POSTGRAD. MED. J. (i962), 38, iS

SURGICAL ASPECTS OFCEREBROVASCULAR DISEASE

W. F. C. KENNEDY, B.M., B.CH., JOE PENNYBACKER, M.D., F.R.C.S.The Department of Neurological Surgery, The Radcliffe Infirmary, Oxford

VASCULAR lesions affecting the brain are importantbecause of their frequency, their incidence in allage groups, their often profound and long-lastingclinical effects, and their social and economicrepercussions on the community. Such lesionspresented to a surgical department are usuallyhighly selected, but for the purpose of this studyone of us (W.F.C.K.) saw all, or practically all,cases of cerebrovascular disease admitted to theRadcliffe Infirmary for a period of I2 months in theyears 1958-59. We found, among other things,;that diagnosis can be very difficult, that it is some-times impossible to say whether a patient has had'a cerebral haemorrhage or a thrombosis, and thatspecial investigations will not always clarify theissue. The therapeutic problems confronting thesurgeon include the decisions about whether andwhen to operate, and also the technical problemsof operation. The increasing use of angiographyin the last two decades has made an importantcontribution to the awareness of physicians andsurgeons of the suitability of certain vascularlesions for surgical treatment, and what follows isan attempt to indicate the possibilities.

.Congenital AbnormalitiesAneurysms

Cerebral aneurysms most commonly declaretheir presence by rupture, but may sometimespresent without rupture on account of localpressure effects, or pain, or occasionally they aremassive enough to behave as space-occupyinglesions.The particular susceptibility to aneurysm forma-

tion of the arteries composing the Circle of Willis,its branches and tributaries is due to localizedcongenital defects of the muscle coat (tunica media)which commonly occur in the angles of branchingand which may be demonstrated histologically.Such defects alone may account for the develop-ment of aneurysms in young people, but otherimportant factors are the degenerative changes inthe muscle coat of atheromatous vessels and highblood pressure. When severe these latter factorsmay be entirply responsible for aneurysm forma-

tion without an underlying congenital defect, andthey particularly give rise to the fusiform type ofaneurysmal dilatation. Mycotic aneurysms arisingat the site of impaction of infective emboli aremuch rarer, but are an occasional cause of intra-cranial hamorrhage or suppuration.The wall of an aneurysm consists only of a thin

fibrous layer, reinforced to a variable extent bymural thrombus, and it contains no muscle coat.Crawford (1959) found the site of rupture at thefundus of the sac in 84% of IO5 post-mortem casesin which it could be identified by careful dissection.Rupture at the neck was found in only three casesand rupture of the lateral wall accounted for theremainder. These findings, have an obviousbearing on the rationale of treatment by clippingor ligation of the neck.The pathology of aneurysmal hzemorrhage is not

limited to the aneurysm itself,, for the state of thebrain bears a more direct relation to the clinicalevents and to the prospects of surgical treatment.The actual hamorrhage probably ceases within afew seconds or minutes of rupture, but during thisinterval blood at arterial pressure escapes into theadjacent tissue plane, which is usually the sub-arachnoid space, often the substance of the brain,and occasionally the subdural space. Thus, anintra-cerebral hlmatoma is a common complica-tion, or the haemorrhage may burst through tothe ventricular system, or a subdural haematomamay occasionally be formed, although it is rarely alarge one. Massive subarachnoid bleeding maylead to the filling of the basal cisterns with solidclot, and this is usually present at autopsies onfatal cases, when often the ventricles are full ofclot as well. Less severe hamorrhage may alsointerfere with the patency of the intra-cerebraland extra-cerebral pathways of the cerebro-spinal fluid and may thus cause internal or externalhydrocephalus, which contributes to the increasein intracranial pressure in the acute phase. Laterthe pathways affected in this manner may beobliterated by fibrosis and surgical relief with aby-pass operation may be necessary after recoveryfrom the initial effects of the haemorrhage hastaken place, as in certain cases of infantile hydro-

copyright. on A

pril 20, 2022 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.38.435.18 on 1 January 1962. D

ownloaded from

Jamnary I962 KENNEDY and PENNYBACKER: Surgical Aspects of Cerebrovascular Disease

cephalus following birth trauma (Kibler, Couchand Crompton, I961).

