Management of subarachnoid haemorrhage · SAH and occur during haemorrhage in approximately...

J7ournal ofNeurology, Neurosurgery, and Psychiatry 1993;56:947-959

NEUROLOGICAL EMERGENCY

Management of subarachnoid haemorrhage

Thomas A Kopitnik, Duke S Samson

Overview of subarachnoid haemorrhageThe brain is unique in its structure and devel-opment. Unlike other organs, as the cerebralblood vessels penetrate the cranial cavity, thevessels form a collateral network along thebase of the brain and only the smaller vesselspenetrate the brain substance. The larger ves-sels are contained within arachnoidal cisterns,designated the subarachnoid space. This sub-arachnoid space is a well-formed fluid con-taining compartment which contains andcirculates CSF.'-' Sheets of arachnoid parti-tion the subarachnoid space into distinctchambers which provide a fragile barrier tomigration of CSF, infection, or blood,throughout the subarachnoid space. It iswithin this fragile network of arachnoidalreflections that subarachnoid haemorrhagemay occur.

Subarachnoid haemorrhage (SAH) is acondition, not a disease, that can be pro-duced by a multitude of aetiologies. The trueincidence of SAH varies considerably on ageographic basis. In the USA, SAH is thecause of death in 16 per 100 000,4 whileJapan reports rates of 25 deaths per 100 000people.5 On the other hand, Rhodesia reportsonly 3-5 deaths from SAH per 100 000people per year.6 In the 1966 CooperativeStudy of intracranial aneurysms and sub-arachnoid haemorrhage, Locksley found 50%of 2627 subjects with aneurysmal subarach-noid haemorrhage were female. Under age40, SAH occurs more commonly in males butafter the age of 50 more commonly infemales.7 In contrast to spontaneous intra-parenchymal brain haemorrhage, SAH doesnot appear to have a consistent seasonalprevalence.8 Some authors have reported onincreased incidence of SAH in the Spring andAutumn,910 while Ohno reported a peak sea-sonal incidence in Japan in the Wintermonths. 11-13The data of the Cooperative Study shows

that regardless of the aetiology, SAH mostfrequently occurs between ages 40 and 60,with the peak frequency between 55 and 60years of age. The Cooperative Study alsofound that intracranial aneurysms were thecausative factors in 54% of the initial SAH,while arteriovenous malformation (AVM)accounted for 6%, and other aetiologies theremaining 40%. The peak incidence of SAHattributed to aneurysm occurred at slightlyolder ages than that for AVM. Sixty three percent of the first haemorrhagic episodes from

AVM occurred between ages 30 and 40.14A third of patients who develop SAH do so

while they are asleep, while another third suf-fer occurrence during routine daily activities,and a third during strenuous activity.Bending and lifting activities have the highestassociation with SAH among those activitiesconsidered strenuous.'4

Death rates from the initial haemorrhagerange from 40-60%.16 Freytag reported 250consecutive deaths from SAH and found 60%of the deaths were immediate, 20% within 24hours of the haemorrhage, and only 11% ofthe patients lived over 24 hours.'5 Rupturedcerebral aneurysm is the most common causeof non-traumatic SAH, although hypertensiveSAH was the most common cause of earlydeath in the Cooperative Study. HypertensiveSAH accounted for 52% of the deaths,whereas aneurysmal SAH accounted for36%. Ninety per cent of all patients dyingwithin 72 hours had an intracranialhaematoma in the first Cooperative Study.'7The phase I report of the 1966

Cooperative Study recorded 6368 patientsexperiencing spontaneous SAH over a periodof eight years. Of these patients, 51% hadcerebral aneurysms subsequently diagnosed,while the remaining patients had some othercause for the haemorrhage. 18 Stehbensreviewed 11 series reported from 1950-69and found aneurysm as the source of SAH in18-76% of the cases.'9 Other causes includetrauma, cerebral and spinal vascular malfor-mations, intrinsic and extrinsic cranial andspinal neoplasms, pathological and iatrogeniccoagulopathy, collagen vascular disease,sickle cell anaemia, cerebral infarction, anddrug abuse. For the purposes of discussionwe will divide SAH into distinct categories ofaneurysmal and non-aneurysmal SAH.Although some overlap exists in the medicalmanagement of SAH in both categories,aneurysmal SAH presents unique surgicalmanagement circumstances that will be dis-cussed separately. We will focus the followingdiscussion on the diagnosis and treatment ofboth aneurysmal and non-aneurysmal SAH.

Diagnosis of subarachnoid haemorrhageHeadache is the most common clinical symp-tom of SAH and occurs in 85-95% ofpatients.'>2 At least one third of patients withaneurysmal SAH will have a minor leak,referred to as a sentinel haemorrhage." The

The University ofTexas, SouthwesternMedical Center,Dallas, TexasT A KopitnikD S SamsonCorrespondence to:Dr Kopitnik Jr.UT Southwestern MedicalCenter, 5323 Harry HinesBoulevard, Dallas, TX75235-8855, USA

947

on February 10, 2020 by guest. P

rotected by copyright.http://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.56.9.947 on 1 S

eptember 1993. D

ownloaded from

http://jnnp.bmj.com/

Kopitnik, Samson

sentinel haemorrhage may occur hours ordays before a major aneurysmal haemorrhage.Many authors have emphasised that a suddenminor, but unusual, headache may herald ahaemorrhage in the near future.'>'6 In 2621cases reviewed by the 1966 CooperativeStudy for premonitory symptoms, the follow-ing were present immediately before a majorSAH: headache (48%), orbital pain (7%),diplopia (4%), ptosis (3%), visual loss (4%),seizures (4%), motor or sensory deficit (6%),dysphasia (2%), bruit (3%), dizziness (10%),and other (13%).14-7 Shields noted thatminor bleeds were often misdiagnosed asinfluenza, migraine, sinusitis, headache, stiffneck, and malingering.27 Missing the diagno-sis of aneurysmal SAH may have disastrousconsequences for the patient presenting withminor neurological symptoms. Kassell stud-ied 150 consecutive patients with proven rup-tured aneurysms and found that only 38%were referred to neurosurgeons within 48hours of the first symptoms of SAH. Themost common cause for referral delay wasphysician misdiagnosis in 37%, followed byadministrative referral delays in 23% ofpatients.28 Nearly a half of all patients withSAH enrolled in a recent international coop-erative study had delays in excess of threedays from onset of SAH to transfer to a neu-rosurgical centre.29 Delay in referral to a neu-rosurgical centre seriously affects patientoutcome; therefore, efforts in the educationof primary-care physicians towards rapiddiagnosis and prompt referral seem war-ranted.

If a significant SAH occurs, the suddenonset of intense headache is the usual pre-senting symptoms, followed by pain radiatinginto the occipital or cervical region. As bloodflows into the spinal canal, cervical pain ornuchal rigidity develops. The duration andintensity of the nuchal rigidity depends on themagnitude of the SAH, although symptomsvary between patients. Signs and symptomssimilar to infectious meningitis are typicallyseen with SAH, due to an inflammatory reac-tion of the leptomeninges to the extravasationof blood. Kernig's or Lesegue's sign may bepresent if substantial meningeal irritationexists.

Other symptoms of SAH include photo-phobia, nausea, vomiting, lethargy or alteredmentation. Brief loss of consciousness occursin most patients suffering SAH, which is fol-lowed by various levels of mentation. Afterthe haemorrhage the patient may regain alert-ness and orientation, or may remain withvarying degrees of lethargy, confusion orobtundation. The altered level of conscious-ness is related to haematoma formation,hydrocephalus, increased intracranial pres-sure (ICP), vasospasm, or reduced cerebralblood flow.30 Other signs of neurologicalinvolvement may include motor or sensorydeficits, upper motor neuron reflex changes,visual field deficits, abnormal brainstemreflexes, or abnormal motor posturing.Unlikely causes of motor deficits includeemboli from the aneurysm sac, brain com-

pression by a large or giant aneurysm, orseizures. Seizures may occur at the time ofSAH and occur during haemorrhage inapproximately 10% of patients."'4

Clinical signs of SAH that often accom-pany the presenting symptoms include mildhyperpyrexia, hypertension, and ophthalmo-logical findings. Intraocular haemorrhagesmay occur in the vitreous or retina, but sub-hyaloid or preretinal haemorrhages are moreindicative of SAH." Subhyaloid haemor-rhages appear as bright red, sharply demar-cated regions adjacent to the optic disc. Othersigns may include one or more cranial nervepalsies, depending upon the location of thehaemorrhage and whether it is aneurysmal inaetiology.

Cranial nerve palsies can be seen withSAH, especially in SAH due to rupturedcerebral aneurysms. Oculomotor nerve palsyfrequently occurs with posterior carotid wallaneurysms associated with the posterior com-municating artery. An oculomotor nerve palsymay be seen less frequently with aneurysms ofthe carotid bifurcation, the posterior cerebralartery, the basilar bifurcation, and the supe-rior cerebellar artery. Third cranial nervepalsy associated with aneurysmal SAH ordirect aneurysms mass effect typically resultsin a dilated pupil, ptosis, or deficits in eyemobility. Compression of the third nervewithin the cavernous sinus may present with amidpoint pupil secondary to compression ofsympathetic fibres en route to the iris.Trigeminal nerve distribution pain can resultfrom SAH or aneurysm compression, but israre and more commonly seen from giantaneurysms within the cavernous sinus.Abducens nerve palsy is frequently seen fol-lowing SAH and is thought to be related toincreased ICP and traction on the nerve dur-ing downward brainstem herniation duringthe haemorrhage.CT scanning of the brain is the procedure

of choice to confirm the diagnosis of sub-arachnoid haemorrhage. The CT scan candemonstrate the magnitude and location ofthe SAH, give clues as to the probable loca-tion of an aneurysm, and assess ventricularsize. The success of detecting SAH with a CTscan is dependent upon the length of timeafter SAH until the scan is obtained. If theCT scan is obtained within five days of thehaemorrhage, the probability is high that thescan can confirm the diagnosis. Eighty fiveper cent of patients scanned within 48 hoursof SAH and 75% of patients scanned withinfive days will have subarachnoid blooddetectable on CT scanning.3639The distribution of blood on the CT scan

after SAH may give an indication of the prob-able location of an intracranial aneurysm.The presence of acute blood within thesupratentorial ventricular system is often dueto SAH from a ruptured anterior communi-cating artery aneurysm. With focal bloodwithin the fourth ventricle, a vertebral arteryaneurysm in the vicinity of the posterior infe-rior cerebellar artery should be suspected.Intracerebral haematomas are not uncommon

948



j.com/

J Neurol N


eptember 1993. D

ownloaded from


Management ofsubarachnoid haemorrhage

in aneurysmal SAH, and are most frequentlyseen with ruptured middle cerebral or distalanterior cerebral artery aneurysms. Inferiorfrontal lobe haematomas commonly occurwith ruptured anterior communicating arteryaneurysms and are a highly accurate CT scanfinding for localising the source of the SAH.40

Fisher has developed a grading scale forthe CT scan appearance of SAH dependentupon the severity and location of subarach-noid blood.4' In the Fisher grading system,Grade I had no blood detectable, whereasGrade II patients had a layer of blood lessthan 1 mm thick diffusely spread throughoutthe subarachnoid cisterns. Grade III patientshad CT scan appearance of SAH greater than1 mm thick, while Grade IV patients hadintraventricular or intracerebral blood. TheFisher grading system is used to relate theamount of subarachnoid blood on a CT scanto the probability of developing delayedischaemia secondary to vasospasm.

