-A BRIEF REVIEW
A subarachnoid haemorrhage is bleeding into the subarachnoid space—the area between the arachnoid mater and the pia mater surrounding the brain. This may occur spontaneously, usually from a ruptured cerebral aneurysm, or may result from head injury.
SAH is a form of stroke and comprises 1–7% of all strokes. It is a medical emergency and can lead to death or severe disability—even when recognized and treated at an early stage. Up to half of all cases of SAH are fatal and 10–15% dies before reaching a hospital, and those who survive often have neurological or cognitive impairment.
The20th century has seen great advances in diagnosis, startingwith the ability to recognize the condition at all during life. Advances in treatment and preventionof complications have also occurred, but these have led to onlymodest improvement in overall outcome; hencethere are still formidable challenges ahead for neurologists,neurosurgeons and radiologists.
In the recent years, there are various researches, clinical trials, review works about every aspects of subarachnoid haemorrhage including its risk factors, path physiology, clinical features and management sides. So, the aim of this review article is to boost up our thinking about subarachnoid haemorrhage with the updated knowledge.
The incidence of SAH has remained stable over the last 30 years.In a meta-analysis of relevant studies in 1996, the pooled incidencerate was 10.5 per 100 000 person per year.  There seemed to be a decline over time, but this was causedby diagnostic bias. That more recent studies reported lowerincidence rates than older studies could be entirely explainedby the increasing proportion of patients investigated with CTscanning. In a virtual study in which CT is applied to all patients,the incidence is calculated to be 5.6 per 100 000 patient years  (Table 1).
Table 1 Epidemiological characteristics of SAH [4, 5]
To update 1996 review on the incidence of subarachnoid haemorrhage (SAH), another study published in 2007 (December) which reveals the overall incidence of SAH is approximately 9 per 100 000 person-years. Studies from Japan and Finland show higher rates in those countries (22.7 and 19.7, respectively), for reasons that are not entirely understood. South and Central America, in contrast, have a rate of 4.2 per 100,000 on average.<href=”#cite_note-DeRooij2007-47>The decline in incidence of SAH over the past 45 years is relatively moderate compared with that for stroke in general. Between 1950 and 2005, the incidence decreased by 0.6% per year.<href=”#cite_note-DeRooij2007-47>
Although the group of people at risk for SAH is younger than the population usually affected by stroke,<href=”#cite_note-Feigin05-1> the risk still increases with age,with a mean age at presentation of 55 years. Young people are much less likely than middle-aged people to suffer a subarachnoid haemorrhage.<href=”#cite_note-DeRooij2007-47> The risk continues to rise with age and is 60% higher in the very elderly (over 85) than in those between 45 and 55. <href=”#cite_note-DeRooij2007-47>
The incidence is 1.6 times higher in women than in men,[8,9] althoughthis difference does not carry across all populations. Studieshave suggested that the gender difference is related to hormonalstatus. Premenopausal women are at reduced risk for subarachnoid haemorrhage, especially those without a history of smoking or hypertension. Increased SAH risk was associated with (1) earlierage at menarche and(2) null gravidity. No significant association of SAH risk was found with regularityof menstrual cycle, age at pregnancy, age at first birth, andnumber of births.  There appear to be racialdifferences in risk of SAH. Black people have a 2.1 times higher riskthan whites. 
Fig.1. Average number of people with SAH per 100,000 person-years, broken down by age.
An important, but non-modifiable risk factor is familial predispositionto SAH. Between five and 20% of patients with SAH have a positivefamily history. First-degree relatives ofpatients with SAH have a 3- to 7-fold increased risk of beingstruck by the same disease.[15-19] In second-degree relatives, the incidenceof SAH is similar to that found in the general population. In a review in 2007, study showed three- to fivefold increased risk in first-degree relatives of people who have suffered a subarachnoid haemorrhage.<href=”#cite_note-oxford-3>
The occurrence of SAH is also associated with specific heritabledisorders of connective tissue, but these patients account foronly a minority of all patients with SAH. Even though autosomaldominant polycystic kidney disease (ADPKD) is the most commonheritable disorder associated with SAH, it is found in only2% of all patients with SAH. Othergenetically determined disorders that have been associated withSAH are Ehlers–Danlos disease IV and neurofibromatosistype 1, but these associations are weaker than between ADPKD and these syndromes are seldom found in patientswith SAH.[22,23] Marfan’ssyndrome has often been associated with SAH, but in a clinicalcohort of 129 patients with Marfan’s syndrome, none had a historyof SAH.
Modifiable risk factors
To identify modifiable risk factorsfor subarachnoid haemorrhage (SAH), Teunissen et al performed a Medlinesearch from 1966 through 1994. Significant risk factors were as follows: (1) smoking, (2) hypertension and (3) drinking150 g or more of alcohol per week. Use of oral contraceptives, hormonereplacement therapy, hypercholesterolemia, and physical activity werenot significantly related to the risk of SAH. 