Subdural haematoma is a rare complication ofaneurysmal rupture and can only be diagnosedwith certainty by angiography; its presence con-stitutes an indication for operation in its ownright, whatever the nature and site of the under-lying vascular anomaly. It can be associated withsevere and rapidly fatal hlemorrhage, but some-times it is not discovered until some days or weeksafter the onset (Clarke and Walton, I953).

Narrowing, or spasm, of the cerebral vessels isa common angiographic or operative finding in theacute stage of subarachnoid haemorrhage (Potter,1959) and one peculiar to this condition. It is notnecessarily confined to the vessel bearing theaneurysm which has ruptured, but is usuallygreatest in this vessel and therefore it can be avaluable guide to the site of rupture in a case wheremultiple aneurysms are present. It usually lastsfrom one to three weeks and may be severe enoughto cause local or widespread infarction. Certainlyit renders the brain exceptionally liable to cedemaor infarction and vulnerable to injury by even thegentlest surgical manipulation. Ischaemic damagemay also result from mechanical stretching, dis-ruption and thrombosis of vessels, some of themost sensitive being the small perforating branchesof the Circle of Willis, which ascend to the vitaldeep cerebral structures.

Occasionally an intracavernous carotid aneurysmruptures to form a carotid-cavernous fistula. Otheraneurysms in this situation may enlarge to fill thesinus without rupturing, acquiring thereby anextra coat of dura, and further enlargement maythen take place, sometimes to fill the middle cranialfossa. These aneurysms often cause pain, especiallyround the eye, and local pressure effects whichinclude vascular congestion of the orbital contents,proptosis, and lesions of the second, third, fourthand fifth cranial nerves, and of the ocular sym-pathetic fibres. They may, therefore, simulate anintrasellar or parasellar tumour. X-rays may showlocal bone erosion in the optic foramen, superiororbital fissure, sella turcica, and floor of the middlecranial fossa, and the aneurysm may calcify.Long-term survival with no more than ocularmanifestations is common, but surgical treatmentmay be required on account of pain, progressivevisual failure or proptosis (Meadows, I96I).The definitive treatment of ruptured aneurysm

has to be considered in relation to what is knownof the natural course of the disease and eachindividual case presents its own particular problemsbecause so many variable factors influence thepossibility and effectiveness of surgical treatment.Salient amongst these are the site, size and shapeof the aneurysm, the presence of multiple

aneurysms, the severity and site of hiemorrhage,the time since it occurred, and the clinical state ofthe patient. Thus, there are rapidly fatal cases inwhich, at autopsy, the cerebral damage is so ex-tensive that it is obvious that no treatment couldhave been worth while. Other factors which haveto be taken into account are the number of previoushlemorrhages, the adequacy of the collateral cir-culation as demonstrated by careful angiography,and the age and general health of the patient,with special reference to tie cardiovascular system.The results of treatment are also particularly

difficult to assess against the background of such avariable illness and the differing objectives of sur-gical treatment must also be kept in mind. Theseare to exclVde the aneurysm from the circulationin order to prevent recurrent haemorrhage and toevacuate an intracranial haematoma when such ispresent. Less commonly, operation is performedfor pain or ocular manifestations of unrupturedaneurysms. To complicate the situation further isthe variety of different operations undertaken: thetheoretical procedure of choice is obliteration ofthe aneurysm by clipping or ligating its neckwithout jeopardizing the arteries in relation to it.This is not always practicable without sacrifice ofthe circulation distal to the aneurysm, and it may beimpossible to predict the effect of intervention.Other possibilities include packing and wrappingthe aneurysm with muscle, or acrylic resin (Dutton,I959), ligation or clipping of the feeding vesselproximal to the aneurysm, or both proximal anddistal to it (' trapping '), or ligation of the commonor internal carotid arteries in the neck.

