Subarachnoid Hemorrhage

The term subarachnoid hemorrhage (SAH) refers to extravasation of blood into the subarachnoid space between the pial and arachnoid membranes. SAH constitutes half of all spontaneous atraumatic intracranial hemorrhages; the other half consists of bleeding that occurs within the brain parenchyma.

Subarachnoid hemorrhage (see the image below) occurs in various clinical contexts, the most common being head trauma. However, the familiar use of the term SAH refers to nontraumatic (or spontaneous) hemorrhage, which usually occurs in the setting of a ruptured cerebral aneurysm or arteriovenous malformation (AVM).

CT scan reveals subarachnoid hemorrhage in the right sylvian fissure; no evidence of hydrocephalus is apparent.

View Media Gallery

Intracranial saccular aneurysms (“berry aneurysms”) represent the most common etiology of nontraumatic SAH; about 80% of cases of SAH result from ruptured aneurysms. SAH is responsible for the death and/or disability of 18,000 persons each year in North America alone. In the United States, it is associated with an annual cost of $1.75 billion. Unfortunately, the difficulties in detecting unruptured aneurysms in asymptomatic patients practically preclude the possibility of preventing most instances of SAH.

About 6-8% of all strokes are caused by SAH from ruptured berry aneurysms. Over the past several decades, the incidence of other types of strokes has decreased; however, the incidence of SAH has not decreased.

The history and physical examination, especially the neurologic examination, are essential components in the diagnosis and clinical staging of SAH (see Presentation). The diagnosis is confirmed radiologically via urgent computed tomography (CT) scan without contrast. Traditionally, a negative CT scan is followed with lumbar puncture. However, noncontrast CT followed by CT angiography (CTA) of the brain can rule out SAH with greater than 99% sensitivity. ^[2] (See Workup.)

Current treatment recommendations involve management in an intensive care unit setting. The blood pressure is maintained with consideration of the patient’s neurologic status, and additional medical management is directed toward the prevention and treatment of complications. Surgical treatment to prevent rebleeding consists of clipping the ruptured berry aneurysm. Endovascular treatment ^[1] (ie, coiling) is an increasingly practiced alternative to surgical clipping (see Treatment).

Aneurysms are acquired lesions related to hemodynamic stress on the arterial walls at bifurcation points and bends. Saccular or berry aneurysms are specific to the intracranial arteries because their walls lack an external elastic lamina and contain a very thin adventitia—factors that may predispose to the formation of aneurysms. An additional feature is that they lie unsupported in the subarachnoid space.

Aneurysms usually occur in the terminal portion of the internal carotid artery and the branching sites on the large cerebral arteries in the anterior portion of the circle of Willis. The early precursors of aneurysms are small outpouchings through defects in the media of the arteries.

These defects are thought to expand as a result of hydrostatic pressure from pulsatile blood flow and blood turbulence, which is greatest at the arterial bifurcations. A mature aneurysm has a paucity of media, replaced by connective tissue, and has diminished or absent elastic lamina.

The probability of rupture is related to the tension on the aneurysm wall. The law of La Place states that tension is determined by the radius of the aneurysm and the pressure gradient across the wall of the aneurysm. Thus, the rate of rupture is directly related to the size of the aneurysm. Aneurysms with a diameter of 5 mm or less have a 2% risk of rupture, whereas 40% of those with a diameter of 6-10 mm have already ruptured upon diagnosis.

Although hypertension has been identified as a risk factor for aneurysm formation, the data with respect to rupture are conflicting. However, certain hypertensive states, such as those induced by use of cocaine and other stimulants, clearly promote aneurysm growth and rupture earlier than would be predicted by the available data.

Brain injury from cerebral aneurysm formation can occur in the absence of rupture. Compressive forces can cause injury to local tissues and/or compromise of distal blood supply (mass effect).

When an aneurysm ruptures, blood extravasates under arterial pressure into the subarachnoid space and quickly spreads through the cerebrospinal fluid around the brain and spinal cord. Blood released under high pressure may directly cause damage to local tissues. Blood extravasation causes a global increase in intracranial pressure (ICP). Meningeal irritation occurs.

Rupture of AVMs can result in both intracerebral hemorrhage and SAH. Currently, no explanation can be provided for the observation that small AVMs (< 2.5 cm) rupture more frequently than large AVMs (>5 cm).