After a 1996 review, there is review study by the same group in 2005. Results reveal Smoking, hypertension, and excessive alcohol againremain the most important risk factors for SAH. Having smoked in the past confers a doubled risk of SAH compared to those who have never smoked. The seeminglyprotective effects of white ethnicity compared to nonwhite ethnicity,hormone replacement therapy, hypercholesterolemia, and diabetesin the etiology of SAH are uncertain.<href=”#cite_note-Feigin05-1>
Causes of subarachnoid haemorrhage
According to the recent review study report, in 85% of cases of spontaneous SAH, the cause is rupture of a cerebral aneurysm—a weakness in the wall of one of the arteries in the brain that becomes enlarged. They tend to be located in the circle of Willis and its branches. While most cases of SAH are due to bleeding from small aneurysms, larger aneurysms (which are less common) are more likely to rupture.<href=”#cite_note-vanGijn-0>
In 15–20% of cases of spontaneous SAH, no aneurysm is detected on the first angiogram.<href=”#cite_note-Rinkel93-10> About half of these are attributed to non-aneurysmal perimesencephalic haemorrhage, in which the blood is limited to the subarachnoid spaces around the midbrain (i.e. mesencephalon). In these, the origin of the blood is uncertain.<href=”#cite_note-vanGijn-0> The remainder are due to other disorders affecting the blood vessels (such as arteriovenous malformations), disorders of the blood vessels in the spinal cord, and bleeding into various tumors.<href=”#cite_note-vanGijn-0> Cocaine abuse and sickle cell anemia (usually in children) and, rarely, anticoagulant therapy, problems with blood clotting and pituitary apoplexy can also result in SAH.<href=”#cite_note-OTV3-5>[26, 27]
Subarachnoid blood can be detected on CT scanning in as many as 60% of people with traumatic brain injury.<href=”#cite_note-Armin06-11> Traumatic SAH (tSAH) usually occurs near the site of a skull fracture or intracerebral contusion.<href=”#cite_note-Rinkel93-10>
Subarachnoid haemorrhage should always be suspected in patientswith a typical presentation which includes a suddenonset of severe headache (frequently described as the “worstever”) associated with nausea, vomiting, neck pain, photophobia, seizures (1 in 14) and lossof consciousness. Physical examination may reveal meningismus, a diminished level of consciousness, and localizingneurologic signs. In patients with such neurological deficits, it is straightforwardthat they should be referred for further investigation. 
About one-third of sufferers have no symptoms apart from the characteristic headache, and about one in ten people who seek medical care with this symptom are later diagnosed with a subarachnoid haemorrhage.<href=”#cite_note-vanGijn-0> In patientsin whom headache is the only symptom, it is often more difficultto recognize the seriousness of the underlying condition.
Classically,the headache from aneurysmalrupture develops in seconds. Thereforeit is important to make specific enquiries about how quicklythe headache developed; patients often complain only about theseverity of the headache and do not know that the speed of onsetis a pivotal piece of information. However, even an accuratehistory does not reliably distinguish between aneurysmalruptureand innocuous forms of headache, such as benign vascular headacheor a muscle contraction headache. First, only half the patientswith aneurysm rupture describe the onset as instantaneous, theother half describe it as coming on in seconds to even a fewminutes. Secondly, in the group of patientswhose headache came on within a split second, innocuous formsof headache outnumber SAH by 10 to one. Other features are equally unhelpful in making the distinction:the severity of headache is rated similar, vomiting occurs in70% of patients with aneurysmalrupture, but also in 43% ofpatients with innocuous thunderclap headache. Also, precedingbouts of similar headaches are recalled in 20% of patients withaneurysmalrupture and 15% of patients with innocuous thunderclapheadache. Neck stiffness is a common signin SAH of any cause, but takes hours to develop and thereforecannot be used to exclude the diagnosis if a patient is seensoon after the sudden-onset headache. It does not occur if patientsare in deep coma.
If explosive headache is the only symptom, the chance of SAHbeing the cause is only 10%. Nevertheless,the lack of clinical features that distinguish reliably andat an early stage between SAH and innocuous types of suddenheadache necessitate a brief consultation in hospital for allpatients with an episode of severe headache that comes on withinminutes. Such an approach serves the patient’s best interestsand is also cost effective. The discomfort and cost of referringthe 90% of patients with innocuous headache is outweighed byavoidance of the disaster in the other 10% so that a rupturedaneurysm is avoided. 
It is even more difficult to suspect aneurysmalrupture if thepatient does not report a history of sudden headache, or ifother symptoms seem to prevail over the headache, such as inpatients presenting with a seizure or a confusional state, orif there is an associated head trauma. Epileptic seizures atthe onset of aneurysmalSAH occur in ~6–16% of patients. Of course the majority of patients with de novo epilepsy aboveage 25 years will have underlying conditions other than SAH,but the diagnosis should be suspected if the post-ictal headacheis unusually severe. One to 2% of patients with SAH presentwith an acute confusional state and in most such patients ahistory of sudden headache is lacking. The differential diagnosis of acute confusional state is extensiveand SAH accounts for, at most, a few percent of all patientsseen in an emergency ward because of an acute confusional state. In such patients, the diagnosis emergesonly if the careful history of an eyewitness reveals the suddenonset of the symptoms; also detection of focal deficits or absenceof a psychiatric history should raise the index of suspicionand lead to a brain imaging study.
Trauma and spontaneous SAH are sometimes difficult to disentangle.Patients may be found alone after having been beaten in a brawlor hit by a drunken driver who made away, without external woundsto indicate an accident, with a decreased level of consciousnessor with retrograde amnesia, making it impossible to obtain ahistory and with neck stiffness, causing the patient to be workedup for SAH. Conversely, patients may cause an accident whilstriding a bicycle or driving a car at time of the aneurysmalrupture. The diagnostic conundrum is particularly difficultwhen patients sustain a skull fracture having fallen after aneurysmrupture or when head trauma causes an aneurysmto burst. Meticulous reconstructionof traffic or sports accidents may therefore be rewarding, especiallyin patients with disproportionate headache or neck stiffness.
Patient may have confusion, decreased level of consciousness or coma, neck stiffness and other signs of meningism.<href=”#cite_note-vanGijn-0> Neck stiffness usually presents six hours after initial onset of SAH.<href=”#cite_note-OTV3-5> Localizingneurologic signs include third-nervepalsy (posterior communicating aneurysm), sixth-nerve palsy(increased intracranial pressure), bilateral lower-extremityweakness or abulia (anterior communicating aneurysm), and thecombination of hemi paresis and aphasia or visuospatial neglect(middle cerebral-artery aneurysm).