It is the general experience that the risks of alloperations for ruptured aneurysms are greatest inthe first few days after haemorrhage; at this stagethe technical problems of the intracranial approachare the most difficult, and the incidence of hemi-plegia after carotid ligation in the neck is thehighest. Also, the adequate demonstration of thecerebral vascular tree by angiography is oftenlimited by the presence of spasm. Yet if themortality and morbidity from early recurrenthaemorrhage are to be reduced, investigation andtreatment must be instituted in the early days afterrupture, for the peak incidence of this complicationis in the second and third weeks (Walton, 1956).At present it is by no means certain that opera-

tion in the acute stage of bleeding reduces mortalityand morbidity significantly. Many of the goodresults claimed for surgery are now thought to beattributable mainly to selection of more pro-mising cases and to failure to assess the resultsagainst a truly comparable control series. ThusMcKissock, Paine and Walsh (Ig6ob), in a retro-spective analysis of 599 cases of ruptured aneurysmin which the lesion could be angiographically

19copyright.

on April 20, 2022 by guest. P

rotected byhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.38.435.18 on 1 January 1962. Dow

nloaded from

POSTGRADUATE MEDICAL JOURNAL

demonstrated, found that their overall mortalityfor the surgically treated cases was 33% (47°%in the conservatively treated group). The mor-tality of operations performed within three daysof the hamorrhage was nearly 50%. The mortalityin conservatively treated patients who had alreadysurvived a corresponding interval of time wasnearly 75%, but this figure was weighted by theunusually high mortality (92%) in those admittedwithin 24 hours of the hemorrhage. For thosewho survived from two to three weeks beforeoperation the mortality was still as high as 30%,whilst amongst the non-surgical group of survivorsafter this interval the mortality was zo%. For allthose who survived four or more days after thehiemorrhage the natural mortality was 30% and theoperative mortality 26%. Furthermore, there wasa much higher proportion of totally and partiallydisabled survivors in the surgical group than in theconservative group, whereas the proportion ofthose who could return to full work was roughlythe same in each. The operative mortality whenthe patient was still in coma at the time of opera-tion was 8o%, whereas it was zo% in alert patients,and the results of conservative treatment in corre-sponding groups of patients were only a littleworse. The authors emphasize that, owing to thefactors of selection and difficulty of control, nofinal conclusions can yet be drawn, and they aremaking a prospective clinical trial to establish thefacts more clearly.On the other hand, if investigation and operation

are delayed until three or four weeks after rupture,the risks are much lower, but so also are the risks ofrecurrent hxmorrhage. After an interval of a fewmonths from a single hlmorrhage it is questionablewhether the risks of operation are less than therisks of doing nothing. Furthermore, one is nowconcerned only with those who survived the acuteillness and who naturally carry a more favourableprognosis.

In this department the policy is to delay angio-graphy until the patient has made a satisfactoryclinical recovery from the immediate effects of thehiemorrhage with respect to his state of alertness,or until failure to improve gives rise to the sus-picion of an intracranial hamatoma. When ahaematoma is present operation for its removal isundertaken and, when possible, the causativevascular anomaly is dealt with. If the clinicaldisability is severe in the absence of an intracranialhematoma, operation is delayed until it is evidentthat a useful degree of recovery is going to takeplace.

Operations on intracranial aneurysms carrysomewhat exceptional risks. The aneurysm mayrupture during exposure and the control of thehbmorrhage may necessitate the occlusion of a

more important vessel than had been contem-plated. This may lead to a serious and permanentdisability where none existed before. Sometimes,in apparently straightforward procedures, sub-sequent thrombosis of adjacent vessels may alsocause profound disability. Sometimes a clip whichseemed securely applied works off the aneurysmand hence it is important to check the result bypost-operative angiography. Not infrequently theprecise shape and size of the aneurysm as seen onthe angic.gram bears little relation to reality asdisclosed at operation, so that it may not bepossible to carry out the planned procedure.Angiography, after all, only visualizes the interiorof the aneurysm.To some extent these risks and the technical

