Long-Term Results of Endarterectomy of theInternal Carotid Artery for Cerebral

Ischemia and InfarctionBy ALBERT HEYMAN, M.D., W. GLENN YOUNG, JR., M.D.,IVAN W. BROWN, JR., M.D., AND KEITH S. GRIMSON, M.D.

SUMMARYThe immediate and long-term results of carotid endarterectomy for treatment of acute

cerebral ischemia or infarction were studied in 95 patients who were observed for anaverage period of 40 months after the surgical procedure. The operation resulted indeath for 12 patients and worsening of the neurological deficit in six others. Duringthe follow-up period, there were 22 additional deaths, most of which were caused bymyocardial infarction or other complications of systemic atherosclerotic vascular disease.The significant factors adversely affecting the prognosis of these patients consisted

of severe electrocardiographic abnormalities, the presence of bilateral carotid occlusivelesions, cerebral infarction prior to endarterectomy, and the age of the patient.

Following endarterectomy only five patients were observed to have recurrent attacksof transient cerebral ischemia which were few in number and stopped spontaneously.Recurrent cerebral infarction developed in only seven patients, six of whom had un-

successful or incomplete removal of the carotid occlusive lesions. Although this lowincidence of recurrent ischemia and infarction seems to be evidence for the therapeuticvalue of carotid endarterectomy, further observations on a comparable group of patientstreated nonsurgically are needed to support this conclusion.

Additional Indexing Words:Electrocardiograms Age MortalHypertension

C URGICAL REMOVAL of atheroscleroticL)occlusive lesions in the internal carotidarteries is now generally recommended as auseful measure for the treatment of acutetransient cerebral ischemia, but the long-termprognosis for patients undergoing such pro-cedures is not so well documented. Althoughthere have been several reports recently oflarge groups of patients with cerebrovascu-lar disease observed for 5 years or more afterthis form of treatment, -4 the lack of an ade-

From the Neurology and Vascular Surgery Divisionsof Duke University Medical Center and the Centerfor Cerebrovascular Research, Durham, North Caro-lina.

This work was supported by Grants NB-6233and HE-7369 from the National Institutes ofHealth, U. S. Public Health Service.

lity Neurological episode

quate population for comparison makes inter-pretation of the surgical survival rate difficult.In one study' the 5-year survival rate of pa-tients treated surgically for extracranial cere-brovascular disease was compared with thesurvival rate of the normal United Statespopulation of comparable age, with gross dif-ferences favoring the latter. In another study2of 175 patients undergoing carotid endarterec-tomy for cerebrovascular disease, only half thegroup, or 87 patients, were found to be curedor improved during an observation periodaveraging 26 months after the operation. Theremainder had died, worsened, or showed nochange. It is not known whether the outcomeof these patients following surgical therapyis better than that obtained with nonsurgicalmethods. Although a definite answer to this

Circulation, Volume XXXVI, Augurst 1967212

by guest on June 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

CAROTID ENDARTERECTOMY

problem must await appropriately controlledstudies of large numbers of cases, it should benoted that in most studies of carotid endar-terectomy the patients surviving the operativeprocedures seem to have a low recurrence rateof cerebral infarction.

This report presents our observations onpatients with extracranial cerebral occlusivedisease during a follow-up period of 2 to 8years after endarterectomy of the internalcarotid arteries. The prognostic factors influ-encing the outcome are also presented, par-ticularly as to the effects of the pre-existingneurological deficit, the extent of the occlu-sive disease in the carotid arteries, and thepresence of coronary vascular disease.

Cases StudiedNinety-five patients who had undergone en-

darterectomy for occlusive disease of one orboth internal carotid arteries at Duke UniversityHospital from 1958 to 1964 were studied. Eachof the patients was examined by one of us (A.H.)before the surgical operation and at regularintervals thereafter. Sixty-two of the patientswere men, 33 were women. All but two (bothNegroes) were white. The average age of thepatients was 59, with an age range from 42 to 81years. Approximately 15% of the patients wereunder the age of 49; an equal percentage wasover the age of 70 years. Of the remainingpatients an almost equal number were in the50 to 59 age range and in the 60 to 69 age range.Seventy-four of the patients had been admittedto the private service of the hospital and wereoperated on by one of three vascular surgeonson the senior staff. The remaining 21 patientswere on the public wards and had endarterectomyby one of the surgical house officers. The publicand private patients also differed somewhat inthe severity of their vascular disease. Althoughslightly younger than the private patients, thepublic group showed a greater percentage ofpatients with cerebral infarction in contrast totransient ischemia, and a higher incidence ofelectrocardiographic abnormalities indicating priorsevere myocardial disease.Twelve patients died shortly after the surgical