Intraocular haemorrhage (bleeding into the eyeball) may occur in response to the raised pressure: subhyaloid haemorrhage (bleeding under the hyaloid membrane, which envelops the vitreous body of the eye) and vitreous haemorrhage may be visible on fundoscopy. This is known as Terson syndrome (occurring in 3–13% of cases) and is more common in more severe SAH.<href=”#cite_note-mccarron-6>
As a result of the bleeding, the body releases large amounts of adrenaline and similar hormones. This leads to a sharp increase in the blood pressure; the heart comes under substantial strain and neurogenic pulmonary edema (accumulation of fluid in the lungs), cardiac arrhythmias (irregularities in the heart rate and rhythm)<href=”#cite_note-OHA-7> and cardiac arrest (in 3% of cases) may occur rapidly after the onset of haemorrhage.[2,<href=”#cite_note-8>42]
Clinical clues to the cause of SAH
Past history may contain useful information. Patients may report symptoms consistent with a minor haemorrhagebefore a major rupture, which has been called a sentinel bleedor warning leak.  The majority of these minor haemorrhagesoccur within 2 to 8 weeks before overt SAH. The headache associatedwith a warning leak is usually milder than that associated witha major rupture, but it may last for a few days. [44, 45] Nauseaand vomiting may occur, but meningismus is uncommon after asentinel haemorrhage.
In patients withprevious head injury, and particularly with a skull fracture,a Dural arteriovenous malformation (AVM) should be suspected,since healing of the fracture may be accompanied by the developmentof such a malformation. Although SAHfrom a septic aneurysm is a rare presentation of infective endocarditicin patients not known to have a disorder of the heart valves,[47,48] this diagnosisshould be considered in patients with a history of malaise inthe days or weeks preceding the haemorrhage, even more so ifthe haemorrhage is located at the convexity of the brain. Usuallyit will not be hard for the physician to get acquainted withthe existence of sickle cell disease, a history of cardiac myxoma,or coagulation disorders. Pain at onset in the lower part ofthe neck (upper neck pain is common also with ruptured intracranialaneurysms), or a sudden and stabbing pain between the shoulderblades (coup de poignard or dagger thrust), with or withoutradiation to the arms, suggests a spinal AVM or fistula as thesource of SAH.A history of even quiteminor neck trauma or of sudden, unusual head movements beforethe onset of headache may provide a clue to the diagnosis ofvertebral artery dissection as a cause of SAH. Cocaine ingestionas a risk factor may not immediately be known in the case ofan unconscious patient. Even if the family turns up in largenumbers, one may find that not every relative is aware of illicitdrugs being used or willing to volunteer this information evenif they are. In cocaine-associated SAH there is often an underlyinganeurysm. [50, 51]
The physical examination can also provide an indication aboutthe cause of SAH. Monocular blindness may result from anteriorcommunicating artery aneurysms if it is exceptionally large. Complete or partial third nerve palsy isa well-recognized sign after rupture of an aneurysm of the internalcarotid artery at the origin of the posterior communicatingartery. The third nerve can alsobe involved with aneurysms of the basilar bifurcation or thesuperior cerebellar artery, but these are relatively infrequentsites. Sixth nerve palsies, oftenbilateral in the acute stage, usually result from a non-specificand sustained rise of cerebrospinal fluid pressure, either atthe time of rupture or later. A combination of visual and occulomotordeficits should raise the suspicion of a pituitary apoplexy.  Usually, the underlying adenoma hasinsidiously manifested itself before the dramatic occurrenceof the haemorrhage by a dull retro-orbital pain, fatigue, gradual decrease of visual acuity or a constriction of the temporalfields. Lower cranial nerve palsies point to dissection of thevertebral artery, through direct compression of the ninth ortenth nerve. Lower cranial nerve palsies(ninth to twelfth nerve) may also accompany dissection of thecarotid artery in the neck, but this is an extremely uncommoncause of SAH. Deficits indicatinglesions of the cerebellum or brainstem, such as dysmetria, scanningspeech, rotatory nystagmus or Horner’s syndrome, also stronglysuggest vertebral artery dissection. Thepresence or absence of hemiparesis does not contribute muchto the diagnosis of uncommon causes, because the rare occurrenceof hemiparesis with a ruptured aneurysm (mostly of the middlecerebral artery) will still outnumber all other potential causesof SAH, in which hemiparesis may be relatively common, for examplewith septic aneurysms.
Missed diagnosis &deferential diagnosis
In the absence of the classic signs and symptoms, subarachnoidhaemorrhage may be misdiagnosed. [59, 60] The frequency of misdiagnosismay be up to 50 percent in patients presenting for their firstvisit to a physician. The common incorrect diagnoses are migraineand tension-type headaches. Failure to obtain the appropriateimaging study accounts for 73 percent of cases of misdiagnosis,and failure to perform or correctly interpret the results ofa lumbar puncture accounts for 23 percent. Misdiagnosed patientstend to be less ill and have a normal neurologic examination.However, in such cases, neurologic complications occur laterin as many as 50 percent of patients, and these patients havean associated higher risk of death and disability. [59, 60]
Up to 40 percent of patientsheadachemay be the only presenting symptom and may abate completely within minutes or hours;  these arecalled sentinel or thunderclap headaches or “warning leaks.”Emergency evaluation of sentinel headaches is required sincepatients may have a serious subarachnoid haemorrhage within threeweeks.  In many instances, no reliable clinical features distinguisha sentinel headache from a benign headache.