In a 25-year autopsy study of 125 patients with ruptured or unruptured aneurysms conducted at Johns Hopkins, the following conditions correlated positively with the formation of saccular aneurysms:

• Hypertension

• Cerebral atherosclerosis

• Vascular asymmetry in the circle of Willis

• Persistent headache

• Pregnancy-induced hypertension

• Long-term analgesic use

• Family history of stroke

The occurrence of aneurysms in children indicates the role of intrinsic vascular factors. A number of disease states resulting in weakness of the arterial wall are associated with an increased incidence of berry aneurysms.

Mechanisms and disease states associated with higher incidence of berry aneurysms include the following:

• Increased blood pressure: Fibromuscular dysplasia, polycystic kidney disease, aortic coarctation

• Increased blood flow: Cerebral arteriovenous malformation (AVM); persistent carotid-basilar anastomosis; ligated, aplastic, or hypoplastic contralateral vessel

• Blood vessel disorders: Systemic lupus erythematosus (SLE), Moyamoya disease, ^[3] granulomatous angiitis

• Genetic disorders: Marfan syndrome, Ehlers-Danlos syndrome, Osler-Weber-Rendu syndrome, pseudoxanthoma elasticum, Klippel-Trenaunay-Weber syndrome

• Congenital conditions: Persistent fetal circulation, hypoplastic/absent arterial circulation

• Metastatic tumors to cerebral arteries: Atrial myxoma, choriocarcinoma, undifferentiated carcinoma

• Infections: Bacterial, fungal

Complications

Complications of SAH include the following:

• Hydrocephalus

• Rebleeding

• Delayed cerebral ischemia from vasospasm

• Intracerebral hemorrhage

• Intraventricular hemorrhage

• Left ventricular systolic dysfunction

• Subdural hematoma

• Seizures

• Increased intracranial pressure

• Myocardial infarction ^[4]

Hydrocephalus

SAH can cause hydrocephalus by 2 mechanisms: obstruction of CSF pathways (ie, acute, obstructive, noncommunicating type) and blockage of arachnoid granulations by scarring (ie, delayed, nonobstructive, communicating type). Acute hydrocephalus is caused by compromise of CSF circulation pathways by interfering with CSF outflow through the sylvian aqueduct, fourth ventricular outlet, basal cisterns, and subarachnoid space. CSF production and absorption rates are unaltered.

Intraventricular blood is the strongest determinant for the development of acute hydrocephalus. Other risk factors include the following:

• Bilateral ambient cisternal blood

• Increased age

• Vasospasm

• Use of antifibrinolytic drugs

• Intraventricular hemorrhage

• Left ventricular systolic dysfunction

• Subdural hematoma

• Seizures

Rebleeding

Rebleeding of SAH occurs in 20% of patients in the first 2 weeks. The rebleeds in the first days ("blow out" hemorrhages) are thought to be related to the unstable nature of the aneurysmal thrombus, as opposed to lysis of the clot sitting over the rupture site. Clinical factors that increase the likelihood of rebleeding include hypertension, anxiety, ^[5] agitation, and seizures.

Cerebral ischemia

Delayed cerebral ischemia from arterial smooth muscle contraction is the most common cause of death and disability following aneurysmal SAH. Vasospasm can lead to impaired cerebral autoregulation and may progress to cerebral ischemia and infarction. ^[6] Most often, the terminal internal carotid artery or the proximal portions of the anterior and middle cerebral arteries are involved. The arterial territory involved is not related to the location of the ruptured aneurysm.

Vasospasm is believed to be induced in areas of thick subarachnoid clot. The putative agent responsible for vasospasm is oxyhemoglobin, but its true etiology and pathogenesis remain to be elucidated.

Intracerebral hemorrhage

The mechanism of intracerebal hemorrhage (ICH) is direct rupture of aneurysm into the brain. ICH commonly results from internal cerebral artery (ICA), pericallosal, and anterior cerebral artery (ACA) aneurysms. Secondary rupture of a subarachnoid hematoma into the brain parenchyma most commonly arises from middle cerebral artery aneurysms.

Intraventricular hemorrhage

Found in 13-28% of clinical cases of ruptured aneurysms and in 37-54% of autopsy cases, intraventricular hemorrhage (IVH) is a significant predictor of poor neurologic grade and outcome. Sources of IVH include the following:

• Anterior cerebral artery (40%)

• Internal cerebral artery (25%)

• Middle cerebral artery (21%)

• Vertebrobasilar artery (14%)

Left ventricular systolic dysfunction

LV systolic dysfunction in humans with SAH is associated with normal myocardial perfusion and abnormal sympathetic innervation. These findings may be explained by excessive release of norepinephrine from myocardial sympathetic nerves, which could damage both myocytes and nerve terminals. ^[7]

Subdural hematoma

Subdural hematoma (SDH) is rare following aneurysmal SAH, with reported incidence of 1.3-2.8% in clinical series and as high as 20% in autopsy series. The mechanisms of SDH involve tearing of arachnoid adherent to the dome of the aneurysm at the time of rupture, direct tearing of arachnoid by a jet of blood, and disruption of arachnoid by ICH, with secondary decompression of ICH into the subdural space.