difficulties of exposure can be reduced by ancillaryaids, such as hypothermia (which reduces brainvolume and allows temporary obliteration of thecerebral circulation without infarction) (McKis-sock, Paine and Walsh, iq6a), hypotensiveanesthesia (which reduces the risk of ruptureduring exposure and facilitates its control shouldit occur), and intravenous urea (which reducesthe amount of retraction necessary by decreasingbrain volume) (Stubbs and Pennybacker, I960).Even so, the state of the brain after a recenthemorrhage is unfavourable for precise and,pro-longed dissection, and at least some of its in-creased vulnerability remains. These operationsare therefore sometimes followed by an increase inthe existing neurological deficit, or the appearanceof such where none existed before. This complica-tion may be accompanied by obstruction of acerebral vessel demonstrable by angiography, butoften it is not. Thus, the treatment may bringabout the very disability it was undertaken toprevent.

Operations are undertaken for unrupturedaneurysms when there is pain, progressive ocularpalsy or proptosis. The lesion is usually on thedistal segment of the internal carotid artery andligation of the common carotid artery in the neckis a satisfactory form of treatment. Sometimes asimilar operation is undertaken for similar reasons,or for a troublesome bruit, in carotid-cavernousfistula, although there is a chance of spontaneouscure (Potter, I954) or long survival without dis-abling symptoms in this condition.

In recent years there have been reports ofaneurysms (and angiomatous malformations) inthe posterior fossa successfully treated accordingto the principles employed for those situated in thesupra-tentorial compartment. Undoubtedly theselesions are no longer to be regarded as uniformlyinoperable (Logue and Monckton, 1954; Dimsdaleand Logue, 1959).

Yanuary i 96220copyright.

on April 20, 2022 by guest. P

rotected byhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.38.435.18 on 1 January 1962. Dow

nloaded from

January i962 KENNEDY and PENNYBACKER: Surgical Aspects of Cerebrovascular Disease

Angiomatous MalformationsThe essential component of these lesions is an

arterio-venous anastomosis, which is not neo-plastic, but which may progressively enlarge, withincreasing dilatation of its feeding and drainingvessels, on account of the local h2modynamicconditions and by the excavation of small bloodcysts following hamorrhage (Potter, I955). Unlikethe comparable arterio-venous malformations ofthe limbs, it is unusual to find an associatedsecondary circulatory disturbance (cardiomegaly,tachycardia, plethora).

These malformations may be small or large,may occur anywhere in the brain, and are some-times associated with aneurysms. They commonlygive rise to epilepsy with focal features, by ruptureto subarachnoid or intra-cerebral hxmorrhage(about 5% of such hbamorrhages result from angio-matous malformations) or occasionally to pro-gressive neurological disability by causing ischamiaof the adjacent brain tissue, or by gradual enlarge-ment. They may cause headache simulatingmigraine. Occasionally they obstruct the cerebro-spinal fluid pathways and cause obstructive hydro-cephalus; they may calcify and can then besuspected from the radiological appearance. Exceptin the rare cases which have a bruit or an associatedexternal vascular anomaly of the scalp or face(Sturge-Weber syndrome), we rely on angiographyfor final diagnosis. The crucial feature is venousfilling in the arterial phase, which also providesfurther evidence of their essential nature. This,together with the characteristic tangle of tortuousand dilated arteries and veins, distinguishes themfrom the abnormal vascular patterns of certaintumours.The satisfactory delineation of an angioma may

demand bilateral carotid and vertebral angiographyif its total extent, arterial supply and venousdrainage are to be demonstrated. Moreover, as thespeed of circulation through them may varyenormously, serial angiography is almost essential.Some of these lesions are so extensive and in sucheloquent parts of the brain that they cannot beremoved. But some are accessible and if one ormore hamorrhages have occurred, or if epilepsyproves refractory to medical treatment and con-stitutes a disability, the lesion should be extirpated(Paterson and McKissock, I956). It should besaid that, in our experience, hemorrhage fromangiomatous malformations is much less lethalthan from aneurysms. It should also be said thatwe have no satisfactory long-term assessment ofthe results of these operations undertaken for therelief of epilepsy. Excision of an angioma leaves ascar which may be as potently epileptogenic as wasthe angioma.