As only 10% of people admitted to the emergency department with a thunderclap headache are suffering from an SAH, other possible causes are usually considered simultaneously, such as meningitis, migraine, tension headache and cerebral venous sinus thrombosis. Intracerebral haemorrhage, in which bleeding occurs within the brain itself, is twice as common as SAH and is often misdiagnosed as the latter.<href=”#cite_note-Teunissen96-13>
Brain scanning (CT and MRI)
If SAH is suspected, CT scanning is the first line in investigationbecause of the characteristically hyperdense appearance of extravagatedblood in the basal cisterns. The pattern of haemorrhage oftensuggests the location of any underlying aneurysm, although with variable degrees of certainty. Head CT scanning can also demonstrate intraparenchymalhematomas, hydrocephalus, and cerebral edema .
This has a high sensitivity and will correctly identify over 95% of cases—especially on the first day after the onset of bleeding.  Becauseof rapid clearance of blood, delayed head CT scanning may benormal despite a suggestive history. In the first12 hours after SAH, the sensitivity of CT for SAH is 98% to100%, declining to 93% at 24 hours [68-72] and to 57% to85% 6 days after SAH. [73, 74]
Fig.2.CT scans of the brain showing subarachnoid haemorrhage as a white area in the center
A false-positive diagnosis of SAHon CT is possible in the presence of generalized brain edema,with or without brain death, which causes venous congestionin the subarachnoid space and in this way may mimic SAH.  The CT scanshould be carefully scrutinized because small amounts of subarachnoidblood may easily be overlooked (Fig.3). If after a thoroughreview no blood is found, aneurysmal SAH cannot be excludedeven if CT is performed within 12 h because studies are negative in ~2% of patientswith SAH. 
Fig. 3 Sedimentation in the left occipital horn as the only sign of SAH on CT.
Brain CT may also help in distinguishing primary SAH from traumaticbrain injury, but the aneurysmalpattern of haemorrhage is notalways immediately appreciated in patients admitted with a trauma. If trauma is the cause of SAH, the bloodis usually confined to the superficial sulci at the convexityof the brain, adjacent to a fracture or to an intracerebralcontusion; these findings dispel any lingering concern aboutthe possibility of a ruptured aneurysm. Nevertheless, patientswith basal-frontal contusions may show a pattern of haemorrhageresembling that of a ruptured anterior communicating arteryaneurysm, and in patients with blood confinedto the sylvian fissure or ambient cistern it may also be difficultto distinguish trauma from aneurysmalrupture by the patternof haemorrhage alone. In patients withdirect trauma to the neck or with head injury associated withvigorous neck movement, the trauma can immediately be followedby massive haemorrhage into the basal cisterns resulting froma tear or even a complete rupture of one of the arteries ofthe posterior circulation, which is often rapidly fatal.[80,81]
MRI with FLAIR (fluid attenuated inversion recovery) techniquesdemonstrates SAH in the acute phase as reliably as CT, but MRI is impracticable because the facilitiesare less readily available than CT scanners, and restless patientscannot be studied unless anaesthesia is given. After a few days(up to 40), however, MRI is increasingly superior to CT in detectingextravagated blood.[83,84] This makes MRI a unique method for identifying the site of thehaemorrhage in patients with a negative CT scan but a positivelumbar puncture (see below), such as those who are not referreduntil 1 or 2 weeks after symptom onset.
Lumbar puncture, in which cerebrospinal fluid (CSF) is removed with a needle from the lumbar sac, will show evidence of haemorrhage in 3% of people in whom CT was found normal; lumbar puncture is therefore regarded as mandatory in people with suspected SAH if imaging is negative.<href=”#cite_note-vanGijn-0> At least three tubes of CSF are collected.<href=”#cite_note-OTV3-5> If an elevated number of red blood cells is present equally in all bottles, this indicates a subarachnoid haemorrhage. If the number of cells decreases per bottle, it is more likely that it is due to damage to a small blood vessel during the procedure (known as a “traumatic tap”).<href=”#cite_note-Suarez-2> The CSF sample is also examined for xanthochromia—the yellow appearance of centrifugated fluid. More sensitive is spectrophotometry (measuring the absorption of particular wavelengths of light) for detection of bilirubin, a breakdown product of hemoglobin from red blood cells.[2,88] Xanthochromia and spectrophotometry remain reliable ways to detect SAH several days after the onset of headache.<href=”#cite_note-Cruickshank-12> An interval of at least 12 hours between the onset of the headache and lumbar puncture is required, as it takes several hours for the hemoglobin from the red blood cells to be metabolized into bilirubin.[2,88]
A normal CTscan and CSF examination exclude a subarachnoid haemorrhage and predict a more favorable prognosis in the setting of severeand/or sudden headache.[89,90]It has been recommended thatpatients with a normal CT scan and CSF examination be offeredreassurance, symptomatic headache treatment, and appropriateconsultative referral as indicated.
Transcranial droppler ultrasound assessment of proximal middle, anterior and posterior cerebral and basilar artery flow is helpful in detecting the onset of vasospasm, even prior to symptoms and following its course and response to therapy.  Skull radiographs sometimes reveal calcification in the AVM or increased vascular markings in the overlying bone.
Associated Systemic Changes
Acute subarachnoid haemorrhage is associated with several characteristic responses in the systemic circulation, water balance, and cardiac function. The ECG changes include symmetrically large peaked T waves and other alterations suggesting subendocardial ischemia. Also there is a tendency to develop hyponatremia; this abnormality and its relationship to intravascular volume depletion play a key role in treatment. Albuminuria and glycosuria may be present for a few days. Rarely, diabetes insipidus occurs in the acute stages, but water retention or a natriuresis is more frequent. There may be a leukocytosis of 15,000 to 18,000 cells per cubic millimeter, but the sedimentation rate is usually normal. 