Visual examination of CSF obtained bylumbar puncture can confirm the diagnosis ofSAH when the CT scan is negative. Lumbarpuncture after SAH is not without risk. Duffyreviewed 54 patients who had lumbar punc-ture following spontaneous SAH. Thirteenper cent had significant neurological deterio-ration following lumbar puncture. Six of theseven patients who deteriorated had evidenceof brain shifts seen on follow up CT scans.42Because lumbar puncture carries the risk ofbrain herniation or aneurysm rebleeding, theprocedure should only be performed if thediagnosis remains in question following theCT scan, or when CT scanning is unavail-able. Lumbar puncture is also useful in rulingout infectious meningitis, which may mimicsymptoms of SAH. In 1901, Sicard foundthat yellow discolouration of CSF after cen-trifugation was a reliable diagnostic sign ofprevious subarachnoid haemorrhage.43 Theterm xanthochromia (xanthochromie) wasfirst used in 1902 to describe the yellowcolour of CSF in a case of pneumococcalmeningitis, and was later used in the 1920s torefer to the colour of CSF several hours afterSAH.44-48 The CSF supernatant does notdemonstrate discolouration immediately fol-lowing SAH, but only after red blood cellshaemolyse and release oxyhaemoglobin.Xanthochromia can usually be detected fourhours after SAH; it becomes maximum oneweek after haemorrhage, and is usually unde-tected at three weeks.49 If xanthochromia ispresent in the CSF, SAH has probablyoccurred. If a traumatic lumbar puncture issuspected, partial or total clearing of the CSFmay occur during collection. If bloody CSF isallowed to stand undisturbed in a test tube, aclot will usually not form in CSF bloody fromSAH. Repeat lumbar punctures hours follow-ing a traumatic tap will be of little diagnosticvalue, since blood contaminating the CSFwill also show xanthochromia.

After the diagnosis of SAH has been estab-lished, patients are assigned a clinical gradebased on one of the accepted grading sys-tems. Grading systems for SAH have been

reported since the 1 930s, when Bramwellgraded patients either apoplectic or para-lytic.50 Botterell et al introduced a useful grad-ing scale in 1956 which has undergoneseveral modifications, including one in 1973by Lougheed and Marshall.5152 One of themore universally accepted grading scales forpatients with SAH is that of Hunt and Hess(1968),53 which was later modified by Huntin 1974.54 This grading system classifiespatients as follows:Grade 0: Unruptured aneurysm without

symptomsGrade 1: Asymptomatic or minimal

headache and slight nuchal rigid-ity

Grade 1 a: No acute meningeal or brain reac-tion, but with fixed neurologicaldeficit

Grade 2: Moderate-to-severe headache,nuchal rigidity, no neurologicaldeficit other than cranial nervepalsy

Grade 3: Drowsy, confused, or mild focaldeficit

Grade 4: Stupor, moderate-to-severe hemi-paresis, possible early decerebraterigidity and vegetative distur-bances

Grade 5: Deep coma, decerebrate rigidity,moribund appearance

Both Botterell and Hunt grading scales putthe patient into the next worse grade if seri-ous systemic disease or vasospasm is present.Hunt has further specified that attention tothe name and date of the classification systemused is important to ensure comparability ofvarious patients or patient series reported.55

After SAH has been confirmed, four-vesselcerebral angiography should be performed assoon as possible. The angiographic investiga-tion should visualise all intracranial vesselswith multiple views and clearly demonstratethe origins of each posterior inferior cerebel-lar artery (PICA). We have seen SAH pro-duced from a vertebral aneurysm arisingwithin the cervical canal in association withan abnormally proximal PICA origin. Thegoals of the angiogram are to demonstrate thecause of the SAH, define the neck of ananeurysm (if possible), delineate the vesselsarising adjacent to the aneurysm, determine ifmultiple aneurysms are present, and assessthe degree of vasospasm, if present. In 1977Nibbelink reported a significant complicationrate for cerebral angiography in acute SAH."6Complications and their frequencies were:transient hemiparesis, 2%; permanent neuro-logical deficits, 2-5%; death, 2-6%; worseningof ischaemic deficit, 3%; and aneurysmalrebleeding, 1-5%. The present complicationrate for cerebral angiography should be lessthan 1%, with an experienced neuroradio-logist.57 Aneurysm rupture during angiogra-phy has been reported, but is fortunately aninfrequent occurrence.5>6MRI has not proved useful in the acute

diagnosis of SAH. MRI has, however, provedvaluable in the localisation of subarachnoidclot beyond the time the blood is detectable

949



j.com/

J Neurol N


eptember 1993. D

ownloaded from


Kopitnik, Samson

with CT scanning and to identify the mostprobable source of haemorrhage when multi-ple aneurysms are found on angiography.61We have found MRI invaluable in the evalua-tion of SAH secondary to giant intracranialaneurysms. Giant aneurysms are often par-tially thrombosed and incompletely opacifywith angiography. MRI is useful in demon-strating the magnitude and location of theaneurysm sac, which angiography fails to elu-cidate. We have used magnetic resonanceangiography to follow the size of giantaneurysm, but because of resolution limita-tions do not use it in evaluating acute SAH.

Non-aneurysmal subarachnoidhaemorrhageSeventy five per cent of patients who suffer aspontaneous SAH will be found to have acerebral aneurysm.62 Arteriovenous malfor-mation (AVM) will be discovered in 5%,while 20% of SAH patients will have variousother causes to which the haemorrhage isattributed, or no cause found. When the ini-tial angiogram does not demonstrate thecause of the SAH, further investigation withrepeat angiography is controversial. Earlierstudies found that a significant number ofpatients had an aneurysm demonstrated on asecond angiogram or at necropsy that was notevident on the initial study.6364 As angio-graphic techniques have improved, the yieldof repeat angiography has decreased. Forsterreported only one patient in whom a secondangiogram diagnosed a previously occultcerebral aneurysm out of 56 SAH patientswith initially negative studies.65 Others havereported higher diagnostic yields of 3-4%,although Suzuki reported repeat angiogramdiagnosed aneurysms in 22% of patients.6-9It has been our policy to tailor each diagnos-tic evaluation to the patient's specific find-ings. If the initial angiogram fails todemonstrate a cause of the SAH, but showsfocal vasospasm, the angiogram is repeated in5-7 days. We also advocate repeat angiogra-phy if a portion of the cerebral vasculature isnot adequately visualised on the initial studyor in patients who have a large amountof subarachnoid blood visualised on CTscanning.SAH is a condition, not a disease, that is

not amenable to immediate intervention inorder to lessen the severity of the initialhaemorrhage. The management goals inspontaneous SAH are similar to those in headtrauma, namely diagnose the condition andminimise the potential for further injury.Patients who present with non-aneurysmalSAH are typically in better neurological con-dition than those patients with SAH fromruptured aneurysms.7>7' Although the sourceof SAH remains undiscovered in 20% ofpatients, the mortality rate for this group ofpatients is less than 3%. The incidence ofrebleeding is 4% in the first six months andranges from 0-2-0-86% per year after sixmonths.74 The outcomes of a number ofreported series concerning SAH with negative

angiography have found that 80% of patientswith SAH of undetermined aetiology willhave a good outcome and return to gainfulemployment, as opposed to 50% of patientswith aneurysmal SAH.627' The patient's clini-cal status usually corresponds to the amountof subarachnoid blood present on the CTscan.68 The magnitude of the haemorrhageseen on CT scan relates to the developmentof complications secondary to the haemor-rhage. These complications frequentlyinclude cerebral vasospasm, hydrocephalus,seizures, memory disturbances, headache,and psychological disturbances. Stober hasreviewed the blood distribution on CT scansfollowing both aneurysmal and non-aneurys-mal SAH and determined that SAH ofunknown aetiology is unlikely to result inblood in the Sylvian or interhemispheric fis-sures.75 We have frequently observed thatthe interpeduncular or perimesencephaliccisterns often demonstrate focal blood collec-tions when SAH of unknown aetiologyoccurs.The differential diagnoses that must be

considered in non-aneurysmal SAH areextensive. Trauma is a frequent cause of non-aneurysmal SAH. It can be difficult to deter-mine if the SAH was the result or the cause ofthe patient's injuries. Other causes includeangiographically demonstrable and angio-graphically occult vascular malformations,coagulopathic conditions, granulomatousangitis, venous thrombosis, central nervoussystem infection, intra- and extra-axialtumours, hypertension, drug abuse, and vari-ous aetiologies within the spinal canal.7$78SAH of spinal origin occurs most commonlyfrom spinal arteriovenous malformations, butmay also be related to spinal neoplasms oruse of systemic anticoagulants.The treatment of SAH of unknown aetiol-

ogy is aimed at preventing secondary injuryand relief of symptoms. Patients are initiallygiven bed rest with close nursing observation.Blood pressure is controlled with antihyper-tensives and the patients are well hydrated.Headache and cervical pain are treated withnarcotic analgesics as needed and pro-phylactic anticonvulsants are administered.Corticosteroids in the form of dexamethasone4 mg every 6 hours are used to symptomati-cally alleviate signs of blood-induced menin-gitis. We also administer an oral calciumchannel-blocking agent, nimodipine 60 mgevery 4 hours, to reduce the effects of cere-bral vasospasm, should it occur. Other symp-tomatic treatment is initiated as necessary.Serial CT scans are performed daily for thefirst couple of days after the SAH to detecthydrocephalus or rebleeding. Hydrocephalusmay symptomatically present with headache,drowsiness, confusion or agitation. The in-cidence of acute hydrocephalus followingSAH varies extensively among reportedseries. Bohn and Hugosson found 1% oftheir patients operated on for rupturedcerebral aneurysms required shunting forhydrocephalus.7 In contrast, Modesti andBinet found a 63% incidence of abnormal