Sturge-Weber SyndromeThis consists of a diffuse angiomatous mal-

formation of the pia-arachnoid and cortex of onecerebral hemisphere associated with a similarcutaneous lesion (port-wine stain) in some or allof the ipsilateral trigeminal distribution, andoccasionally also in other parts of the body. Itis not often a surgical problem and haemorrhageis unusual, but fits, mental retardation, abnor-malities of behaviour, hemiplegia and sometimesathetosis may present a problem of social manage-ment which may be partly solved by hemi-spherectomy (Falconer, I960).Subarachnoid Hcemorrhage with NormalAngiogramsA considerable proportion of a series of cases of

subarachnoid hxemorrhage will have normal carotidangiograms. Walton (I956) quotes various authorswho found io% to 20% of such cases in theirseries of patients with subarachnoid hamorrhage.This proportion will be reduced if vertebral angio-graphy is carried out. This investigation will some-times display aneurysms or angiomas- in theposterior fossa, or in the posterior part of thesupratentorial compartment when the posteriorcerebral arteries have failed to fill on carotidangiography. Thus, Spatz and Bull (I957) dis-played such lesions in 26% of 6o cases of sub-arachnoid hamorrhage with negative carotidangiograms (the ages ranging from 6 to 66 years),even when the posterior cerebral arteries were filledby the carotid injections in a third of these cases.Half of the lesions were aneurysms, half wereangiomatous malformations.The greater difficulty of vertebral angiography

militates against its routine use in the investiga-tion of subarachnoid hamorrhage, and it may evenbe necessary to inject both sides to display orexclude with certainty an aneurysm at the originof the ipsilateral posterior inferior cerebellarartery (Dimsdale and Logue, I959). When bilateralcarotid angiograms have proved negative vertebralangiography is usually employed in the younger,healthy patients. If this is also negative, a favour-able prognosis can be given, as the lesion has pre-sumably destroyed itself, or undergone a naturalcure by thrombosis. But occasionally there will bea recurrence and in these cases the lesion is prob-ably too small to be visualized by angiography.We also employ vertebral angiography when thereare clinical symptoms or signs to suggest a lesionon the vertebral system.

Other Causes of Subarachnoid and Intra-cerebralHamorrhage

Although most cases will have demonstrableaneurysms or angiomatous malformations, and

21

copyright. on A

pril 20, 2022 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.38.435.18 on 1 January 1962. D

ownloaded from

POSTGRADUATE MEDICAL JOURNAL

some will have no demonstrable vascular anomaly,a small proportion will have other forms of intra-cranial pathology (when we exclude those due toconstitutional diseases, such as hypertension andblood diseases). Thus, a spontaneous subarachnoidhoemorrhage may cause the first symptoms andconstitute the presenting clinical picture of chronicsubdural haematoma or intracranial tumour; some-times a traumatic subarachnoid hxmorrhage maymasquerade as a spontaneous one if a detailedhistory is not available. We have seen cases ofmeningioma, glioblastoma multiforme, oligo-dendroglioma, acoustic nerve tumour, subduralhlmatoma, sagittal sinus thrombosis and traumaticextradural hamatoma present in this way, withsudden headache, perhaps with depression of con-sciousness, with fresh blood in the cerebro-spinal fluid and without previous symptoms. Inthese cases angiography will usually indicate thenature of the lesion.

Spontaneous Intra-cerebral HEkmorrhage in YoungPeople without Visible AnomaliesThere is a group of cases of intra-cerebral

hmmorrhage in otherwise perfectly healthy youngpeople, usually in the second, third or fourthdecade, in whom angiography, although showingthe hamatoma, fails to display an underlyingvascular anomaly (aneurysm or angioma); neitheris one seen at operation. Such hxamorrhages aregenerally held to have originated in small angio-matous malformations which have been destroyedby the hamorrhage. Sometimes the remnants canbe found by careful dissection at autopsy (Russell,1954). The prognosis, is favourable in those whosurvive if no irreversible neurological damage hasbeen sustained, and if the hematoma is evacuated,when a remarkable degree of neurological recoverymay occur. The hamorrhage is often more super-fidially placed in the cerebral hemisphere, in thesubcortical layer, than the deeply placed liemor-rhage of arteriosclerosis and hypertension. It oftenappears much larger at operation than would havebeen anticipated from the angiograms.