Increased intracranial pressure

Elevations in ICP are due to mass effect of blood (subarachnoid, intracranial, intraventricular, or subdural hemorrhage) or acute hydrocephalus. Once ICP reaches mean arterial pressure (MAP), cerebral perfusion pressure becomes zero and cerebral blood flow stops, resulting in loss of consciousness and death.

Of nontraumatic subarachnoid hemorrhages, approximately 80% are due to a ruptured berry aneurysm. Rupture of arteriovenous malformations (AVMs) is the second most identifiable cause of SAH, accounting for 10% of cases of SAH. Most of the remaining cases result from rupture of the following types of pathologic entities:

• Mycotic aneurysm

• Angioma

• Neoplasm

• Cortical thrombosis

SAH may reflect a secondary dissection of blood from an intraparenchymal hematoma (eg, bleeding from hypertension or neoplasm).

Both congenital and acquired factors are thought to play a role in SAH. Evidence supporting the role of congenital causes in aneurysm formation includes the following:

• Clusters of familial occurrence, such as in Finland, where the incidence of familial cerebral aneurysm is 10%

• Significant incidence of multiple aneurysms in patients with SAH (15%)

• The association of aneurysms with specific congenital diseases (eg, coarctation of the aorta, Marfan syndrome, Ehlers-Danlos syndrome, fibromuscular dysplasia, polycystic kidney disease)

Familial cases of AVM are rare, and the problem may result from sporadic abnormalities in embryologic development. AVMs are thought to occur in approximately 4-5% of the general population, of which 10-15% are symptomatic. Congenital defects in the muscle and elastic tissue of the arterial media in the vessels of the circle of Willis are found in approximately 80% of normal vessels at autopsy. These defects lead to microaneurysmal dilation (< 2 mm) in 20% of the population and larger dilation (>5 mm) and aneurysms in 5% of the population.

Acquired factors thought to be associated with aneurysmal formation include the following:

• Atherosclerosis

• Hypertension

• Advancing age

• Smoking

• Hemodynamic stress

Less common causes of SAH include the following:

• Fusiform and mycotic aneurysms

• Fibromuscular dysplasia

• Blood dyscrasias

• Moyamoya disease

• Infection

• Neoplasm

• Trauma (fracture at the base of the skull leading to internal carotid aneurysm)

• Amyloid angiopathy (especially in elderly people)

• Vasculitis

Reversible cerebral vasoconstriction syndrome (RCVS) is characterized by recurrent thunderclap headaches and reversible segmental multifocal cerebral artery narrowing, and it results in SAH in more than 30% of cases. Muehlschlegel and colleagues found that clinical and imaging findings can differentiate RCVS with SAH from other causes of SAH. ^{[8, 9]}

After analyzing clinical and imaging features of 38 patients with RCVS-SAH, 515 patients with aneurysmal SAH, and 93 patients with cryptogenic (angiogram negative) SAH, Muehlschlegel et al identified clinical characteristics and radiological findings that can differentiate RCVS-SAH from aneurysmal SAH or cryptogenic SAH. These researchers concluded that these differences may be useful for improving diagnostic accuracy, clinical management, and resource utilization. ^{[8, 9]}

Risk factors

Although risk factors for SAH have been evaluated extensively, little conclusive evidence has been derived. Smoking appears to be a significant risk factor, as does heavy alcohol consumption. The risk of AVM rupture is greater during pregnancy. Data regarding the relationship between hypertension and SAH are conflicting. Previously documented acute severe hypertension with diastolic pressure over 110 mm Hg has been linked to SAH.

The following do not appear to be significant risk factors for SAH:

• Use of oral contraceptives

• Hormone replacement therapy

• Hypercholesterolemia

• Vigorous physical activity

United States statistics

The frequency of ruptured and unruptured aneurysms has been estimated at 1-9% in different autopsy series, with a prevalence of unruptured aneurysms of 0.3-5%. Retrospective arteriographic studies show a prevalence of less than 1% with the limitation that some cases did not receive adequate evaluation and thus some aneurysms may have been missed. Annual incidence increases with age and probably is underestimated because death is attributed to other reasons that are not confirmed by autopsies.

The annual incidence of aneurysmal SAH in the United States is 6-16 cases per 100,000 population, with approximately 30,000 episodes occurring each year. Unlike other subcategories of stroke, the incidence of SAH has not decreased over time. However, since 1970, population-based survival rates have improved.