950



j.com/

J Neurol N


eptember 1993. D

ownloaded from



ventricular enlargement on CT scanningwithin 24 hours of SAH.80 In 1985, van Gijnreported a series of 174 patients who sufferedSAH. Thirty four (20%) developed acutehydrocephalus within 72.hours of the haem-orrhage. Although intraventricular blood was

closely associated with development of hydro-cephalus, the extent of cisternal haemorrhagewas not. The mortality rate among thosepatients with acute hydrocephalus was sig-nificantly higher than in those without thiscomplication.8' If hydrocephalus developsfollowing subarachnoid haemorrhage, clinicaljudgement should be used to assess the sever-

ity of the hydrocephalus and the need forCSF diversion. We have found that some

patients will transiently exhibit asymptomaticventricular enlargement following subarach-noid haemorrhage which spontaneouslyresolves. If the hydrocephalus causes clinicalmanifestations, continuous external ventricu-lar drainage is performed until the CSF clearsof haemorrhagic debris. Continuous externalventricular drainage can usually be performedwith little additional morbidity or mortality.82After the CSF clears of blood, an attempt isthen made to wean the ventricular drainageby gradual elevation of the external CSF col-lection chamber. If patients tolerate ventricu-lostomy weaning both clinically and by CTscan appearance of the ventricular size, theventricular drain is then discontinued. If theexternal ventriculostomy cannot be weanedwithout concurrent hydrocephalus, and inter-nal ventriculoperitoneal shunt is placed.

Symptomatic vasospasm occurs in a smallpercentage of patients who present with SAHof undetermined aetiology. The incidence isless than in those patients found to haveaneurysmal SAH, and is related to the magni-tude of the haemorrhage seen on the present-ing CT scan.67 We have frequently observedangiographic vasospasm in non-aneurysmalSAH, although clinically symptomaticvasospasm in this specific group of patients israre in our experience. The treatment ofsymptomatic vasospasm will be further dis-cussed in this section to follow on aneurysmalsubarachnoid haemorrhage.

Aneurysmal subarachnoid haemorrhageGENERAL CONSIDERATIONSRuptured cerebral aneurysms constitute 77%of the cases of spontaneous SAH.83 Chasonand Hindman were able to demonstrate 137cerebral aneurysms, a 5% incidence, in a

necropsy study of 2786 patients who died ofcauses unrelated to SAH. They also deter-mined that 42 per cent of the aneurysms hadruptured at some point previously.84Necropsy studies will demonstrate a higheroccurrence rate of ruptured aneurysms thanclinical or radiological studies because

aneurysm rupture is a frequent cause of sud-den death. Although many series reflectwidely variable epidemiological statistics, an

approximate occurrence rate for aneurysmrupture is 10 per 100 000 population peryear. There is an average prevalence rate of

unruptured aneurysms of 5% of the adultpopulation.

There are three basic theories to the patho-genesis of cerebral aneurysms. One theoryproposes that a congenital weakness in themuscular layer of cerebral arteries allows theintimal layer to herniate and eventually dis-tend and destroy the elastic membrane, lead-ing to outpouching of an aneurysmal sac.Other theories have attributed aneurysm for-mation to postnatal degeneration within thevessel wall that leads to deterioration of theinternal elastic lamina and resultant aneurysmformation. Others have postulated that it is acombination of congenital and degenerativeeffects that lead to aneurysm formation.4Aneurysms tend to occur at vascular bifurca-tions, although they infrequently may occurunassociated with vessel branches.85 Forbusused a rigid glass model to demonstrate thatthe point of greatest stress on the artery walloccurs at the apex of a vascular bifurcation inline with the direction of flow.86 The law ofLaplace relates wall stress to radius and trans-mural pressure and can be used to show thatas the radius of the aneurysm enlarges, signif-icantly less force is required to cause furtherenlargement of the sac.87 The average size ofruptured aneurysms is 7-5 mm. Two per centof aneurysms under 5mm rupture in contrastto 40% of those between 6 mm and 10 mmin external diameter.88 Sixty per cent ofaneurysms discovered in people under 60years of age have previously haemorrhaged.89Unruptured but symptomatic giantaneurysms carry a grave prognosis related toboth mass effect and future rupture. Thecommonly held perception that giantaneurysms do not bleed and can be managedconservatively is dangerously misleading.Between 30% and 70% of giant aneurysmsthat become symptomatic are associatedwith subarachnoid haemorrhage.9091 Approxi-mately 20% of SAH patients will have multi-ple aneurysms. Frequent sites of multipleaneurysms are the internal carotid arteries onboth sides.4 This makes it imperative todemonstrate all cerebral vessels on diagnosticangiography when SAH is investigated. Thepresence of multiple aneurysms will signifi-cantly affect the planning of surgical proce-dures aimed at aneurysm obliteration. Whenmultiple aneurysms are present, the mostproximal and the largest will be the mostlikely source of SAH.9' Aneurysms with smallsecondary outpouchings are thought to beparticularly prone to rupture, and these out-pouchings may actually be false sacs fromprevious haemorrhages.9394 DuBoulay hashypothesised that secondary aneurysm locula-tions are regions where the aneurysm wall ismost unstable, and found asymptomaticaneurysms rarely had secondary loculations.He also found the mortality rate of loculatedaneurysms to be twice that of smooth-walledlesions.95

Examinations that are helpful in determin-ing which of the multiple aneurysms is mostlikely to be the source of SAH, include thehistory of the ictal event, clinical examination,

951



j.com/

J Neurol N


eptember 1993. D

ownloaded from


Kopitnik, Samson

CT scan, angiogram, and MRI study. Thepatient may be able to lateralise the initialheadache symptoms when bilateralaneurysms are present, and the clinical exam-ination may demonstrate unilateral weaknessor cranial nerve palsy. When multipleaneurysms are diagnosed, CT localisation ofsubarachnoid blood, ventricular shift, and thesite of an intraparenchymal haematoma arehelpful findings. Local vasospasm may bepresent near the ruptured lesion on angiogra-phy. The aneurysm most likely to have bledwill be the largest, have the most irregularcontour, or have a nipple-like secondary locu-lation. Nehls found statistical evidence thataneurysms associated with the anterior com-municating complex, the basilar apex, andthe posterior inferior cerebellar artery-verte-bral junction were the most likely aneurysmsto bleed when multiple aneurysms were diag-nosed.96 MRI can be valuable in detectingsubarachnoid clot beyond the time clot is visi-ble on the CT scan and in localising thecausative source of haemorrhage in cases withmultiple aneurysms. Focal increased signalintensity is often found around a rupturedaneurysm.6'We plan the primary surgical approach to

the ruptured lesion and clip asymptomaticaneurysms that are accessible through thesurgical exposure. There appears to be apoorly understood phenomenon of enlarge-ment and rupture of previously asymptomaticaneurysms in patients who undergo surgeryto clip the ruptured lesion. Others have notedan increased frequency of asymptomaticaneurysm rupture after treatment of thesymptomatic aneurysm as well.9798 A possibleexplanation may relate to increased haemody-namic stress in the perioperative period pro-ducing an increase in the transmural pressuregradient within asymptomatic aneurysms.Haemorrhage from a previously asympto-matic aneurysm following surgery to treat theruptured aneurysm is an extremely rareoccurrence in our practice. Others havereported that postoperative volume expansionand induced hypertension after surgery for aruptured aneurysm are both safe and effica-cious, and do not appear to promote ruptureof asymptomatic lesions.99100

After the diagnosis of ruptured cerebralaneurysm has been confirmed as the cause ofSAH, treatment plans need to be considered.The treatment for SAH secondary to a rup-tured aneurysm is primarily surgical, withnonoperative treatment reserved for thosepatients in the poorest grades. Yasargil per-formed a retrospective analysis comparingsurgical to nonsurgical management inpatients with ruptured aneurysms in one spe-cific region of Switzerland. Of 624 provenruptured aneurysms, 349 (55-9%) underwentoperation with 5 (1-4%) deaths. Two hun-dred and seventy five (44 1%) patients didnot have surgery and only 4 (0-6%) patientssurvived, resulting in a mortality rate of98-5% in the nonoperated patients.'0'Therefore SAH from a ruptured cerebralaneurysm has a high mortality and nonopera-

tive management is rarely indicated.Expectant management should be confined topoor-grade patients who would not beexpected to tolerate a surgical procedure.

Patient selection and timing of surgery areimportant factors that determine outcome.The dilemma of which patient shouldundergo surgery and when surgery should beperformed, in relation to the onset of SAH orvasospasm, remains an unresolved issue.Current trends in neurosurgery and our ownresults support early operation in patientswith good clinical grades. Early surgery isperformed to secure the ruptured aneurysm,prevent rebleeding, and to remove as muchsubarachnoid clot as possible. We generallyoffer surgery to those patients in Hunt-Hassgrades55 I to III as soon as they are fullyradiologically evaluated and medically pre-pared following hospital admission. Poor-grade patients, grades IV and V55 are treatednon-operatively until their clinical conditionimproves. Diagnosis and treatment of acutehydrocephalus by external ventriculardrainage will often result in patients improv-ing by one grade early after SAH. In a similarfashion, patients who are in a poor grade, butharbour a significant intraparenchymalhaematoma, may show significant improve-ment with surgery to obliterate the aneurysmand remove the mass effect from thehaematoma.The rationale for conservative management

of poor-grade patients stems from previousstudies evaluating surgical results related tothe patient's preoperative status. Hunt andHess categorised patients into grades toinvestigate prognosis of the preoperative neu-rological status, and found that operativemortality approached 75% in poor-grade(grades IV and V) patients." The explanationfor dismal surgical results in poor-gradepatients is probably multifactorial. A combi-nation of elevated intracranial pressure,reduced cerebral blood flow, poor tissue tol-erance to manipulation, and poor tolerance totemporary occlusion, all contribute to poorsurgical results. Some have seen encouragingresults with aggressive surgical treatment ofpoor-grade patients,'02 although our experi-ence and that of others'0l has reaffirmed that,in general, poor-grade patients have poor out-comes.