Acquired DiseaseApoplexy

Cerebral hamorrhage occurring in the olderage-groups is due to atheroma and hypertension(occasionally to aneurysm) and the prognosis isinfluenced adversely thereby. The hiemorrhage isoften small and deeply placed in the cerebralhemisphere, with correspondingly more severeneurological damage by smaller lesions. These donot often benefit from attempts at surgical evacua-tion, which can only add to the neurologicaldamage in the process. Sometimes, however, thehimorrhage is large enough and superficial enough

to allow surgical access, but the general-experienceis that little is to be gained by attempting suchoperations within the first fortnight or so. If, afterthat time, significant clinical recovery has not yetbegun, aspiration of the liquefied hlmatoma willsometimes be followed by partial recovery. Forthis purpose a simple burr-hole made under localanesthesia is adequate, and such a limited pro-cedure has obvious advantages when dealing withthis group of cases.The prognosis is influenced by the pathology:

the underlying vascular disease is usually diffuseand there are corresponding risks of furtherepisodes of cerebral hemorrhage, or of ischxemicdisease in the brain or other organs.

Occasionally we have encountered cases of cere-bellar haemorrhage due to degenerative arterialdisease; these are not common, but sometimesthe haematoma requires evacuation, which usuallyentails a full cerebellar exploration rather than asimple burr-hole and aspiration.

Cerebral ThrombosisCerebral thrombosis and thrombo-phlebitis are

not amenable to surgical treatment and concernthe surgeon only as a diagnostic problem or as acomplication of any operation in the elderly, orsometimes after operations for aneurysms inyounger people.Any of the intracranial vessels may be affected

and commonly it is one too small to visualize byangiography and in these cases the thrombosesare often multiple. Thrombosis of the largerintracranial vessels, and especially of the middlecerebral artery, may occur spontaneously in theyoung and usually it is not possible to make aclinical distinction between this lesion and carotidthrombosis. Thrombosis of the posterior cerebralarteries may be spontaneous, or secondary to atentorial pressure cone, and in the latter case itmay give rise to hemianopia as a false localizingsign in cases of intracranial tumour.

Ischiemic damage to the brain does not neces-sarily lead to infarction, for there may be fullrecovery after a temporary and severe disruptionof cerebral function lasting sometimes as long as aweek or a fortnight before any improvement isevident. The recovery is presumably accountedfor by the elaborate anastomoses between the intra-cranial vessels and by recanalization.There is still no certain way of distinguishing

between cerebral hlemorrhage, thrombosis andembolism. The latter is usually diagnosed onlywhen there is a clinically recognizable source ofemboli; biemorrhage and thrombosis are eachcompatible with a sudden or progressive onset ofsymptoms, with retention or depression of con-sciousness, with presence or absence of headache,

3tanuary I96222

copyright. on A

pril 20, 2022 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.38.435.18 on 1 January 1962. D

ownloaded from

January I962 KENNEDY and PENNYBACKER: Surgical Aspects of Cerebrovascular Disease

with slight or severe neurological deficit, withpresence or absence of subarachnoid blood andwith presence or absence of a space-occupyinglesion on the angiograms.

Occasionally such lesions have an atypicalclinical development which may be graduallyprogressive over several days and associated withfever, meningeal irritation and a pleocytosis in thecerebrospinal fluid (perhaps as many as a thousandcells, though usually a few hundred). In thesecases an initial clinical diagnosis of meningitis orabscess may be made.The slow progression over months of small

ischiemic lesions in the brain may cause confusionwith tumour, and this may be impossible toresolve except by the demonstration of cerebralatrophy by air-encephalography or ventriculo-graphy. Commonly such cases present with pro-gressive mental deterioration and evidence of dif-fuse pyramidal affection, but sometimes the signsare remarkably focal, even with focal epilepsy oflate onset.