International statistics

The reported incidence of subarachnoid hemorrhage is high in the United States, Finland, and Japan, while it is low in New Zealand and the Middle East. In Finland, the estimated incidence based on different studies is 14.4-19.6 cases per 100,000 population, although numbers as high as 29.7 have been reported.

In Japan, the reported rates vary between 11 and 18.3 cases per 100,000 population, with one study showing an incidence of 96.1 cases per 100,000 population (this study included only patients aged 40 and older in the data collection, and results were not adjusted for sex and age to the same reference population). In New Zealand, age-adjusted incidence was reported as 14.3 cases per 100,000 population.

An Australian study reported an incidence of 26.4 cases per 100,000 population but only for patients older than 35 years, as age was not adjusted in the reference population. In the Netherlands, the age-specific incidence was reported as 7.8 cases per 100,000 population (this is believed to be an underestimate).

Iceland reported 8 cases per 100,000 population, but a significant portion of the affected rural population was believed to be missed. Greenland Eskimos had 9.3 cases per 100,000 population; ethnic Danes there had an incidence of 3.1 cases per 100,000 population. This latter figure is consistent with the figures in Denmark—marked differences are postulated to be related to genetic factors. On the Faeroe Islands (part of Denmark with an isolated population of the same genetic ancestry), the reported incidence is 7.4 cases per 100,000 population.

In China, the reported incidence is low, but no good studies have been published to support this statement. The incidence among Indians and Rhodesian Africans is significantly lower than in those from European nations; this can be explained partly by the low incidence of atherosclerosis in these populations. In the Middle East, the numbers are very low as well; the best available estimate is 5.1 cases per 100,000 population in Qatar.

Race-, sex-, and age-related demographics

The risk is higher in blacks than in whites; however, people of all ethnic groups develop intracranial aneurysms. The disparity in frequency of rupture has been attributed to population variance with respect to prevalence of risk factors and age distribution.

The incidence of SAH in women is higher than in men (ratio of 3 to 2). The risk of SAH is significantly higher in the third trimester of pregnancy, and SAH from aneurysmal rupture is a leading cause of maternal mortality, accounting for 6-25% of maternal deaths during pregnancy. A higher incidence of AVM rupture also has been reported during pregnancy.

Incidence increases with age and peaks at age 50 years. Approximately 80% of cases of SAH occur in people aged 40-65 years, with 15% occurring in people aged 20-40 years. Only 5% of cases of SAH occur in people younger than 20 years. SAH is rare in children younger than 10 years, accounting for only 0.5% of all cases.

Although mortality rates of SAH have decreased in the past 3 decades, it remains a devastating neurologic problem. An estimated 10-15% of patients die before reaching the hospital. Approximately 25% of patients die within 24 hours, with or without medical attention. Hospitalized patients have an average mortality rate of 40% in the first month. About half of affected individuals die in the first 6 months. Rebleeding, a major complication, carries a mortality rate of 51-80%.

Age-adjusted mortality rates are 62% greater in females than in males and 57% greater in blacks than in whites. Morbidity and mortality increase with age and are related to the overall health status of the patient.

More than one third of survivors have major neurologic deficits. Cognitive deficits are present even in many patients considered to have a good outcome.

Al-Khindi et al found that survivors of aneurysmal SAH commonly experience deficits in memory, executive function, and language that affect their day-to-day functioning, including activities of daily living, instrumental activities of daily living, return to work, and quality of life. Deficits in cognition and day-to-day functioning are further compounded by depression, anxiety, fatigue, and sleep disturbances. ^[10]

Factors that affect morbidity and mortality rates are as follows:

• Severity of hemorrhage

• Degree of cerebral vasospasm

• Occurrence of rebleeding

• Presence of comorbid conditions and the hospital course (eg, infections, myocardial infarction)

Other factors that affect the prognosis of patients who have suffered an SAH include age, Hunt and Hess grade (see below), smoking history, and location of the aneurysm. Younger patients do better. Patients with a history of cigarette smoking have a poorer prognosis. Anterior circulation aneurysms carry a more favorable prognosis.

Acute cocaine use was associated with higher rates of in-hospital death and a significantly increased risk for aneurysm rerepture in a retrospective study of 1134 patients with aneurysmal SAH. Compared with patients who had not used cocaine in the 72 hours preceding their event, those who had used cocaine had a nearly 3-fold increased risk for in-hospital mortality. Mortality remained higher among cocaine users after patients with rerupture were excluded from the analysis, suggesting that rerupture was not entirely responsible for the higher mortality rate in these patients. ^[11]

Clinical grading scales

Clinical assessment of SAH severity commonly utilizes grading scales. The 2 clinical scales most often employed are the Hunt and Hess and the World Federation of Neurological Surgeons (WFNS) grading systems. A third, the Fisher scale, classifies SAH based on CT scan appearance and quantification of subarachnoid blood.