procedure, that is, before discharge from thehospital; 22 others died during the period offollow-up observation. The surviving 61 patientshave been observed for an average period of 40months: 17 of them 13 to 24 months, 24 for 25to 48 months, and the remaining 20 patientsfrom 49 to 96 months. Ninety-one of the 95patients had definite focal neurological mani-Circulation, Volume XXXVI, August 1967

festations of cerebrovascular disease prior tosurgical therapy; the remaining four had no neu-rological complaints. In these four "asympto-matic" patients, carotid endarterectomy wascarried out as a prophylactic measure, that is,to prevent cerebral infarction during subsequentsurgical operation for vascular disease of theabdominal aorta or renal arteries. The diagnosisof carotid occlusive disease was made in thesefour patients on the basis of loud carotid bruits,decreased ipsilateral ophthalmic artery pressures,and subsequent arteriographic examination.The predominant or presenting neurological

signs or symptoms consisted of weakness of thelimbs (45 patients), hemisensory loss (eightpatients), aphasia (four patients), transient al-terations of consciousness (four patients), andtransient monocular blindness or visual field de-fects (30 patients). These symptoms either ap-peared alone or in combination. Forty-nine pa-tients were considered to have recurrent transientcerebral ischemia, their neurological manifesta-tions having either cleared completely at thetime of surgery or were manifested by veryminor residual neurological deficits. A diagnosiswas made of cerebral infarction in the remaining42 patients who showed mild to moderatelysevere hemiplegia, sensory loss, or aphasia overa period of several days to weeks. Eighteen ofthese 42 patients were admitted to the hospitalwith what appeared to be gradually progressiveneurological deficit (thrombosis-in-evolution), butfor the purpose of this report they are consideredas having cerebral infarction. Each of these 91symptomatic patients had their last episode oftransient cerebral ischemia or the onset of cere-bral infarction within a few weeks prior tosurgical therapy. None of the patients was un-conscious or had severe mental changes at thetime of surgery. An attempt was made to excludepatients from this study whose major symptomsconsisted of vague dizziness, mental changes, orvisual disorders which were not clearly relatedto cerebral ischemic disease.

Both carotid arteries, as well as one or bothvertebral arteries, were visualized by arterio-graphic examination in 57 patients. Bilateralcarotid arteriography without vertebral arteryvisualization was carried out in 36 patients. Inthe two remaining patients, only one carotidartery was examined by arteriography becauseof sudden worsening during the arteriographicprocedure and the need for immediate surgicalintervention.

Patients with symptoms predominantly of ver-tebral-basilar artery insufficiency and those withsignificant subclavian arterial disease (as mani-fested by differences of blood pressure in thetwo arms) were excluded from this study to

213

by guest on June 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

HEYMAN ET AL.

obtain a homologous group of patients in whomthe clinical symptoms were definitely related toocclusive disease of the carotid arteries.

Lesions were present in only one carotid arteryof 62 patients, 16 of whom had complete ob-struction (hereafter called "occlusion") and 46of whom had partial obstruction (hereafter des-ignated as "stenosis"). Thirty-three others hadinvolvement of both carotid arteries: 11 withbilateral stenosis, two with bilateral occlusion,and 20 with stenosis of one internal carotidartery and complete occlusion of the contra-lateral internal carotid artery or common carotidartery. In all of the patients in this study at least30% of the diameter of one or both carotid ar-teries was obstructed by the atheromatous plaqueas shown by roentgenographic examination. Inmost of the patients, however, the degree ofstenosis was considerably greater. In almost everypatient, the obstructive lesion was located in theinternal carotid artery at or immediately distalto the bifurcation of the common carotid artery.In general, patients with complete arterial oc-clusion tended to present with cerebral infarctions,whereas those with stenotic lesions only, hadfewer cerebral infarcts and a higher percentageof transient cerebral ischemic attacks. Ten ofthe 95 patients gave a history of a previouscerebral infarction, usually a mild hemiparesiswhich occurred within 5 years prior to admissionto the study.

Severe electrocardiographic abnormalities werepresent in 29 patients indicating old myocardialinfarction, myocardial ischemia, or disturbancesin rhythm such as atrial fibrillation or flutter.The 17 patients with electrocardiographicchanges, such as questionable ventricular ische-mia, ventricular hypertrophy, or extrasystoles,were classified as having borderline abnormalitiesand are considered together with 42 patientswith normal records. Electrocardiograms werenot made on seven patients.Twenty-two of the patients had systemic blood

pressure persistently greater than 180/110 mmHg, and 35 others had moderately elevated bloodpressures, in the range of 150/90 to 180/110 mmHg. The remaining 38 patients were normo-tensive.