Carotid Artery DiseaseThe recent advances in the surgery of arterial

reconstruction have led to renewed interest inthrombosis and stenosis of the carotid and ver-tebral arteries in their extracranial course (Rob,1959). The arterial lesion may range from a totalobliteration by thrombosis, or congenital atresia,to a slight degree of stenosis by atheromatousplaques. These plaques have a particular affinityfor the origins of the main vessels and for theirpoints of bifurcation, in particular the origin of theinternal carotid artery.The associated cerebral lesion may be a massive

infarction sufficient to cause death, or total andpermanent loss of function of the appropriatecerebral hemisphere, or it may be much less thanthis, and partly or wholly recoverable spon-taneously, or there may be no detectable abnor-mality. There may be intermittent and transientattacks of cerebral disturbance, such as hemi-paresis, perhaps associated with monocular blind-ness. These intermittent symptoms are especiallylikely to occur at times of general circulatory stress(e.g. hypotensive phases or ana-mia) or when thecollateral and compensatory blood supply isthreatened by mechanical means, such as certainmovements of the neck and shoulder girdle, or bydiffuse arterial disease. Thus, a complete carotidocclusion may cause no symptoms at all until suchtime as the other carotid and the vertebral arterialsystems become narrowed sufficiently to causesymptoms of cerebral ischamia. These may,therefore, be precipitated by the failure of thecollateral circulation rather than by the originalarterial obstruction. The important consideration

is the state of all the extra-cranial cerebral arteriesrather than that of any one of them alone (Hutchin-son and Yates, 1956, 1957; Schwartz andMitchell, 196I).

This conclusion agrees with the common sur-gical experience of the rarity of deficiency symp-toms after ligation of the common carotid arteryin the neck for aneurysm, provided this is doneafter the acute effects of the hxmorrhage have sub-sided. Although ischaemic manifestations, such ashemiplegia or aphasia, may appear at the time ofoperation or within a few days thereafter, latecomplications of this type are very rare. Also,cases of long-standing bilateral carotid occlusionare known in which there have been no features ofcerebral ischaemia at all; in these cases the wholecerebral blood supply must have passed up thevertebral and basilar arteries. There is, however,no information about what happens when peoplewho have had successful carotid ligations foraneurysm before middle age enter the commonage-group of cerebral arteriosclerosis. It would beinstructive to know whether they are more or lessprone to ischaemic lesions than a controlgroup.The abundance of the normal cerebral collateral

circulation renders the assessment of results ofreconstruction operations on the carotid arteries inthe neck particularly difficult. A priori, it wouldseem illogical to undertake these operations in thepresence of established cerebral infarction, as wit-nessed by severe or complete loss of function in ahemisphere, and, indeed, the restoration of arterialflow through an area of infarction may well pre-cipitate a fatal issue owing to haemorrhage. Whenthe neurological deficit has been temporary and isalready recovering spontaneously, this fact isevidence of the present adequacy of the collateralcirculation. An operation on the carotid couldthen only help the patient by preventing a recur-rence of symptoms in the future and would haveto be undertaken without jeopardizing the existingcirculation. The tortuous course and relations ofthe carotid artery, and the sensitivity of its area ofsupply to anoxia, add technical difficulties andproblems which are absent from the surgery ofother peripheral vessels. At present it seems thatpatients most likely to benefit from this kind ofsurgery are those with localized stenosis in oneartery, causing frequent, intermittent symptoms.

ConclusionIt is hardly surprising that problems abound in

the diseases of so complex an arrangement as thecerebral vascular system, on which the working ofthe brain depends. It is also clear that under-standing of these varied disorders is still incom-

23

copyright. on A

pril 20, 2022 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.38.435.18 on 1 January 1962. D

ownloaded from

24 POSTGRADUATE MEDICAL JOURNAL January 1962

plete and much more will have to be learned aboutthem before ideas of management cease to changeas restlessly as in recent years.