According to the AHA/ASA guideline published in 2009
Manifestations and Diagnosis of SAH: Summary and
- SAH is a medical emergency that is frequently misdiagnosed.A high level of suspicion for SAH should exist in patients withacute onset of severe headache (Class I, Level of Evidence B).
- CT scanning for suspected SAH should be performed (Class I,Level of Evidence B), and lumbar puncture for analysis of CSFis strongly recommended when the CT scan is negative (ClassI, Level of Evidence B).
- Selective cerebral angiography shouldbe performed in patientswith SAH to document the presence andanatomic features of aneurysms(Class I, Level ofEvidence B).
- MRA and CTA may be considered when conventional angiographycannot be performed in a timely fashion (Class IIb, Level ofEvidence B).
The main cause
Approximately 85% of all spontaneous haemorrhages into the subarachnoidspace arise from rupture of saccular aneurysms at the base ofthe brain. [96-98] such aneurysms are not congenital, butdevelop during the course of life. Cerebral aneurysms almostnever occur in neonates and they are also rare in children.  In those exceptional cases, there is usually a specificunderlying cause for the aneurysm, such as trauma, infectionor connective-tissue disorder. [100,101]
Approximately 90 to 95 percent of saccular aneurysm lies on the anterior part of the circle of Willis. The four most common sites are
1. The proximal portions of the anterior communicating artery.
2. At the origin of the posterior communicating artery from the stem of the internal carotid.
3. At the first major bifurcation of the middle cerebral artery.
4. At the bifurcation of the internal carotid into middle and anterior cerebral arteries.
Other sites include the internal carotid artery in the cavernous sinus, at the origin of the ophthalmic artery, the junction of the posterior communicating and posterior cerebral arteries, the bifurcation of the basilar artery, and the origins of the three cerebeller arteries. Aneurysms that rupture in the cavernous sinus may give rise to an arteriovenous fistula.
Fig.4. Common sites of berry aneurysms in the circle of Willis
Source: Robin’s Patholgy (6th edn)
The etiology and pathogenesis of intracranial aneurysms are clearly multifactorial.It is largely unknown why only some adults develop aneurysmsat arterial bifurcations and most do not. The once popular notionof a congenital defect in the muscle layer of the wall (tunicamedia) being a weak spot through which the inner layers of thearterial wall would bulge has been largely dispelled by a numberof contradictory observations. First, gaps in the muscle layerof intracranial arteries are equally common in patients withand without aneurysms and are usually strengthenedby densely packed collagen fibrils.[105,106] Secondly, if an aneurysm has formed, any defectin the muscle layer is located not at the neck of the aneurysm,but somewhere in the wall of the aneurysmalsac.
Evidence suggests that both genetic and environmental factors contribute to the development of saccular aneurysms.A role of acquired changes in the arterial wall is likely becausehypertension, smoking and alcohol abuse are risk factors forSAH in general. It may well be theinfluence of these factors that leads to local thickening ofthe intimal layer (intimal pads’) in the arterial wall, distaland proximal to a branching site, changes that some investigatorsregard as the earliest stage in the formation of aneurysms.[109,110] The formation of these pads,in which the intimal layer is inelastic, may cause increasedstrain in the more elastic portions of the vessel wall. Also, structural abnormalities in structural proteinsof the extracellular matrix have been identified in the arterialwall at a distance from the aneurysm itself.
Some neoplastic conditions may lead to the formation of aneurysms,i.e. cerebellar haemangioblastoma ormetastasis from bronchial carcinoma. Iatrogenic causes include radiation therapy, acrylate applied externally for microvascular decompression and operation for a superficial temporalartery-middle cerebral artery bypass, with the aneurysm at thesite of the anastomosis.
An unruptured berry aneurysm is a thin-walled out pouching at arterial branch points along the circle of Willis or major vessels just beyond. Berry aneurysms measure a few mm to 2 to 3 cm and have a bright red, shiny surface and a thin translucent wall. Athermanous plaques, calcification, or thrombotic occlusion of the sac may be found in the wall or lumen of the aneurysm. Brownish discoloration of the adjacent brain and meanings is evidence of prior haemorrhage. The neck of the aneurysm may be either wide or narrow. Rupture usually occurs at the apex of the sac with extravasations of blood into the subarachnoid space, the substances of the brain, or both. The arterial wall adjacent to the neck of the aneurysm often shows some intimal thicking and gradual attenuation of the media as it approaches the neck. At the neck of the aneurysm, the muscular wall and intimal elastic lamina are usually absent or fragmented, and the wall of the sac is made up of thickened hyalinized intima. The adventitia covering the sac is continuous with that of the parent artery.
Risk of rupture: Between 3.6 and 6% of the population harbour an unruptured intracranial aneurysm. Risk of rupture is related to aneurysm site and size and whether or not the patient has already had a subarachnoid haemorrhage from another aneurysm. In ISUIA 2, the rupture rate for anterior circulation aneurysms <7 mm was 0% per year in patients with no prior sub arachnoids haemorrhage and 0.3% per year in patients with previous subarachnoid haemorrhage; 7-12 mm aneurysms, 0.5% per year (both groups); 13-24 mm aneurysms, 3% per year; and gaint aneurysms 8% per year. Rupture rate for posterior circulation is higher at all sizes; <7mm was 0.5% per year in subjects with no prior SAH, 0.7% in those with prior SAH; 7-12 mm, 3% per year; 13-24 mm, 3.7% per year; and giant aneurysms, 10% per year.
Non-invasive tests like contrast enhanced magnetic resonance angiography (MRA) and multislice computered tomographic angiography (CTA) are alternatives to intra-arterial digital subtraction angiography to detect aneurysms. Although there are promising techniques, the quality of data testing their accuracy remains limited and single slice CTA and time-of-flight MRA are poorer at detecting aneurysms <5mm diameter, which account for up to 1/3 of unroptured aneurysms. 