After it has been determined that a patientwith aneurysmal SAH is a candidate forsurgery, timing of surgery becomes a consid-eration. The overall management morbidityand mortality must be a consideration in thesurgical planning. The major factors thatmust be considered are: 1) aneurysm rebleed-ing; 2) delayed ischaemia deficit due tovasospasm, and 3) technical considerations ofthe operative procedure. The predominanthistorical opinion has been that if surgery wasintentionally delayed for 1-2 weeks followingSAH, the surgical outcomes were much morefavourable compared with early surgery.Norlen and Olivecrona published the resultsof 100 consecutive anterior communicatingartery aneurysms managed in this fashion and

952



j.com/

J Neurol N


eptember 1993. D

ownloaded from


Management of subarachnoid haemorrhage

reported a remarkable 3% mortality rate.'03 Itwas presumably conceived in these early stud-ies that technical difficulties associated withearly surgery would negate any potential ben-efit early surgery was affording to preventrebleeding and facilitate the management ofvasospasm. There was also concern that earlysurgery could worsen the effects of vasospasmin the face of disturbed autoregulation.104'05Because of improvements in neuroanaesthe-sia, neurosurgical instrumentation and theadvent of the operative microscope, investiga-tors have readdressed the optimal timing ofaneurysm surgery.

Recent studies investigating the timing ofsurgery for ruptured intracranial aneurysmshave shown that early surgery was not signifi-cantly more difficult than delayed surgeryfrom a technical standpoint, as perceived bythe operating surgeons. Although the resultsof surgery delayed until after post-bleed day10 were superior to results of early surgery,the morbidity and mortality of rebleeding(12% at 2 weeks) and other complicationsassociated with delayed surgery, negated anybenefit of delaying the procedure. Earlysurgery did result in a decreased incidence ofaneurysm rebleeding, but did not significantlyaffect the incidence of subsequentvasospasm.'9106 The overall results of thesecontemporary cooperative studies show thatoverall management morbidity and mortalityof early versus delayed surgical therapy arenot significantly different. The possibleexception may be the patients who were alertand in excellent grade upon admission andhad the most favourable results with earlysurgery. The surgeon must weigh all factors,including his/her technical capabilities, therisk of rebleeding, and potential managementdifficulties, including vasospasm, when decid-ing on the best time to perform the surgicalprocedure.

Preoperative management of aneurysmalsubarachnoid haemorrhageWe believe in early surgical repair of the rup-tured aneurysm followed by aggressive med-ical therapy, if necessary, for vasospasm. It ispreferable for referral patients to be trans-ferred to a neurosurgical centre as soon aspossible following the ictus of SAH. Thethought that patients need to be observed fora period following SAH before transfer to aneurosurgical centre is an erroneous conceptand only serves to delay transfer unnecessarilyand place patients at risk for rebleeding. Wehave found that immediate transfer once thediagnosis of SAH is suspected provides thebest chance for optimal patient outcome. It isimperative that pertinent radiographic studiesalso be sent to the neurosurgeon along withthe patient. This is especially true forangiograms performed before transfer.Sending poor quality copies of radiographs ornot sending the complete angiogram becauseof individual hospital administrative policy,results in needless and dangerous repetitionof vital diagnostic examinations.

We admit patients with SAH directly to thesurgical intensive care unit, where furtherassessment is performed. Patients are gradedaccording to the Hunt-Hess grading sale,5354and accompanying radiographic studies areevaluated. In the presence of obvious SAH onCT scan, a cerebral angiogram is immediatelyperformed, if not previously obtained.Angiography is performed as soon as possibleafter admission, with each case individualisedfor the patient. Early angiography is beneficialfor early diagnosis, even if patients are notoffered immediate surgery. An angiogramobtained early after SAH will confirm thediagnosis, and preoperative planning orimmediate surgery can be undertaken.Preoperative preparation usually includesroutine blood tests, type and crossmatch ofpacked red blood cells, placement of centralvenous access and radial arterial monitoringcatheters, and premedication.

Preoperative medication includes adminis-tration of anticonvulsants, corticosteroids,calcium channel blockers, antihypertensivesand analgesics. Although we intially attemptto normalise the patient's intravascular vol-ume status, we do not prophylactically inducehypervolaemia. Prophylactic hypervolaemiahas been shown to be hazardous and offers noclear benefit.'07 Strict attention to blood pres-sure control before aneurysm surgery hasbeen shown to reduce the rate of aneurysmrebleeding.'08 The effect of calcium channelblockers is less clear. Reports on calciumchannel blockers show that prophylactic useof these agents does not prevent vasospasm,but may decrease the overall managementmorbidity following SAH.'09-"2 In the light ofthe referenced reports on calcium channelblocking agents, we place all patients on thisdrug immediately upon admission to the hos-pital. We have not seen a reduction in theseverity or morbidity from vasospasm sincewe added this treatment to our managementprotocol. Clearly, further studies need to beperformed on this important issue.The administration of antifibrinolytic ther-

apy designed to minimise clot lysis is contro-versial."'3 14 Epsilon-aminocaproic acid(AMICAR) is one of the more widely usedantifibrinolytics which primarily functions toinhibit the conversion of plasminoid intoplasmin. The main function of plasmin is todigest fibrin and aid in clot lysis. Intravenousinjection of AMICAR gives a peak plasmalevel 20 minutes after injection, and 75% willbe excreted unchanged in the urine within 12hours. The drug crosses the blood-brain bar-rier and achieves maximal antifibrinolyticactivity 48 hours after therapy is initiated."5Patients who receive this therapy are given 2grams per hour intravenously for 48 hours,then 1-5 grams per hour for the duration oftherapy or until surgery is performed. Reviewof the 1966 Cooperative Study found a signif-icantly decreased incidence of aneurysmrebleeding with the use of antifibrinolytics.Adams et al reported that the rate of rebleed-ing and death was approximately 21% at 14days among patients treated with bed rest

953



j.com/

J Neurol N


eptember 1993. D

ownloaded from


Kopitnik, Samson

alone following SAH. The addition of anti-fibrinolytic therapy lowered the incidence ofrebleeding and death to 10% at two weeks inthose patients studied."16 Although antifibri-nolytic therapy decreases rebleeding by 50%during the first two weeks following SAH, anincrease in management complications are amajor concern. The most frequent complica-tion of antifibrinolytic therapy is diarrhoea,occurring in 24% of patients. Communi-cating hydrocephalus is 25% more frequentwith antifibrinolytic therapy according toPark's report."'7 The greatest concern overthe use of antifibrinolytic agents has been theassociated increase in ischaemic neurologicaldeficits, that has counterbalanced its efficacyin some studies."3 Our view is that anti-fibrinolytics have little role to play in acuteaneurysmal subarachnoid haemorrhage ifsurgery is anticipated within two days ofadmission. These agents may be of value if anoperative procedure is delayed longer than 48hours.Aneurysm rebleeding is a catastrophic

event that may occur relatively soon after theinitial SAH."8 The frequency of aneurysmrebleeding is highest within the first 48 hoursfollowing the initial SAH with an averageincidence of 4%.108 "19 The mortality rate fromaneurysm rebleeding carries at least a 70%mortality rate. Surgical treatment offers thebest protection against rebleeding and earlysurgery after hospital admission and affordsthe best chance of minimising this highlylethal complication.Some neurosurgeons are concerned that

surgery soon after acute SAH is technicallydifficult due to brain swelling and obscura-tion of vital structures by acute blood withinthe subarachnoid space. Despite these con-siderations, the majority of our patients havesurgery early after SAH to minimise the inci-dence of rebleeding. Although the surgerycan be technically more demanding, we havenot found technical problems to adverselyinfluence the management morbidity andmortality of our patients. We have foundintraoperative ventricular puncture andaggressive gravity drainage of CSF to be anextremely useful adjunct in overcoming aninitially swollen brain soon after SAH.'20 Theresults of the most recent cooperative studyconfirm that although most surgeons reportthat the brain is significantly more swollenwith early surgery, the majority of surgeonsreport that early surgery is not technicallymore difficult.'06 In this cooperative study,early surgery did reduce the incidence ofrebleeding, but had no effect in decreasingthe incidence of vasospasm, which theoreti-cally might have been expected by early sub-arachnoid clot removal.'21-'23

Surgical considerations andcomplicationsThe goals of aneurysm surgery followingSAH are: (1) aneurysm obliteration withpreservation of normal vasculature; (2) mini-mal brain tissue disruption, and (3) removal

of as much subarachnoid clot as is safely fea-sible. Our patients undergoing uncomplicatedaneurysm repair following SAH had morbid-ity and mortality rates of approximately 10%.When intraoperative rupture occurred, themorbidity and mortality rates increased toapproximately 20 per cent. Factors such asthe phase of the dissection, the use of bluntversus sharp dissection techniques, and com-plete aneurysm neck dissection play key rolesin minimising intraoperative aneurysm rup-ture. Adequate depth of anaesthesia, strictblood pressure control, and appropriately sit-uated craniotomies, which necessitate mini-mal brain retraction will aid in reducing theincidence of aneurysm rupture before sub-arachnoid dissection.The most frequent time for intraoperative

aneurysm rupture is during subarachnoid dis-section before clip application.'24'26 In ourexperience, ill-advised blunt dissection tech-niques are the most frequent cause of intra-operative rupture. Aneurysm ruptureproduced by blunt dissection typically pro-duces a large tear at the aneurysm sac-neckjunction, and the amount of bleeding is usu-ally torrential. This is in contrast to aneurysmbleeding resulting from sharp dissection.Bleeding from sharp dissection is usuallypunctate and more controllable than bleedingfrom blunt dissection. The dense clot sur-rounding the vessels and the aneurysm shouldbe sharply divided with microscissors ormicroarachnoid knives. Dissection shouldalso follow the normal vasculature when dis-secting into the vicinity of the aneurysm.Proximal and distal vascular control early inthe dissection and before aneurysm dissectionis mandatory in all cases. Definitive neck dis-section before aneurysm clip application willalso reduce the likelihood of clip-inducedrupture of the aneurysm, which is similar toaneurysm rupture from blunt dissection.