We are particularly grateful for the friendly co-opera-tion of the physicians of the Radcliffe Infirmary and fortheir generosity in allowing us to see so many of theirpatients.

REFERENCESCLARKE, E., and WALTON, J. N. (1953): Subdural Hlematoma Complicating Intracranial Aneurysm and Angioma,

Brain, 76, 378.CRAWFORD, T. (1959): Some Observations on the Pathogenesis and Natural History of Intracranial Aneurysms, J.

Neurol. Neurosurg. Psychiat., 22, 259.DIMSDALE, H., and LOGUE, V. (1959): Ruptured Posterior Fossa Aneurysms and their Surgical Treatment, Ibid., 22, 202.DUTTON, J. (1959): Acrylic Investment of Intracranial Aneurysms, Brit. med. J., ii, 597.FALCONER, M. A. (I960): The Treatment of Encephalotrigeminal Angiomatosis (Sturge-Weber Disease) by Hemi-

spherectomy, Proceedings of the Society of British Neurological Surgeons in J. Neurol. Neurosurg. Psychiat., 23, 8i.HUTCHINSON, E. C., and YATES, P. 0. (1956): The Cervical Portion of the Vertebral Artery: A Clinico-Pathological

Study, Brain, 79, 319.(1957): Carotico-Vertebral Stenosis, Lancet, i, 2.

KIBLER, R. F., COUCH, R. S. C., and CROMPTON, M. R. (I96I): Hydrocephalus in the Adult following SpontaneousSubarachnoid Hemorrhage, Brain, 84, 45.

LOGUE, V., and MONCKTON, G. (1954): Posterior Fossa Angiomas. A Clinical Presentation of Nine Cases, Ibid., 77,252.MCKISSOCK, W., PAINE, K. W. E., and WALSH, L. S. (Ig6oa): The Value of Hypothermia in the Surgical Treatment of

Ruptured Intracranial Aneurysms, J. Neurosurg., 17, 700.__, - (Ig6ob): An Analysis of the Results of Treatment of Ruptured Intracranial Aneurysms. Report of 772

Consecutive Cases, Ibid., 17, 762.MEADOWS, S. P. (I96I): Intracranial Aneurysms, 'Medical Annual', p. 325.PATERSON, J. H., and McKISSOcK, W. (1956): A Clinical Survey of Intracranial Angiomas with Special Reference to

their Mode of Progression and Surgical Treatment: A Report of i io Cases, Brain, 79, 233.POTTER, J. M. (I954): Carotid-Cavernous Fistula. Five Cases with ' Spontaneous' Recovery, Brit. med. Y., ii, 786.

(1955): Angiomatous Malformations of the Brain: Their Nature and Prognosis, Ann. roy. Coll. Surg. Eng., I6, 227.(1959): Redistribution of Blood to the Brain Due to Localized Cerebral Arterial Spasm, Brain, 82, 367.

ROB, C. (1959): The Surgical Treatment of Stenosis and Thrombosis of the Internal Carotid, Vertebral and CommonCarotid Arteries, in Symposium on Cerebrovascular Disease, Proc. roy. Soc. Med., 52, 549.

RUSSELL, D. (1954): Discussion: The Pathology of Spontaneous Intracranial Hxemorrhage, Ibid., 47, 689.SCHWARTZ, C. J., and MITCHELL, J. R. A. (1961): Atheroma of the Carotid and Vertebral Arterial Systems, Brit. med. 7.,

ii, 1057.SPATZ, E. L., and BULL, J. W. D. (1957): Vertebral Arteriography in the Study of Subarachnoid Haemorrhage, J7.

Neurosurg., 14, 543.STUBBS, J., and PENNYBACKER, J. (I960): Reduction of Intracranial Pressure with Hypertonic Urea, Lancet, i, 1094.WALTON, J. N. (I956): ' Subarachnoid Hxmorrhage '. Edinburgh: Livingstone.

copyright. on A

pril 20, 2022 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.38.435.18 on 1 January 1962. D

ownloaded from