Surgical ProceduresEarly in the study, endarterectomy was car-

ried out under local anesthesia. In the last 1 or2 years of the study, general anesthesia wasemployed by all but one of the surgeons (K.S.G.)who prefers- cervical block anesthesia. Patientswith severe occlusive disease of both carotidvessels were usually operated on under hypo-thermic conditions, the body temperature beingredUiced to 32 C by surface cooling. In the

early part of the study, a shunt was insertedinto the internal carotid artery to provide circu-lation to the brain during clamping of the ves-sels and removal of the atheromatous plaque.Later in the study, however, shunts were notemployed, cerebral protection being providedby hypercarbia induced by rebreathing andmaintenance of blood pressure by hypertensiveagents. The use of vein grafts or plastic patchesto widen the lumen of the endarterectomizedvessel was a routine practice early in the studybut was rarely employed in the later period.The carotid artery was clamped during removalof the atheromatous material for 5 to 30 minutes,with an average of 15 minutes. In most instances,the atheromatous plaque in the carotid arteryproduced a greater degree of stenosis than wasapparent on arteriographic examination, theeffective lumen of the vessel often being assmall as 1 to 2 mm in diameter. In a smallpercentage of cases, possibly 10%, a fresh throm-bus was found attached to the plaque, andhemorrhage into the plaque was occasionallyobserved. In several instances, the plaque showedconsiderable grumose material which could havebeen the source for dissemination of plateletor cholesterol emboli to the brain.The operation was unsuccessful in removing

the lesions in 11 of the 16 patients with uni-lateral occlusion, in four of the 46 with unilateralstenoses, and in both patients with bilateral carot-id occlusion. In addition, endarterectomy ofboth carotid arteries was attempted in six ofthe 20 patients with unilateral stenosis and con-tralateral occlusion, but was unsuccessful in re-moving the complete occlusions in four of the sixpatients. Unsuccessful endarterectomy was us-ually evident during the operation of patientswith total occlusion since no back flow wasobserved from the intracranial or distal portionsof the internal carotid arteries. In the four pa-tients with unsuccessful operations on stenoticlesions, recurrent thrombi developed shortlyafter the surgical procedure as indicated bymeasurement of ophthalmic artery pressures,arteriography, or both.

Prior to endarterectomy, measurements ofophthalmic artery pressures provided useful di-agnostic information in 67 of the 92 patientsin whom this procedure was carried out. Ab-normal pressure readings (that is, differencesin either systolic or diastolic pressures of 15 gor more) were obtained in 34 of the 46 patientswith unilateral carotid stenosis, four of 11 withbilateral stenosis, 14 of 20 patients with bilateralocclusion or occlusion of one artery and stenosisof the other, and in each of 15 patients withunilateral occlusion. Ophthalmic artery pres-sures were also measured several times in each

Circulation, Volume XXXVI, August 1967

214

by guest on June 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

CAROTID ENDARTERECTOMY

patient during the first postoperative year todetermine whether the carotid arteries remainedpatent. In addition, arteriography or thermog-raphy was carried out in approximately onethird of the patients several months after theoperation. Only five of the patients were givenlong-term anticoagulant therapy postoperatively.A definite bruit was heard over one or both

carotid arteries in 71 of the 94 patients in whomauscultation of the neck vessels was carried outprior to the surgical procedure. In 14 of the 46patients with unilateral carotid stenosis, the bruitwas heard only over the site of the lesion. Bruitswere heard bilaterally in 18 of these 46 patientsand over the normal carotid artery in threepatients and were not heard in the remaining11 patients. Bruits were present in seven of the16 patients with unilateral occlusion and wereheard over the nonnal carotid vessel in fourpatients, over the obstructed artery in one, andbilaterally in two patients.

ResultsMorbidity and Mortality Related toEndarterectomyThe surgical procedure produced more se-

vere neurological manifestations in six of the95 patients. The neurological worsening wastransient in three patients but resulted inoptic atrophy in another patient and moderatehemiparesis in the other two patients. Fourother patients were found to have recurrentthrombosis at the site of the arteriotomy soonafter the surgical procedure, but none of themhad residual clinical manifestations.The operation was considered to be respon-

sible for deaths of 12 patients. Two of themdied as a result of myocardial infarction whichdeveloped a few days after operation. Infec-tion developed in the operative wound ofanother, with formation of a false aneurysm;this patient died after recurrent cerebral in-farction several weeks later. The nine remain-ing patients died soon after the surgicalprocedure as a result of anoxia or hypotension.Most of them failed to regain consciousnessafter the operation and developed severehemiplegia. Surgical re-exploration or autopsyin three of them revealed a fresh thrombus atthe site of the arteriotomy or in the contra-lateral carotid artery. One of the patients withhypotension during the surgical procedurewas found at autopsy to have a large freshCirculation, Volume XXXVI, August 1967

hemorrhage in the area of the cerebral in-farction.