Intraoperative aneurysm rupture is aninevitable complication of intracranialaneurysm surgery. Close blood pressure con-trol, basal craniotomy flaps, minimal brainretraction, strict use of sharp dissection tech-niques, dissection along normal vascularanatomy, and appropriate use of availableclips, can potentially minimise the incidenceof intraoperative rupture and provide theoptimal chance for a good surgical outcome.

In the event of intraoperative aneurysmrupture, we have found several techniquesuseful. Temporary arterial occlusion on allafferent and efferent vessels involved with theaneurysm, is one of the most useful measures.We induce electroencephalographic burstsuppression during temporary occlusion tooptimise cerebral protection during tempo-rary occlusion. We use etomidate for cerebralprotection because of its ability to suppresscerebral metabolic activity and oxygenrequirements, its rapid reversibility, and mini-mal toxicity.'27 We do not use induced hypo-tension to control intraoperative aneurysmrupture because of the theoretically deleteri-ous effects of hypotension on patients specifi-cally susceptible to either acute or delayed

954



j.com/

J Neurol N


eptember 1993. D

ownloaded from



ischaemia. Giannotta found that inducedhypotension influenced outcome negativelywhen used to control intraoperative aneurysmrupture.'26We attempt to remove as much subarach-

noid clot during surgery as is deemed safe.The variable consistency of acute and sub-acute subarachnoid blood at various stages offibrinolysis renders success of clot removalunpredictable. Previous reports have foundsome potential benefit in preventingvasospasm or lessening its severity by aggres-sive subarachnoid clot removal.'28129 Recentreports have focused on the use of topicalthrombolytic agents instilled into the sub-arachnoid space during surgery. The chiefagent being investigated is tissue-type plas-minogen activator (tPA) substance. Use oftPA has been reported to show some effect onchemical thrombolysis of subarachnoid clot inthe postoperative period.30-"33 In our limitedexperience, tPA does appear to aid clearanceof subarachnoid blood seen on CT scan.Whether this chemically induced thromboly-sis will be beneficial in decreasing the mor-bidity and mortality of vasospasm remains tobe demonstrated. The addition of tPA atsurgery carries some risk of increased bleed-ing in the operative site postoperatively.

Postoperative managementPatient care following surgery to obliterate aruptured cerebral aneurysm is complex andmust be based on many possible complica-tions. The major problems in the immediatepostoperative period include brain swelling,bleeding in the operative site, fluid and elec-trolyte disturbances, hydrocephalus, and theonset of cerebral vasospasm. Close observa-tion is necessary because most postoperativecomplications produce similar symptoms butrequire entirely different treatment. As well asthe neurological examination, frequent CTscanning, blood electrolyte determination,and transcranial Doppler (TCD) evaluationsare particularly helpful.

Patients who have had surgery to repair aruptured aneurysm are predisposed todevelop brain swelling and oedema. The sub-arachnoid clot is irritating to the brain sur-face, cerebral vascular autoregulation isdisturbed, intracranial pressure is often ele-vated, and regions of infarction related to sur-gical occlusion are all contributing factors topostoperative brain swelling. Patients are typ-ically maintained on dexamethasone, 16 mgper day, for at least one week postoperatively.Intravascular volume status is closely assessedwith either central venous access monitors orpulmonary artery catheters, depending on thepatient's condition. At the time of surgery thedegree of swelling and the necessity of frontalor temporal lobectomy are assessed. In a sim-ilar fashion, lobectomy is considered postop-eratively if clinical deterioration occursconcomitant with CT evidence of brainswelling.

Fluid and electrolyte disturbances are alsorelatively common following SAH and

surgery to repair a ruptured aneurysm.Takaku found an 8-8% incidence of elec-trolyte disturbances following aneurysmsurgery."34 Hyponatraemia is the most com-mon abnormality, occurring in 53%, whereashypernatraemia had the highest mortality rate(42%). Hyponatraemia can be due to eitherinappropriate secretion of antidiuretic hor-mone or true natriuresis due to cerebral saltwasting. Both syndromes are characterised bya decrease in plasma sodium levels and osmo-lality, associated with increased urine sodiumconcentration greater than 25 mmol per litre.Clinical differentiation of these syndromes isimportant because patients with primary saltwasting syndrome are hypovolaemic andrequire sodium and fluid replacement.Conversely, true inappropriate antidiuretichormone secretion is treated with fluidrestriction.As previously discussed, communicating

hydrocephalus both before and after surgerycan be seen in SAH patients. There havebeen some reports that early operation andsubarachnoid clot removal may decrease theincidence of postoperative hydrocephalus.135Others have proposed that preoperativeantifibrinolytics contribute to the develop-ment of hydrocephalus.1"7 Regardless of thecause, postoperative hydrocephalus should beruled out in any patient with a decline inmental status following aneurysm surgery.

Delayed ischaemic deficit secondary tocerebral vasospasm is the greatest cause ofmorbidity in patients surviving the initialSAH. Clinically significant vasospasm occursin 30% of patients with SAH from a rupturedcerebral aneurysm. Permanent neurologicaldeficit or death occurs in approximately 12%of patients who develop severe clinicalvasospasm."36 137 Cerebral vasospasm has apeak incidence around the sixth to eighth dayfollowing SAH, although it can occur at anytime following SAH up to about 14 dayspost-bleed, beyond which it is extremelyrare."38 When vasospasm develops, it may lastfor several days or up to several weeks.36 139The most reliable predictor of those patientspredisposed to develop vasospasm is theamount and distribution of subarachnoidblood on the CT scan. Thick layers of bloodin the basal cisterns carries a higher risk ofvasospasm than diffuse or focal loculations ofcisternal blood. Lobar haematomas and inter-hemispheric blood are associated with a lowrisk of vasospasm. Subarachnoid blood in theSylvian fissure appears to carry an intermedi-ate vasospasm risk.Y0' 4' Clinical vasospasmdevelops gradually over hours or days, and istypically associated with gradual, progressivedecline in neurological status. Headache,fever, and leukocytosis are often present andmay herald the onset of vasospasm beforeneurological deterioration.

At present, the mainstay of treatment forclinically significant cerebral vasospasm is theinduction of hypervolaemia and systemichypertension, often referred to as hyper-dynamic therapy. The neurological deficitsseen with vasospasm are the result of arterial

955 on F

ebruary 10, 2020 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.56.9.947 on 1 Septem

ber 1993. Dow

nloaded from


Kopitnik, Samson

narrowing and increased cerebrovascularresistance. Because autoregulation is usuallyimpaired after SAH, manoeuvres thatincrease cerebral perfusion pressure canincrease cerebral blood flow in the ischaemicregions. 142 143 Patients undergoing inducedhypertension and intravascular volumeexpansion are best treated in an intensive careunit with arterial and deep venous pressuremonitoring. An indwelling arterial catheter isused for close blood pressure monitoring, anda Swan-Ganz catheter is used to monitorpulmonary capillary wedge pressures. Trans-cutaneous pulse-oximeters are used to moni-tor oxygen saturation and signs of earlypulmonary decompensation from hyper-volaemic therapy. Fluid balance is assessedhourly, and daily chest radiographs are per-formed and inspected for signs of congestiveheart failure or pulmonary oedema.The initial therapy for symptomatic

vasospasm consists of volume expansion withplasma protein fractionate to create a positivefluid balance of 1-2 litres. The pulmonaryartery wedge pressure is usually maintainedbetween 18 and 20 mmHg, and the centralvenous pressure is kept at approximately 10mmHg. If clinical improvement is not seensoon after volume expansion, arterial bloodpressure is elevated with dopamine and typi-cally maintained with systolic pressuresbetween 180 and 220 mmHg. Kassell hasreported 58 patients treated for cerebralvasospasm with volume expansion andinduced arterial hypertension in which therewas reversal of neurological symptoms in75%. He found neurological improvement tobe permanent in 74% and temporary in 7%.99As intravascular volume is expanded, patientsmay undergo a secondary diuresis which canmake artificial elevation of the pulmonarycapillary wedge pressure difficult. Use oflow-dose vasopressin can help minimise thediuresis and help maintain an elevatedintravascular fluid volume. Hypervolumetricand hypertensive therapy is continued untilthe neurological symptoms resolve or compli-cations from therapy require re-evaluation ofthe risk/benefit ratio of continuing this type oftreatment. Complications of hyperdynamictherapy include pulmonary oedema, conges-tive heart failure, brain oedema, hypertensivecerebral haemorrhage, systemic complicationsof prolonged vasopressor use, and myocardialinfarction. Contraindications to hyperdy-namic therapy include cerebral oedema, cere-bral infarction, pulmonary oedema, adultrespiratory distress syndrome, and increasedintracranial pressure."44When hyperdynamic therapy has proven

unsuccessful or is contraindicated, we havefound other less established manoeuvres help-ful. We have found selective intra-arterialinfusion of papaverine hydrochloride (300 mgover 1 hour) into the symptomatic vascularterritory to be extremely useful in reversingangiographic vasospasm in some patients.The results of this therapy can be clinicallydramatic, but in a similar fashion may beextremely fleeting or completely unsuccessful.

Further investigation needs to be performedto clarify the role of intra-arterial infusions ofvasodilator substances for treatment of cere-bral vasospasm. Anecdotal reports agree withour experience in the limited number ofpatients who have had this therapy.'45 146

Another potentially useful adjunct in thetreatment of posthaemorrhagic cerebralvasospasm is transluminal balloon angioplastyof the large intracranial vessels. A number ofinvestigators have reported encouragingresults with the use of this technique invasospasm treatment. 145147-149 We have alsowitnessed marked improvement in cerebralcirculation following transluminal angio-plasty. We have also seen fatal complicationsof this technique due to vessel rupture andperforation. Linskey reported a fatal SAHthat was produced by rupture of residualaneurysm neck by the angioplasty catheter."50The use of such adjunctive treatment modali-ties should be undertaken with caution untilfurther safety and efficacy of these proceduresis more clearly understood.

Management after hospital dischargePatients that have survived SAH and are ulti-mately discharged from the hospital requireclose follow up care to detect and treat latentcomplications. Communicating hydrocepha-lus may develop following discharge and maybe manifest by increasing headache, lethargy,confusion, or regression of a previouslyimproving neurological status. Occasionallyfluid and electrolyte disturbances do notbecome evident until after discharge, andmay only be suspected with a patient historyof abnormal fluid intake coupled with mentalstatus changes.