Management: For ruptured aneurysms, the only large scale randomized controlled trial comparing surgical and endovascular treatment (ISAT) by coiling, resulted in an absolute 8.8% reduction (updated figure as of June 2003 for 1888) in death or dependency at 1 year compared with surgical clipping.  For unruptured aneurysms, the best available data so far comparing coiling and clipping is form the prospective (but non randomized) arm of ISUIA.  Elective surgical clipping had combined morbidity and mortality at 1 year of 12.2% versus 9.5% for coiling, although the groups were not matched with more high risks patients in the endovascular treatment cohorts.  Nevertheless these data are encouraging for future long term durability of coiling treatment and the fact that complete aneurysm occlusion is not always achieved remain obstacles to its wider use in unruptured aneurysms.
Screening: There is an increased risk of SAH in relative of patients with SAH (highest in those with two or more first degree relatives affected), but most SAH is sporadic and therefore the balance of available evidence indicates that mass screening for aneurysms is not cost effective. There may be a limited role for investigation of high-risk subgroups and ideally such screening should be tested in a randomized trial. The avoidance and active management of vascular risk factors should also be part of the management of at risk subjects. 
Detection of ruptured aneurysms
MR angiography in SAH has evolved over the past decade but has not replacedcatheter-based angiography as the initial test for aneurysmidentification and localization. MRA is safe, but less suitable in the acute stage, becausein the acute stage patients are often restless or need extensivemonitoring. Factorssuch as aneurysm size, acquisition sequences used, and the typeof post processed images used for MRA interpretation can influenceMRA results. The sensitivity of 3-dimensional time-of-flightMRA for cerebral aneurysms is between 55% and 93%.[122-125] With aneurysms 5 mm, the sensitivity is 85%to 100%, whereas the sensitivity of MRA for detecting aneurysms<5 mm drops to 56%.[122,123,126,127] MRA also has limitationsin the characterization of the aneurysm neck and its relationshipto the parent vessels. MRA does not require iodinated contrastand ionizing radiation. This may be helpful in the evaluationof patients during pregnancy. MRA may also be an acceptablemodality for initial screening in patients without SAH, as describedabove. [128,129]
CT angiography is a rapid, readily available, less invasive alternativeto catheter angiography and has demonstrated sensitivities approachingequivalence to catheter angiography for larger aneurysms. Thetechnique uses a rapid intravenous injection of iodinated contrastwith image acquisition during the arterial phase in the areaof interest. Images from a CTA should extend from just belowthe foramen magnum to above the circle of Willis and middlecerebral artery bifurcation. The success of CTA depends in parton imaging through the area of interest during maximal contrastdose. Post–image processing techniques can provide valuable3-dimensional information for developing treatment strategies.Interpretation of CTA should not be based on reconstructed imagesalone. The source images should be the major basis of interpretation,and the 3-dimensional reconstructed images should be used toclarify specific questions. CTA has a reported sensitivityfor aneurysms between 77% and 100% and a specificity between79% and 100%.[131-137] The sensitivity and specificityof CTA for aneurysm detection depend on aneurysm location andsize, radiologist experience, image acquisition, and the presentationof the images. For aneurysms 5 mm, CTA has a sensitivity between95% and 100% compared with between 64% and 83% when aneurysmsare <5 mm. [131-137] Vessel tortuosity decreasesthe specificity of CTA, leading to misinterpretation as an intracranialaneurysm. This occurs most frequently in the region of the middlecerebral artery bifurcation, anterior communicating artery,and the posterior inferior cerebellar arteries. Radiologistexperience is an important factor in the practical accuracyof CTA in detecting cerebral aneurysms. The sensitivity andspecificity for the detection of cerebral aneurysms are increasedwith more experienced observers.[131,132] Among aneurysms detectedon CTA and then undergoing surgery, 100% correlation was observedbetween CTA and catheter angiography.[132,138] Velthuis and colleagues found that CTA is equal to catheter angiography in 80% to 83%of cases. In 74% of patients, catheter angiography performedafter CTA did not reveal any additional information. Fromthese data, many neurosurgeons operate on the basis of CTA alonein cases in which the risk of delaying surgery for a catheterstudy is not justified. A smaller number of neurosurgeons haveused these data to justify routine surgery on CTA alone. 
CTA can also be used to supplement information obtained by catheterangiography. CTA is better able to define aneurysmalwall calcification,intraluminal aneurysm thrombosis, orientation of aneurysm withrespect to intraparenchymal haemorrhage, and the relationshipof the aneurysm with bony landmarks. CTA has been shown to beeffective in determining the presence of severe vasospasm butis less accurate in detecting mild and moderate vasospasm. CTA has advantages related to rapid image acquisition and itswidespread availability, which can make it suitable for criticallyill patients. Disadvantages of CTA include the need for iodinatedcontrast dye administration, the possibility of bony artifactthat interferes with image quality, and the inability to studysmall distal vessels. Artifact interference from metal limitsthe use of CTA in patients with previous aneurysm clips or coils.The use of CTA continues to evolve, and in the future, CTA willincreasingly supplement or selectively replace conventionalangiography in the management of acute SAH. 
The gold standard for detecting aneurysms is conventional angiography,but this procedure can be time consuming and it is not an innocuousprocedure. Theaneurysm may re-rupture during the procedure, as occurs in 1–2%of cases overall. The rupture rate in the 6 h period followingangiography has been estimated at 5%.
