohort studypubmed.ncbi.nlm.nih.gov
16. Cerebral Aneurysmspmc.ncbi.nlm.nih.gov
17. Substantial Within-Country Variation in the Incidence of Subarachnoid Hemorrhage: A Nationwide Finnish Studypubmed.ncbi.nlm.nih.gov
18. Interaction of Optimal Cerebral Perfusion Pressure with Early Brain Injury and its Impact on Ischemic Complications and Outcome Following Aneurysmal Subarachnoid Hemorrhagepmc.ncbi.nlm.nih.gov
19. Cerebral perfusion Pressure and Delayed Cerebral Ischemia After Aneurysmal Subarachnoid Hemorrhagepmc.ncbi.nlm.nih.gov
20. Early CT perfusion changes and blood-brain barrier permeability after aneurysmal subarachnoid hemorrhagepubmed.ncbi.nlm.nih.gov
21. Early blood-brain barrier dysfunction predicts neurological outcome following aneurysmal subarachnoid hemorrhagepmc.ncbi.nlm.nih.gov
22. Methemoglobin Is an Endogenous Toll-Like Receptor 4 Ligand—Relevance to Subarachnoid Hemorrhagepmc.ncbi.nlm.nih.gov
23. Interleukin-1 receptor antagonist is beneficial after subarachnoid haemorrhage in rat by blocking haem-driven inflammatory pathologypubmed.ncbi.nlm.nih.gov
24. Microglia regulate blood clearance in subarachnoid hemorrhage by heme oxygenase-1pmc.ncbi.nlm.nih.gov
25. Subarachnoid hemorrhage: cerebral damage, fluid balance, intracranial pressure and pressure-volume relationpubmed.ncbi.nlm.nih.gov
26. Posthemorrhagic hydrocephalus associates with elevated inflammation and CSF hypersecretion via activation of choroidal transporterspmc.ncbi.nlm.nih.gov
27. Subarachnoid blood acutely induces spreading depolarizations and early cortical infarctionpmc.ncbi.nlm.nih.gov
28. Recommendations | Subarachnoid haemorrhage caused by a ruptured aneurysm: diagnosis and management | Guidance | NICEnice.org.uk
29. 4 Primary thunderclap headache - ICHD-3ichd-3.org
30. Headache attributed to cranial or cervical vascular disorder - ICHD-3ichd-3.org
31. Headache Classification Committee of the International Headache Society (IHS) The International Classification of Headache Disorders, 3rd editionichd-3.org
32. 5.1.1 Acute headache or facial or neck pain attributed to cervical carotid or vertebral artery dissection - ICHD-3ichd-3.org
33. Recommendations | Meningitis (bacterial) and meningococcal disease: recognition, diagnosis and management | Guidance | NICEnice.org.uk
34. Subarachnoid hemorrhage indicator | NIH Common Data Elements Repositorycde.nlm.nih.gov
35. Rationale and impact | Subarachnoid haemorrhage caused by a ruptured aneurysm: diagnosis and management | Guidance | NICEnice.org.uk
36. Differentiation between traumatic tap and aneurysmal subarachnoid hemorrhage: prospective cohort studypmc.ncbi.nlm.nih.gov
37. NG228 Evidence review K: Diagnostic imaging strategiesnice.org.uk
38. NG228 Evidence review C: Severity scoring systemsnice.org.uk
39. Surgical risk as related to time of intervention in the repair of intracranial aneurysmspubmed.ncbi.nlm.nih.gov
40. Pediatric Brain Arteriovenous Malformationshelda.helsinki.fi
41. WFNS grading system: JSON Representationbuild.fhir.org
42. Relation of cerebral vasospasm to subarachnoid hemorrhage visualized by computerized tomographic scanningpubmed.ncbi.nlm.nih.gov
43. The WFNS Classification, Fisher Grading Scale, and Modified Rankin Scalejstage.jst.go.jp
44. Prediction of symptomatic vasospasm after subarachnoid hemorrhage: the modified fisher scalepubmed.ncbi.nlm.nih.gov
45. Claudin proteins and hemorrhage severity in aneurysmal subarachnoid hemorrhage: Correlation with modified Fisher score but not functional outcomepmc.ncbi.nlm.nih.gov
46. NG228 Evidence review H: Managing hydrocephalusnice.org.uk
47. 2023 Guideline for the Management of Patients With Aneurysmal Subarachnoid Hemorrhagemcgill.ca
48. Cardiac dysfunction after aneurysmal subarachnoid hemorrhage: relationship with outcomepubmed.ncbi.nlm.nih.gov
49. Traumatic Brain Injury | Hope Through Researchninds.nih.gov
50. Perimesencephalic hemorrhage: a nonaneurysmal and benign form of subarachnoid hemorrhagepubmed.ncbi.nlm.nih.gov
51. Nonaneurysmal perimesencephalic subarachnoid hemorrhage: CT and MR patterns that differ from aneurysmal rupturepubmed.ncbi.nlm.nih.gov
52. Angiographically Occult Subarachnoid Hemorrhage: Yield of Repeat Angiography, Influence of Initial CT Bleed Pattern, and Sources of Diagnostic Error in 242 Consecutive Patientspubmed.ncbi.nlm.nih.gov
53. Demographic disparities in the incidence and case fatality of subarachnoid haemorrhage: an 18-year nationwide study from New Zealandpmc.ncbi.nlm.nih.gov
54. Return to work after aneurysmal subarachnoid hemorrhagefrontiersin.org
55. International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised trialpubmed.ncbi.nlm.nih.gov
56. NG228 Methodsnice.org.uk
57. NG228 Evidence review Qnice.org.uk
58. 2023 Guideline for the Management of Patients With Aneurysmal Subarachnoid Hemorrhage: A Guideline From the American Heart Association/American Stroke Associationpubmed.ncbi.nlm.nih.gov
59. Unruptured Cerebral Aneurysms and Subarachnoid Hemorrhagecommondataelements.ninds.nih.gov