Seizures following SAH occur in a smallnumber of patients. There is no evidence thatseizures during the initial haemorrhage arelikely to persist or recur later. Patients in ourpractice suffering SAH are usually placed onprophylactic anticonvulsants when they pre-sent to the hospital, and are maintained onanti-seizure medications for approximately sixmonths postoperatively. Hart was unable todemonstrate the benefit of prophylactic anti-convulsant therapy after acute SAH."3 Themost important risk factors determining thedevelopment of a seizure disorder are poorneurological grade and focal neurologicaldeficits. Our belief is that postoperativeseizures are socially stigmatising to such anextent that prophylactic anticonvulsant ther-apy for six months is of potential benefit withvery little risk of side effects.

Because latent complications may develop,our pattern is to follow patients on a monthlybasis following discharge with CT scans andlaboratory testing as necessary. Patients areweaned from anticonvulsants approximatelysix months postoperatively and are followedwith clinic visits until they are neurologicallystable for one year.

ConclusionsThe condition known as subarachnoid haem-orrhage is a complex medical event that

956 on F


http://jnnp.bmj.com

/J N

eurol Neurosurg P


ber 1993. Dow

nloaded from



affects a significant number of people eachyear. The causes can be multifactorial, butare most commonly related to bleeding froman existing cerebral aneurysm. Many patientsdo not survive the initial SAH, and the opti-mal treatment for patients that come to diag-nosis and treatment remains controversial.Diagnostic imaging techniques have signifi-cantly advanced over the past 10 years,although CT scanning and angiographyremain the diagnostic procedure of choice.Controversy continues over the optimal timeto perform aneurysm surgery following SAH.There does not appear to be any significantdifference in overall patient outcome withearly compared to delayed aneurysm surgery.Early surgery and antifibrinolytic medicationcan prevent rebleeding, but add to the man-agement morbidity of acutely ill SAHpatients. Delayed cerebral ischaemia contin-ues to be a major cause of morbidity in SAHpatients, despite medical and surgicaladjuncts such as calcium channel blockers,hyperdynamic hypervolumetric therapy, infu-sion of intra-arterial vasodilators, translumi-nal angioplasty, and use of intraoperativethrombolytic drugs. Induced hypertensionand volume expansion is the most reliablemethod of treating symptomatic vasospasm atthe present time. In spite of therapeutic andsurgical advances over the past two decades, apercentage of SAH patients will ultimatelysuccumb or be neurologically injured, as longas SAH remains a condition that is not widelypreventable.

Physicians who diagnose and managepatients with acute SAH would be welladvised to keep up with the ever-changingdevelopments in the management of thisubiquitous and catastrophic condition thatmay strike healthy and unsuspecting individu-als at any time.

1 Clemente CD. Anatomy, a regional atlas of the humanbody, 3rd ed. Baltimore; Urban and Schwarzenberg,1987:572-3.

2 Liliequist B. The subarachnoid Cisterns. An anatomicand Roentgenologic study. Acta Radiol (Stockholm)1959;suppl: 185.

3 Haines DE, Harkey HL, Al-Mefty 0. The "subdural"space: a new look at an outdated concept. Neurosurg1993;32:111-20.

4 Sahs A, Perret GE, Locksley HB, Nishioka H.Intracranial aneurysms and subarachnoid hemorrhage.Philadelphia: J B Lippincott, 1969.

5 Shokichi U, Masumichi I, Munesuke S, Masayoshi S.Subarachnoid hemorrhage as a cause of death in Japan.Z Rechtsmed 1973;72:151-60.

6 Levy LE, Rachman L, Castle WM. Spontaneous primarysubarachnoid hemorrhage in Rhodesian Africans. Aft JMed Sci 1973;4:77-86.

7 Locksley HB. Report on the cooperative study ofintracranial aneurysms and subarachnoid hemorrhage,Sect 5, P1. Natural history of subarachnoid hemor-rhage, intracranial aneurysms, and arteriovenous mal-formations. JNeurosurg 1966;25:219-39.

8 Talbot S. Epidemiological features of subarachnoid andcerebral haemorrhages. Postgrad MedJ1 1973;49:300-4.

9 Crompton MR The coroner's cerebral aneurysm: achanging animal. JfForensic Sci Soc 1975;15:57-65.

10 Murphy JP. Subarachnoid hemorrhage; intracranialaneurysm. Cerebrovascular disease. Chicago: Year Book,1954:199-241.

11 Ohno Y. Biometeorologic studies on cerebrovascular dis-eases. I. Effects of meteorologic factors on the deathfrom cerebrovascular accident. Jpn Circ J 1969;33:1285-98.

12 Ohno Y. Biometeorologic studies on cerebrovascular dis-eases. II. Seasonal observation on effects of meteoro-logic factors on the death from cerebrovascularaccident. Jpn Circ Y 1969;33:1299-308.

13 Ohno Y. Biometeorologic studies on cerebrovascular dis-eases. m. Effects by the combination of meteorologicchanges on the death from cerebrovascular accident.Jpn CircJ 1969;33:1309-14.

14 Locksley HB. Natural history of subarachnoid hemor-rhage, intracranial aneurysms and arteriovenous mal-formations. Pt I. In: Sahs AL, Perret GE, LocksleyHB, Nishioka H, eds. Intracranial aneurysms and sub-arachnoid hemorrhage. A Cooperative study. Philadelphia:Lippincott 1969:37-57.

15 Freytag E. Fatal rupture of intracranial aneurysms.Survey of 250 medicolegal cases. Arch Pathol 1966;81:418-24.

16 Pakarinen S. Incidence, aetiology, and prognosis ofprimary subarachnoid hemorrhage. Acta Neurol Scand1967;29:Suppl 1-128.

17 Locksley HB. Natural history of subarachnoid hemor-rhage, intracranial aneurysms and arteriovenous mal-formations. Part II. In: Sahs AL, Perret GE, LocksleyHB, Nishioka H, eds. Intracranial aneurysms andsubarachnoid hemorrhage. A Cooperative Study.Philadelphia: Lippincott, 1969:58-108.

18 Sahs AL. Randomized treatmnent study. Introduction.In: Sahs AL, Nibbelink DW, Turner JC, eds.Aneurysmal Subarachnoid Hemorrhage. Report of theCooperative Study. Baltimore: Urban andSchwarzenberg, 1981:19-20.

19 Stehbens WE. Subarachnoid hemorrhage. Pathology ofthe cerebral blood vessels. St Louis CV Mosby, 1972:252-83.

20 Adams HP, Jergenson DD, Sahs AL. Pitfalls in therecognition of subarachnoid hemorrhage. jAMA 1980;244:794-6.

21 Leblanc R. The minor leak preceding subarachnoidhemorrhage. J Neurosurg 1987;66:35-9.

22 Leblanc R, Winfield JA. The warning leak in subarach-noid hemorrhage and the importance of its early diag-nosis. Can MedAssocJ7 1984;131:1235-6.

23 Calvert JM. Premonitory symptoms and signs of sub-arachnoid haemorrhage. MedjAust 1966;53:651-7.

24 Richardson JC, Hyland HH. Intracranial aneurysm. Aclinical and pathological study of subarachnoid andintracerebral hemorrhage caused by berry aneurysms.Medicine 1984;20:1-83.

25 Berman AJ. The problem of the intracranial aneurysm.Angiology 1958;9:136-53.

26 Gillingham FJ. The management of ruptured intracranialaneurysm. Ann Roy Coll Surg 1958;23:89-117.

27 Shields CB. Current trends in management of cerebralaneurysms. J7 Kentucky MedAssoc 1977;75:529-35.

28 Kassell NF, Kongable GL, Torner JC, Adams HP,Mazuz H. Delay in referral of patients with rupturedaneurysms to neurosurgical attention. Stroke 1985;16:587-90.

29 Kassell NF, Torner JC, Haley C, et al. The internationalcooperative study on the timing of aneurysm surgery.Part I: Overall management results. Jf Neurosurg1990;73:18-36.

30 Ito Z, Matsuoka S, Moriyama T, et al. Factors related tolevel of consciousness in the acute stage of rupturedintracranial aneurysms. Brain Nerve (Tokyo) 1975;27:895-901.

31 Austin DC. A review of intracranial aneurysms. HenryFord Hosp Med Bull 1964;12:251-71.

32 Fisher CM. Clinical syndromes in cerebral thrombosis,hypertensive hemorrhage, and ruptured saccularaneurysm. Clin Neurosurg 1975;22:117-47.

33 Hart RG, Byer JA, Slaughter JR, et al. Occurrence andimplications of seizures in subarachnoid hemorrhagedue to ruptured intracranial aneurysms. Neurosurgery1981;8:417-21.

34 Sarner M, Rose FC. Clinical presentation of rupturedintracranial aneurysm. I Neurol Neurosurg Psychiatry1967;30:67-70.

35 Tsementzis SA, Williams A. Ophthaimological signs andprognosis in patients with a subarachnoid hemorrhage.Neurochirurgia 1984;27: 133-5.

36 Kendall BE, Lee BC, Claveria E. Computerized tomog-raphy and angiography in subarachnoid haemorrhage.BRYRadiol 1976;49:4873-501.

37 Liliequist B, Lindquist M, Valdimarsson E. Computedtomography and subarachnoid hemorrhage. Neuro-radiology 1977;14:21-6.

38 Modesti LM, Binet EF. Value of computed tomographyin the diagnosis and management of subarachnoidhemorrhage. Neurosurgery 1978;3: 151-6.

39 Scotti G, Ethler R, Melancon D, et al. Computedtomography in the evaluation of intracranial aneurysmsand subarachnoid hemorrhage. Radiology 1977;123:85-90.

40 Weir B, Miller J, Russell D. Intracranial aneurysms: Aclinical, angiographic, and computerized tomographicstudy. CanY Neurol Sci 1977;4:99-105.

41 Fisher CM, Kistler JP, Davis JM. Relation of cerebralvasospasm to subarachnoid hemorrhage visualized bycomputerized tomographic scanning. Neurosurgery1980;6: 1-9.

42 Duffy GP. Lumbar puncture in spontaneous subarach-noid haemorrhage. BM3t 1982;285:1 163-4.

43 Sicard J-A. Chromodiagnostid cu liquide cephalorachi-dien dans les hemorragies du nevraxe. Valeur de lateinte jaunatre. CR Soc Biol 1901;53:1050-3.