Causes other than saccular aneurysms
Perimesencephalic haemorrhage constitutes ~10% of all episodesof SAH and two-thirds of those with a normal angiogram.[143-148] In this radiologically distinct and strikingly harmlessvariety of SAH, the extravagated blood is confined to the cisternsaround the midbrain, and the centre of the bleeding is immediatelyanterior to the midbrain (Fig. 5).[149-151] In some cases, theonly evidence of blood is found anterior to the pons. For this reason some have proposed the term pre-truncalhaemorrhage, but in other patientsthe blood is found mainly in the ambient cistern (Fig. 6) oronly in the quadrigeminal cistern.[153,154,155] There is no extensionof the haemorrhage to the lateral Sylvain fissures or to theanterior part of the interhemispheric fissure. Some sedimentationof blood in the posterior horns of the lateral ventricles mayoccur, but frank intraventricular haemorrhage or extension ofthe haemorrhage into the brain parenchyma indicates arterialhaemorrhage and rules out this particular condition. this disease entity is defined only by the characteristicdistribution of the extravagated blood on brain CT, in combinationwith the absence of an aneurysm.
Fig. 5 Upper panels: a typical perimesencephalic pattern of haemorrhage. The centre of the bleeding is in the interpeduncular cistern; the haemorrhage extends into both ambient cisterns and the basal parts of the sylvian fissure, but not into the lateral parts of the Sylvain fissures or the anterior interhemispheric fissure. The angiogram shows neither basilar aneurysm, nor a vertebral artery aneurysm on the right. Angiography of the left vertebral artery was also normal (not shown). Lower panels: a patient with the centre of the haemorrhage in the interpeduncular cistern, but with extension into the lateral part of the sylvian fissures and into the anterior interhemispheric fissure. CT angiography shows a basilar tip aneurysm.
Fig. 6 Perimesencephalic haemorrhage, mainly in the ambient cistern.
Perimesencephalic haemorrhage can occur in any patient overthe age of 20 years, but most patients are in their sixth decade,as with aneurysmal haemorrhage. A history of hypertension wasobtained more often than expected in a single study,  but not in another. Inone-third of the patients, strenuous activities immediatelyprecede the onset of symptoms, a proportion similar to thatfound in aneurysmal haemorrhage.[143,159]
Clinically, there is little to distinguish idiopathic perimesencephalichaemorrhage from aneurysmal haemorrhage. The headache onsetis more often gradual (minutes rather than seconds) than withaneurysmal haemorrhage, [143,159] but the predictive value of this feature is poor. Lossof consciousness and focal symptoms are exceptional and thenonly transient; a seizure at onset virtually rules out the diagnosis.  On admission, all patients are, in fact,in perfect clinical condition, apart from their headache.[143,157] Transient amnesiais found in about one-third and is associated with enlargementof the temporal horns on the initial CT scan. Typically, the early course is uneventful: rebleeds and delayedcerebral ischaemia simply do not occur. Approximately 20% ofpatients have enlarged lateral ventricles on their admissionbrain CT scan, associated with extravasations of blood in allperimesencephalic cisterns, which probably causes blockage ofthe CSF circulation at the tentorial hiatus. Only few have symptoms from this ventricular dilatationand even then an excellent outcome can be anticipated.[161,162] The period of convalescence is short andalmost invariably patients are able to resume their previouswork and other activities.[162,163] Rebleeds after the hospital period have not beendocumented thus far and the quality of life in the long term is excellent.
A perimesencephalic pattern of haemorrhage may occasionally(in 2.5–5% of cases) be caused by rupture of a posteriorfossa aneurysm.[165,166] The chance of finding an aneurysmin 5% of patients has to be weighed against the risks of complicationsfrom angiography imposed upon the remaining 95% of patients.In recent years, CTA has been studied as a method to confirmor exclude the presence of an aneurysm in patients with a perimesencephalicpattern of haemorrhage on CT. In a prospectively collected seriesof 40 patients with either a perimesencephalic haemorrhage ora posterior circulation aneurysm in whom CTA and conventionalangiography were performed, radiologists detected an aneurysmin 16 patients and no aneurysm in the remaining 24 patients.These findings were confirmed after reading the angiograms.  A formal decision analysis based onthese observations indicated that a strategy where CTA is performedand not followed by conventional angiography, if negative, resultsin a better utility than a strategy where CTA is followed byconventional angiography or if all patients are initially investigatedby conventional angiography (Y. M. Ruigrok, G. J. E. Rinkel,E. Buskens, B. K. Velthuis and J. van Gijn, unpublished data).
Dissection, in general, tends to be recognized more often inthe carotid than in the vertebral artery, but SAH from a dissectedartery occurs mostly in the vertebral artery (Fig. 7).[168,169] It is unknown what preciseproportion of all SAH cases arise from a dissected vertebralartery. All miscellaneous causes together account for only ~5%,against 85% for aneurysmalhaemorrhages and 10% for idiopathicperimesencephalic haemorrhages. In a post-mortem study of fatalSAH, dissection was found in five of 110 patients 
Fig. 7 Subarachnoid haemorrhage from dissection of a vertebral artery. CT angiogram on the day of admission shows irregular narrowing of the left vertebral artery. Intra-arterial angiography 1 week later shows absence of retrograde filling on injection of the right vertebral artery (lower left panel) and a string sign on injection of the left vertebral artery (lower centre and right panels).
Neurological deficits that may accompany SAH from vertebralartery dissection are palsies of the ninth and tenth cranialnerves, by subadventitial dissection,  or Wallenberg’s syndrome.  Rebreeds occurin between 30 and 70% of cases. [172-174] Theinterval can be as short as a few hours or as long as a fewweeks. The second episode is fatal in approximately half ofthe patients.
Dissection of the intracranial portion of the internal carotidartery or one of its branches as a cause of SAH is much lesscommon than with the vertebral artery. Reported cases have affectedthe terminal portion of the internal carotid artery, [175,176] the middle cerebral artery  and the anterior cerebral artery.