Late Morbidity and Mortality

Of the 83 patients surviving the surgicalprocedure, only seven developed late neu-rological deficits, usually in the third orfourth year of the follow-up period (table 1).Six of these patients experienced hemiparesisof moderate severity (two of whom hadrepeated attacks of transient cerebral ischemiaafter operation). The remaining patient diedfollowing a massive brain stem infarction 4years after an unsuccessful carotid endarter-ectomy. As would be expected, a number ofother patients with severe cerebral infarctionslowly worsened over the years subsequent toendarterectomy, and several were eventuallyconfined to their homes or admitted to nursinghomes or mental institutions. As best could bedetermined, none of them had an overt re-currence of cerebral infarction, and their down-hill course could not be attributed solely toprogression of their cerebrovascular disease.

Five additional patients developed recurrentcerebral ischemia within 1 or 2 months afteroperation, but these episodes were few innumber and stopped spontaneously withoutresidual neurological manifestations. An addi-tional patient developed epileptic seizures. Asseen in table 1, six of the seven patients withrecurrent cerebral infarction had obstructivelesions in one or both carotid arteries, com-plete removal of which was either not at-tempted or was unsuccessful. Bilateral endar-terectomy was also unsuccessful in both ofthe patients with bilateral carotid occlusion.One of them was alive and well 3 years later,the other died 6 months after the operationas a result of myocardial infarction.As mentioned in the foregoing paragraph,

only one of the seven patients developinglate cerebral infarction died as a result ofthe recurrent episode. Of the remaining 21late deaths, nine were due to myocardial in-farction. Four patients died after other vascu-lar illnesses, such as congestive heart failure,renal insufficiency, mesenteric thrombosis, orcerebral hemorrhage. The remaining patients

215

by guest on June 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

t~~~~~~~~~~~~~~~~~~~~~~~c !,4c:

o oQC P4 ax 4 =oQ) Cr C°oz g:C r~~~Q = ) Dc .C t0 a o "° *0 .-0 ~~~~~~~~4cOa-4.CZa)a)* - - ~~~~~~~Q P 4 c

0. C; ,bO, etCct:

U 0 )~ v o4a S

00D C

Cl00 C~~~~~'~~j4 cisJ~C)0'

C)...4 - E.~-- .,IQ

a) c oO U CO a O $0

¢C Cd cn 4m t

'~~~~~~~~~- 4- Ci clt CZ

4 P-14-0 $n4~~~ 4 ,a-C's

a Q) C ,C

4) 4) 4) 4- a

o -~~~~~~~~~~~~-m~~~CIS 0CC)0

a.~)4 01 - dC> to) 0 0J

a) a)0 0

C004 0O4- 00 - 00 coa)

0a40 - ~ -coCO Ca .-

4-'0

cO CO~~~~~~~~~~~~~~~~~~~C

0~~~-

.~~CO Caa) ~~~~CO CJ a)0d c

4 -q4J 440-4Qa4JHo

C)~~~~~~~~~~~~~~~~~~~~~~~~~~~~~a

C) :z~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~aCld~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~C

4 00 10 1 ,CO 0 00)4

C~~~~~~10~ ~ ~ ~ ~C

aO Q- Cl O 'CZ

216 HEYMAN ET AL.

Csicuation, Volume XXXVI, August 1967

by guest on June 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

CAROTID ENDARTERECTOMY

Table 2Relation of Presenting Neurological Episode to Cerebral Infarction or Death afterEndarterectomy

Cerebral infarction Late cerebralPresenting associated with operation infarction Death due toepisode Patients Survived Died Survived Died other causes

Transient cerebralischemiaCerebral infarctionAsymptomatic

Total

49424

95

4206

460

10

2406

1001

1011*2

23

*Includes two patients dying of myocardial infarction shortly after endarterectomy.

Table 3Relation of Pattern of Carotid Occlusive Lesions to Cerebral Infarction or Death afterEndarterectomy

Carotid lesions

Unilateral stenosisUnilateral occlusionBilateral stenosisBilateral occlusionUnilateral stenosis withcontralateral occlusion

Total

*Includes one patient whoectomy.