The WFNS scale is as follows:

• Grade 1 - Glasgow Coma Score (GCS) of 15, motor deficit absent

• Grade 2 - GCS of 13-14, motor deficit absent

• Grade 3 - GCS of 13-14, motor deficit present

• Grade 4 - GCS of 7-12, motor deficit absent or present

• Grade 5 - GCS of 3-6, motor deficit absent or present

The Fisher scale (CT scan appearance) is as follows:

• Group 1 - No blood detected

• Group 2 - Diffuse deposition of subarachnoid blood, no clots, and no layers of blood greater than 1 mm

• Group 3 - Localized clots and/or vertical layers of blood 1 mm or greater in thickness

• Group 4 - Diffuse or no subarachnoid blood, but intracerebral or intraventricular clots are present

The Hunt and Hess grading system is as follows:

• Grade 0 - Unruptured aneurysm

• Grade I - Asymptomatic or mild headache and slight nuchal rigidity

• Grade Ia - Fixed neurologic deficit without acute meningeal/brain reaction

• Grade II - Cranial nerve palsy, moderate to severe headache, nuchal rigidity

• Grade III - Mild focal deficit, lethargy, or confusion

• Grade IV - Stupor, moderate to severe hemiparesis, early decerebrate rigidity

• Grade V - Deep coma, decerebrate rigidity, moribund appearance

In the Hunt and Hess system, the lower the grade, the better the prognosis. Grades I-III generally are associated with favorable outcome; these patients are candidates for early surgery. Grades IV and V carry a poor prognosis; these patients need stabilization and improvement to grade III before surgery is undertaken. Some recommend more aggressive management for patients with poor clinical grade.

Survival correlates with the grade of subarachnoid hemorrhage upon presentation. Reported figures include a 70% survival rate for Hunt and Hess grade I, 60% for grade II, 50% for grade III, 40% for grade IV, and 10% for grade V.

The Hunt and Hess and the WFNS grading systems have been shown to correlate well with patient outcome. The Fisher classification has been used successfully to predict the likelihood of symptomatic cerebral vasospasm, one of the most feared complications of SAH. All 3 grading systems are useful in determining the indications for and timing of surgical management. For an accurate assessment of SAH severity, these grading systems must be used in concert with the patient's overall general medical condition and the location and size of the ruptured aneurysm.

Complications

Complications of SAH include the following:

• Hydrocephalus

• Rebleeding

• Delayed ischemia

• Intracerebral hemorrhage

• Intraventricular hemorrhage (IVH)

• Left ventricular systolic dysfunction

• Subdural hematoma

• Seizures

• Increased intracranial pressure

• Myocardial infarction ^[4]

The incidence of rebleeding complication is greatest in the first 2 weeks. The peak is within 24-48 hours following initial SAH (approximately 6%), with a rate of 1.5% per day for the next 12-13 days. The cumulative 2-week incidence is 20-30% in unoperated patients. After the first 30 days, rebleed rate decreases to 1.5% per year for the first 10 years. In another study, rebleeding was reported at a rate of 3% per year after 6 months, with a 67% mortality rate at 20 years.

Delayed ischemia

Delayed ischemia from cerebral vasospasm is currently the most common cause of death and disability following aneurysmal SAH. It has to some degree cancelled out the improvement in morbidity and mortality from the lower rebleed rate related to early surgical clipping.

An estimated 10-20% of patients with aneurysmal SAH suffer delayed cerebral ischemia, resulting in permanent disability or death. This complication alone accounts for 14-32% of deaths and permanent disability in large studies, while the direct effect of aneurysm rupture accounts for 25% and rebleeding for 17.6%. Approximately 15-20% of patients with symptomatic vasospasm will have a poor outcome despite maximal medical therapy, including mortality in 7-10% of patients and severe morbidity in 7-10% of patients.

Intraventricular hemorrhage

Found in 13-28% of clinical cases of ruptured aneurysms and in 37-54% of autopsy cases, intraventricular hemorrhage (IVH) is a significant predictor of poor neurologic grade and outcome. Patients with IVH are at higher risk of developing hydrocephalus. In one study of 91 patients, IVH was associated with an overall mortality rate of 64%. The key prognostic indicator is the degree of ventricular dilatation.