Subarachnoid bleeding at the convexity of the brain may occurfrom superficial AVMs, but only in <5% of all ruptured AVMsare the extravasations only in the subarachnoid space, withoutintracerebral haematoma (Fig. 8).  Saccular aneurysmsform on feeding arteries of 10–20% of AVMs, presumablybecause of the greatly increased flow and the attendant strainon the arterial wall. If bleeding occurs in these cases, itis more often from the aneurysm than from the malformation.In those cases the site of the aneurysms is different from theclassical sites of saccular aneurysms on the circle of Willisand again the haemorrhage is more often into the brain itselfthan into the subarachnoid space. [180,181]
Fig. 8 Subarachnoid haemorrhage from an arteriovenous malformation on the left middle cerebral artery.
Dural arteriovenous fistulae
Dural arteriovenous fistulae of the tentorium can give riseto a basal haemorrhage that is indistinguishable on CT fromaneurysmalhaemorrhage (Fig. 9). The anomaly is rare and can be found from adolescenceto old age. The risk of haemorrhage from dural AVMs dependson the pattern of venous drainage. Patients with direct corticalvenous drainage have a relatively high risk, which is furtherincreased if a venous ectasia is present. Patients with drainageinto a main sinus have a low risk of haemorrhage and if no refluxoccurs into the smaller sinuses or cortical veins, it is negligible.  After a first rupture, rebleeding mayoccur; in a series of five patients presenting with SAH, threehad one or more rebleeds.
Fig. 9 Subarachnoid haemorrhage in a patient with a Dural arteriovenous malformation. Apart from this malformation no aneurysm was found.
Spinal AVMs present with SAH in ~10% of cases; in >50% ofthese patients, the first haemorrhage occurs before the ageof 20 years. Clues pointingto a cervical origin of the haemorrhage are onset with a suddenand excruciating pain in the lower part of the neck, or painradiating from the neck to the shoulders or arms.In the absence of such symptoms, the true originof the haemorrhage emerges only when spinal cord dysfunctiondevelops, after a delay that may be as short as a few hoursor as long as a few years. Rebleeds may occur, even repeatedly. CT scanning of the brain in patients with a ruptured cervicalAVM may show blood throughout the basal cisterns and ventricles. If a cervical origin of the haemorrhageis suspected, MRI or MRA angiography are the first line of investigation,because spinal angiography is impractical without localizingsigns or symptoms.
Saccular aneurysms of spinal arteries
Saccular aneurysms of spinal arteries are extremely rare, withrecorded incidents in ~12 patients. As with AVMs of the spinal cord, the clinicalfeatures of spinal SAH may be accompanied by those of a transverselesion of the cord, either partial or complete.
Cardiac myxoma are uncommon to start with, and if present theymay in exceptional cases metastasize to an intracranial artery,infiltrate the wall and thus cause an aneurysm to develop, even>1 year after operation on the primary tumour. 
Infected tissue debris entering the blood stream may lodge inthe wall of cerebral arteries and lead to aneurysmal dilatation.The traditional term `mycotic aneurysms’ refers only to fungiand should perhaps be discarded; after all, bacterial endocarditisis more common as an underlying condition than aspergillosis.Most strokes in the context of infective endocarditis are notSAH but (haemorrhagic) infarcts or intracerebral haemorrhagesfrom pyogenic arteritis. Aneurysms associated with infective endocarditisare most often located on distal branches of the middle cerebralartery, but ~10% of the aneurysms develop at more proximal sites. Therefore, rupture of a septic aneurysmcauses an intracerebral haematoma in most patients, but somehave a basal pattern of haemorrhage on CT that is very similarto that of a ruptured saccular aneurysm (Fig. 10). CT-documentedrebleeds have been reported. Usuallypatients present with clinical features of infected heart valvesbefore SAH occurs, but sometimes rupture of a septic aneurysmis the initial manifestation of infective endocarditis. Septic aneurysms can be obliteratedby surgical or endovascular treatment,[192,194] or they may resolve after adequate antibiotictherapy.
Fig. 10 Subarachnoid haemorrhage and an intracerebral haemorrhage in a patient with multiple septic aneurysms from infective endocarditis.
Septic aneurysms in patients with aspergillosis are usuallylocated on the proximal part of the basilar or carotid artery. Rupture of such an aneurysm causes a massiveSAH in the basal cisterns, indistinguishable from that of asaccular aneurysm. Aspergillosis isdifficult to diagnose, but should particularly be suspectedin patients undergoing long-term treatment with antibioticsor immunosuppressive agents. Most patients with haematogenousdissemination have pulmonary lesions, but X-ray films of thechest may be normal early in the course.  Severely HIV-infected children may develop cerebral aneurysmssecondary to generalized arteriopathy. [198,199] In HIV-infectedadults, aneurysmal SAH can also be coincidental.
The precipitating event of arterial haemorrhage occurring ina pituitary tumour is thought to be tissue necrosis, involvingone of the hypophyseal arteries. Several contributing factorsmay precipitate haemorrhagic infarction of a pituitary tumour,such as pregnancy, raised intracranial pressure, anticoagulanttreatment, cerebral angiography or the administration of gonadotrophin-releasinghormone. The initialfeatures are a sudden and severe headache, with or without nausea, vomiting, neck stiffness or adepressed level of consciousness. The hallmarkof pituitary apoplexy is that most patients have a sudden decreasein visual acuity: in one series of 15 patients, only two hadnormal visual acuity. In most patients with pituitary apoplexyeye movements are disturbed as well, because the haemorrhagecompresses the oculomotor, trochlear and abducens nerves inthe adjacent cavernous sinus.  BrainCT or MRI scanning indicate the pituitary fossa as the sourceof the haemorrhage and i