44 Milian G, Chiray. Meningite a pneumocoques.Zanthochromie du liquide cephal-rachidien. Bull SocAnat Paris 1902;4:550-2.

957



j.com/

J Neurol N


eptember 1993. D

ownloaded from


Kopitnik, Samson

45 Froin G. Les Hemorragies Sous-Arachnoidiennes et leMicanisme de l'Hematolyse en gineral. Paris: Steinheil,1904.

46 Collier J, Adie WJ. Cerebral vascular lesions. In: PriceFW, ed. A textbook of the practice of medicine. London:Henry Frowde and Hodder and Stoughton, 1922:1348-65.

47 Symondo CP. Spontaneous subarachnoid haemorrhage.QJMed 1924-25;18:93-122.

48 Greenfield JG, Carmichael EA. The cerebrospinal fluid inclinical diagnosis. London, Macmillan, 1925:50-2.

49 Barrows LJ, Hunter FT, Banker BQ. The nature andclinical significance of pigments in the cerebrospinalfluid. Brain 1955;78:59-80.

50 Bramwell E. The etiology of recurrent ocular paralysis(including periodic ocular paralysis and ophthalmo-plegic migraine). Edinburgh MedJ 1933;40:209-81.

51 Botterell EH, Lougheed WM, Scott JW, et al.Hypothermia, and interruption of carotid, or carotidand vertebral circulation in the surgical management ofintracranial aneurysms. J Neurosurg 1956;13: 1-42.

52 Lougheed WM, Marshall BM. Management ofaneurysms of the anterior circulation by intracranialprocedures. In: Youmans JR, ed. Neurological surgeryvol 2. Philadelphia: Saunders, 1973:731-67.

53 Hunt WE, Hess RM. Surgical risk as related to time ofintervention in the repair of intracranial aneurysms.J Neurosurg 1968;28: 14-20.

54 Hunt WE, Kosnik EJ. Timing and perioperative care inintracranial aneurysm surgery. Clinical Neurosurg 1974;21:79-89.

55 Hunt WE. Grading of patients with aneurysms. Letterto the Editor, J Neurosurg 1977;47:13.

56 Nibbelink DW, Torner J, Henderson WG. Intracranialaneurysms and subarachnoid hemorrhage. Stroke 1977;8:202-18.

57 Hesselink JR. Investigation of intracranial aneurysm. In:Fox JL, ed. Intracranial aneurysms, vol 1. New York:Springer-Verlag. 1983:497-548.

58 Teal JS, Wade PJ, Bergeron RT, et al. Ventricular opaci-fication during carotid angiography secondary to rup-ture of intracranial aneurysm. Radiology 1973;106:581-3.

59 Vines FS, Davis DO. Rupture of intracranial aneurysmat angiography. Radiology 1971;99:353-4.

60 Goldstein SL. Ventricular opacification secondary torupture of intracranial aneurysm during angiography.J Neurosurg 1967;27:265-7.

61 Hackney DB, Lesnick JE, Zimmerman RA, GrossmanRI, Goldberg HI, Bilaniuk LT. MR identification ofbleeding site in subarachnoid hemorrhage with multi-ple intracranial aneurysms. J Comput Assist Tomog1986;10:878-80.

62 Friedman AH. Subarachnoid hemorrhage of unknownetiology. In: Wilkins RH, Rengachary SS, eds.Neurosurgery update II, New York: McGraw-Hill,1991:73-7.

63 Nishioka H, Torner JC, Graf CJ, et al. Cooperative studyof intracranial aneurysms and subarachnoid hemor-rhage: A longterm prognostic study. IH. Subarachnoidhemorrhage of undetermined etiology. Arch Neurol1984;41:1 147-51.

64 Perret G, Nichioka H. Cerebral angiography: Diagnosticvalue and complications of carotid and vertebralangiography. In: Sahs AL, ed. Intracranial aneurysmsand subarachnoid hemnorrhage: a cooperative study.Philadelphia: Lippincott, 1969:109-24.

65 Forster DMC, Steiner L, Hakanson S. The value ofrepeat panangiography in cases of unexplained sub-arachnoid hemorrhage. J Neurosurg 1978;48:712-6.

66 Biller J, Toffol GJ, Kassell NF, et al. Spontaneous sub-arachnoid hemorrhage in young adults. Neurosurgery1987;21:664-7.

67 Juul R, Fredricksen TA, Ringkjob R Prognosis insubarachnoid hemorrhage of unknown etiology.J Neurosurg 1986;64:359-62.

68 Giombini S, Burzzone MG, Pluchino F. Subarachnoidhemorrhage of unexplained cause. Neurosurgery 1988;22:313-6.

69 Suzuki S, Kayama T, Sakurai Y, et al. Subarachnoidhemorrhage of unknown cause. Neurosurgery 1987;21:310-3.

70 Alexander MSM, Dias PS, Uttley D. Spontaneous sub-arachnoid hemorrhage and negative cerebral panan-giography: Review of 140 cases. J Neurosurg 1986;64:537-42.

71 Andrioli GC, Salar G, Rigobello L, et al. Subarachnoidhaemorrhage of unknown etiology. Acta Neurochir(Wien) 1979;48:217-21.

72 Beguelin C, Seiler R. Subarachnoid hemorrhage withnormal cerebral panangiography. Neurosurgery 1983;13:409-11.

73 Brismar J, Sundbarg G. Subarachnoid hemorrhage ofunknown origin: Prognosis and prognostic factors.J Neurosurg 1985;63:349-54.

74 Nishioka H, Tomner JC, Graf CJ, Kassell NF, Sins ALGoettler LC. Cooperative study of intracranialaneurysms and subarachnoid hemorrhage: A long-termprognostic study. II. Ruptured intracranial aneurysmsmanaged conservatively. Arch Neurol 1984;41:1142-6.

75 Stober T, Emde H, Anstatt T, et al. Blood distributionin computer cranial tomograms after subarachnoidhemorrhage with and without an aneurysm on angio-graphy. Eur Neurol 1985;24:319-23.

76 Smith RR, Miller JD. Pathophysiology and clinical

evaluation of subarachnoid hemorrhage. In: YoumansJR. Neurological Surgery. Philadelphia: WB Saunders,1990: 1644-60.

77 Margolis MT, Newton TH. Methamphetamine("speed") arteritis. Neuroradiology 1971;2: 179-82.

78 Weir B. Medical, neurologic, and ophthalmologicaspects of aneurysms, Pt 2: Neurology of aneurysmsand subarachnoid hemorrhage. Aneurysm affecting thenervous system. Baltimore: Williams and Wilkins,1987;74-83.

79 Bohn E, Hugosson R Experiences of surgical treatmentof 400 consecutive ruptured cerebral aneurysms. ActaNeurochir 1978;49:33-43.

80 Modesti LM, Binet EF. Value of computed tomographyin the diagnosis and management of subarachnoidhemorrhage. Neurosurgery 1978;3:151-6.

81 Gijn JV, Hijdra A, Wijdicks EF, Vermeulen M, CrevelHV. Acute hydrocephalus after aneurysmal subarach-noid hemorrhage. J Neurosurg 1985;63:355-62.

82 Bogdahn U, Lau W, Hassel W, Gunreben G, MertensHG, Brawanski A. Continuous-pressure controlled,external ventricular drainage for treatment of acutehydrocephalus-Evaluation of risk factors. Neurosurg1992;31:898-904.

83 Pakarinen S. Incidence, aetiology, and prognosis of pri-mary subarachnoid hemorrhage: A study based on 589cases diagnosed in a defined urban population during adefined period. Acta Neurol Scand 1967;43:(Suppl29): 1-128.

84 Chason JL, Hindman WM. Berry aneurysms of the circleof Willis: Results of a planned autopsy study. Neurology1958;8:41-4.

85 Nakagawa F, Kobayashi S, Takemae T, Sugita K.Aneurysms protruding from the dorsal wall of theinternal carotid artery. JNeurosurg 1986;65:303-8.

86 Forbus WD. On the origin of the miliary aneurysms ofthe superficial cerebral arteries. Bull Johns HopkinsHosp 1930;47:239-84.

87 Early CB, Fink LH. Some fundamental applications ofthe law of LaPlace in neurosurgery. Surg Neurol1976;6: 185-9.

88 Graf CJ. Prognosis for patients with nonsurgically treatedaneurysm. Analysis of the cooperative study ofintracranial aneurysm and subarachnoid hemorrhage.Y Neurosurg 1971;35:438-43.

89 McCormick WF, Rosenfield DB. Massive brain hemor-rhage. A review of 144 cases and an examination oftheir causes. Stroke 1973;4:946-54.

90 Drake CG. Giant intracranial aneurysms: Experiencewith surgical treatment in 174 patients. Clin Neurosurg1979;26: 12-95.

91 Sundt TM, Piepgras DG. Surgical approach to giantintracranial aneurysm. Operative experience with 80cases. J Neurosurg 1979;51:731-42.

92 Crompton MR. Mechanism of growth and rupture incerebral berry aneurysms. BMJ 1966;1:1138-42.

93 Crompton MR. The natural history of cerebral berryaneurysms. Am Hearty 1976;73:567-9.

94 Stehbens WE. Aneurysms and anatomical variation ofcerebral arteries. Arch Pathol 1953;75:45-64.

95 DuBoulay GH. The significance of loculation ofintracranial aneurysms. Bull Schweiz Akad Med Wiss1969;24:480-5.

96 Nehls DG, Flom RA, Carter LP, Spetzler RF. Multipleintracranial aneurysms: Determining the site of rup-ture. J Neurosurg 1985;63:342-8.

97 Heiskanen 0. The identification of ruptured aneurysm inpatients with multiple intracranial aneurysms.Neurochirurgia (Stuntg) 1965;8:102-7.

98 Pool JL, Potts DG. Aneurysms and arteriovenous anomaliesof the brain. Diagnosis and treatment. New York: Harperand Row, 1965.

99 Kassell NF, Peerless SJ, Durward QJ, et al. Treatment ofischemic deficits from vasospasm with intravascularvolume expansion and induced arterial hypertension.Neurosurgery 1982;11:337-43.

100 Swift DM, Solomon RA. Unruptured aneurysms andpostoperative volume expansion. J Neurosurg 1992;77:908-10.

101 Yasargil MG. Unoperated cases. Microneurosurgery.Stuttgart: Georg Thieme Verlag, 1984:329-30.

102 Bailes JE, Spetzler RF, Hadley MN, Baldwin HZ.Management morbidity and mortality of poor-gradeaneurysm patients. J Neurosurg 1990;72:559-66.

103 Norlen G, Olivercrona H. The treatment of aneurysmsof the circle of Willis. YNeurosurg 1953;10:404-15.

104 Weir B, Aronyk K. Management and postoperative mor-tality related to time of clipping for supratentorialaneurysms. A personal series. Acta Neurochir 1982;63:135-9.

105 Wslkins RH. The role of intracranial arterial spasm in thetiming of operations for aneurysm. Clin Neurosurg1977;24: 185-207.

106 Kassell NF, Torner JC, Jane JA, et al. The internationalcooperative study on the timing of aneurysm surgery.Part 2: Surgical results. JNeurosurg 1990;73:37-47.

107 Medlock MD, Dulebohn SC, Elwood PW. Prophylactichypervolemia without calcium channel blockers in earlyaneurysm surgery. Neurosurg 1992;30;12-6.

108 Torner JC, Kassell NF, Wallace RB, Adams HP.Preoperative prognostic factors for rebleeding and sur-vival in aneurysm patients receiving antifibrinolytictherapy: Report of the cooperative aneurysm study.Neurosurgery 1981;9:506-13.

109 Pickard JD, Murray GD, Illingworth R, et al. Effect of

958 on F


http://jnnp.bmj.com

/J N

eurol Neurosurg P


ber 1993. Dow

nloaded from



oral nimodipine on cerebral infarction and outcomeafter subarachnoid haemorrhage: British AneurysmNimodipine Trial. BMJ 1989;298:636-42.

110 Allen GS, Ahn HS, Preziosi TJ, et al. Cerebral arterialspasm: A controlled trial of nimodipine in patients withsubarachnoid hemorrhage. N Engi J Med 1983;308:619-24.

111 Mee E, Dorrance D, Lowe D, et al. Controlled study ofnimodipine in aneurysm patients treated early aftersubarachnoid hemorrhage. Neurosurg 1988;22:484-91.

112 Pretruk KC, West M, Mohr G, et al. Nimodipinetreatment in poor grade aneurysm patients. Results ofa multicenter double-blind placebo-controlled trial.J Neurosurg 1988;68:505-17.

113 Kassell NF, Torner JC, Adams HP. Antifibrinolytictherapy in the acute period following aneurysmal sub-arachnoid hemorrhage: Prelimninary observations fromthe cooperative aneurysm study. J Neurosurg 1984;61:225-30.

114 Adams HP Jr. Current status of antifibrinolytic therapyof patients with subarachnoid hemorrhage. Stroke1982;13:256-9.

115 Burchiel JK, Hoffman JN, Bakay FAR. Quantitativedetermination of plasma fibrinolytic activity in patientswith ruptured intracranial aneurysms who are receivingEaminocaproic acid: Relationship of possible complica-tions of therapy to the degree of fibrinolytic inhibition.Neurosurg 1984;14:57-63.

116 Adams HP, Nibbelink DW, Torner JC, Sahs AL.Antifibrinolytic therapy in patients with aneurysmalsubarachnoid hemorrhage. In: Sahs AL, NibbelinkDW, Torner JC, eds. Aneurysmal subarachnoid hemor-rhage. Report of the Cooperative Study. Baltimore: UrbanSchwarzenberg, 1981:331-9.

117 Park BE. Spontaneous subarachnoid hemorrhage com-plicated by communicating hydrocephalus: Epsilonamino caproic acid as a possible predisposing factor.Surg Neurol 1979;11:73-80.

118 Rosenom J, Eskesen V, Schmidtk, et al. The risk ofrebleeding from ruptured intracranial aneurysms.J Neurosurg 1987;67:329-32.

119 Kassell NF, Tomer JC. Aneurysmal rebleeding: A pre-liminary report from the cooperative study.Neurosurgery 1983;13:479-81.

120 Paine JT, Batjer HH, Samson DS. Intraoperative ven-tricular puncture-Technical note. Neurosurgery 1988;22:1107-9.

121 Mizukami M, Kawase T, Usami T, et al. Prevention ofvasospasm by early operation with removal of sub-arachnoid blood. Neurosurgery 1982;10:301-7.

122 Sano K, Saito I. Early operation and washout of bloodclots for prevention of cerebral vasospasm. In: WilkinsRH, ed. Cerebral arterial spasm. Baltimore: Williamsand WiLkins, 1980:510-3.

123 Taneda M. Effect of early operation for rupturedaneurysm on prevention of delayed ischemic symp-toms. J Neurosurg 1982;57:622-8.

124 BatJer HH, Samson D. Intraoperative aneurysmal rup-ture: Incidence, outcome, and suggestions for surgicalmanagement. Neurosurgery 1986;18:701-6.

125 Yasargil MG. Mirneurosurgery, vol 2. Stuttgart, GeorgeThieme Verlag, 1984:58-9.

126 Giannotta SL, Oppenheimer JH, Levy ML, Zelman V.Management of intraoperative rupture of aneurysmwithout hypotension. Neurosurgery 1991;28:531-6.

127 BatJer HH, Frankfurt AI, Purdy PD, et al. Use ofEtomidate, temporary arterial occlusion, and intraoper-ative angiography in surgical treatment of large andgiant cerebral aneurysms. Y Neurosurg 1988;68:234-40.

128 Mizukami M, Kawase T, Usame I, Tazawa T.Prevention of vasospasm by early operation withremoval of subarachnoid blood. Neurosurgery 1981;10:301-7.

129 Teneda M. Effect of early operation for rupturedaneurysms on prevention of delayed ischemic symp-toms. J Neurosurg 1982;57:622-8.

130 Mizoi K, Yoshimoto T, Fujiwara S, Sugawara T,Takahashi A, Koshu K. Prevention of vasospasm by

clot removal and intrathecal bolus injection at tissue-type Plasminogen Activator: Preliminary report.Neurosurg 1991;28:807-13.

131 Findlay JM, Weir BKA, Steinke D, Tanabe T, GordonP, Grace M. Effect of intrathecal thrombolytictherapy on subarachnoid clot and chronic vasospasmin a primate model of SAH. J Neurosurg 1988;69:723-35.

132 Seifert V, Eisert WG, Stolke D, Goetz C. Efficacy ofsingle intracisternal bolus injection of recombinanttissue plasminogen activator to prevent delayedcerebral vasospasm after experimental subarachnoidhemorrhage. Neurosurg 1989;25:590-8.

133 Findlay JM, Weir RILA, Kassell NF, Disney LB, GraceMGA. Intracisternal recombinant tissue plasminogenactivator after aneurysmal subarachnoid hemorrhage.J Neurosurg 1991;75:181-8.

134 Takaku A, Tanaka S, Mori T, Suzuki J. Postoperativecomplications in 1,000 cases of intracranial aneurysms.Surg Neurol 1979;12: 137-44.

135 Yamamoto I. Early operation for ruptured intracranialanerysms: comparative study with computerizedtomography. Presented at the 32nd Annual Meeting,Congress ofNeurological Surgeons, Toronto, 1982.

136 Heros R, Zarvas N, Varsos V. Cerebral vasospasm aftersubarachnoid hemorrhage: An update. Ann Neurol1983;14:599-608.

137 Ropper AH, Zervas NT. Outcome 1 year after SAHfrom cerebral aneurysm. Management, morbidity,mortality, and functional status in 112 consecutivegood risk patients. JNeurosurg 1984;60:909-15.

138 Fisher CM, Roberson GH, Ojemann RG. Cerebralvasospasm with ruptured saccular aneurysm-Theclinical manifestations. Neurosurgery 1977:1:245-8.

139 Weir BK. Pathophysiology of vasospasm. Int AnesthesiolClin 1982;l0:39-43.

140 Pasqualin A, Rosta L, Da Pian R, Cavazzani P, ScienzaR. Role of computed tomography in the managementof vasospasm after subarachnoid hemorrhage.Neurosurgery 1984;15:344-53.

141 Kistler JP, Crowell RM, Davis KR, Heros R, OjemannRG, Zervas T, Fisher CM. The relation of cerebralvasospasm to the extent and location of subarachnoidblood visualized by CT scan. A prospective study.Neurology 1983;33:424-36.

142 Symon L. Disordered cerebro-vascular physiology inaneurysmal subarachnoid haemorrhage. Acta Neurochir(Wien) 1979;41:7-22.

143 Pritz MB, Giannotta SL, Kindt GW, McGillicuddy JE,Prager RL Treatment of patients with neurologicaldeficits associated with cerebral vasospasm by intravas-cular volume expansion. Neurosurgery 1978;3:364-8.

144 Shimoda M, Oda S, Tsugane R, Sato 0. Intracranialcomplications of hypervolemic therapy in patients witha delayed ischemic deficit attributed to vasospasm.J Neurosurg 1993;78:423-9.

145 Kaku Y, Yonekawa Y, Tsukahara T, Kazekawa K.Superselective intra-arterial infusion of papaverine forthe treatment of cerebral vasospasm after subarachnoidhemorrhage. J Neurosurg 1992;77:842-7.

146 Kassell NF, Helm G, Sommons N, Phillips CD, CailWS. Treatment of cerebral vasospasm with intraarterialpapaverine. J Neurosurg 1992;77:848-52.

147 Higashida RT, Halbach W, Cahan LD, et al.Transluminal angioplasty for treatment of intracranialarterial vasospasm. YNeurosurg 1989;71:648-53.

148 Newell DW, Eskridge JM, Maybery MR, et al.Angioplasty for the treatment of symptomatic vaso-spasm following subarachnoid hemorrhage. J Neurosurg1989;71:654-60.

149 Zubkov YN, Nikiforov BM, Shustin VA. Ballooncatheter technique for dilatation of constricted cerebralarteries after aneurysmal SAH. Acta Neurochir 1984;70:65-79.

150 Linskey ME, Horton JA, Rao GR, Yonas H. Fatal rup-ture of the intracranial carotid artery during translumi-nal angioplasty for vasospasm induced by subarachnoidhemorrhage. J Neurosurg 1991;74:985-90.

959



j.com/

J Neurol N


eptember 1993. D

ownloaded from


Management of subarachnoid haemorrhage · SAH and occur during haemorrhage in approximately...

Documents

Transcript of Management of subarachnoid haemorrhage · SAH and occur during haemorrhage in approximately...