Cerebral infarction Late cerebralassociated with operation infarction Deaths due to

Patients Survived Died Survived Died other causes

46 4 4 1 0 11*16 1 0 1 1 6*11 0 2 1 0 12 0 0 0 0 1

20 1 4 3 0 495 6 10 6 1 23

died as a result of myocardial infarction shortly after endarter-

Table 4Relation of Electrocardiographic Findings to Death after Endarterectomy

Associated DeathsECG

findings

Severe myocardial diseaseBorderline or normalNo data

Total

Patients

29597

95

Associatedwith operation

6 (1)*4 (1)212

Late

12 (4)10 (5)0

22

*Figures in parentheses indicate deaths known to be caused by myocardial infarction.

died as a result of neoplasm or leukemia (fivepatients) or from unknown causes (three pa-tients ) .

Influence of Presenting NeurologicalEpisode and Pattern of Carotid Lesions

Table 2 shows the relationship of the pre-senting neurological manifestations to sub-sequent cerebral infarction or death. Four(8%) of the 49 patients with transient cerebralischemia prior to operation developed seri-Circulation, Volume XXXVI, August 1967

ous cerebral manifestations and died fromthe surgical procedure. Of the 42 patientswith cerebral infarction prior to endarterec-tomy, six (14%) had worsening of the neurolog-ical manifestations immediately after the op-

eration and died within a few days. It is ofinterest that none of the four asymptomaticpatients developed cerebral infarction. Two ofthem died several years after the operation,however; one from myocardial infarction;the other from renal failure.

Total

18142

34

217

by guest on June 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

HEYMAN ET AL.

The relationship of cerebral infarction anddeath to the pattern and severity of the lesionsin the carotid arteries is seen in table 3.Our findings tend to support the current con-cept that the endarterectomy carries greaterrisks in patients with bilateral carotid lesions.Of the 62 patients with unilateral lesions(either occlusion or stenosis), four (6.4%)developed more severe neurological deficitswhich resulted in death shortly after endarter-ectomy. Among the 33 patients with bilateralcarotid lesions, six (19%) developed moresevere cerebral damage and died immedi-ately after the operation.

Effects of Electrocardiographic Abnormalitiesand Hypertension

Patients with electrocardiographic abnor-malities indicating severe myocardial diseaseprior to endarterectomy had an increased mor-tality in the early postoperative period as wellas during the later periods of follow-upobservation (table 4). Eighteen (62%) of the29 patients with marked electrocardiographicabnormalities died in contrast to only 14 (24%)of the 59 patients with normal or borderlineelectrocardiograms. Only five of the 11 patientsdying of myocardial infarction, however,had severe electrocardiographic abnormalitiesprior to the surgical procedure; in the re-maining six, the electrocardiogram wasborderline or normal. Of the two patientsdying of myocardial infarction in the immedi-ate postoperative period, one had severe an-gina and a normal electrocardiogram. Theother had no clinical evidence of myocardialdisease but showed severe electrocardio-graphic abnormalities. He died suddenly 1week after an unsuccessful endarterectomyfor removal of a unilateral carotid occlusion.The patients with hypertension did not

show an increase in mortality. Of the 22patients with arterial blood pressure levelsgreater than 180/100 mm Hg, eight died (fiveearly and three late). There were 11 deathsamong the 35 patients with blood pressures inthe range of 150/90 to 180/110 mm Hg, and15 deaths among the 38 normotensive pa-tients.

Influence of Age and Socioeconomic Status

Of the 16 patients over 70 years of age inthis study, five died soon after the surgicalprocedure and three others succumbed dur-ing the follow-up period. This mortality rate(50%) is, as would be expected, greater thanin the younger age groups in which deathoccurred in approximately one third of thepatients in each of the three age decadesbelow 70 years.The 21 patients admitted to the public

wards of Duke Hospital and operated on bythe resident staff had a higher percentage ofdeaths (five early and five late) than the74 private patients (seven early and 17 late).As mentioned earlier, the public patients hada greater incidence of cerebral infarction thanthe private ones and, probably more signifi-cantly, showed a higher percentage of electro-cardiographic abnormalities indicating severemyocardial disease.

Discussion

The results of this study demonstrate avery low incidence of recurrent cerebral in-farction in patients surviving carotid endarter-ectomy. Of the 83 patients in this study sur-viving the surgical procedure, only seven (mostof whom had unsuccessful or incomplete re-moval of the carotid lesions) developedcerebral infarction during the follow-up pe-riod. This low rate of recurrent infarction hasbeen observed by others and has been con-sidered as strong evidence for the therapeuticvalue of carotid endarterectomy. Unfortunate-ly, there are no reports of large numbers ofcomparable patients treated nonsurgically, andthe incidence of recurrent cerebral infarctionduring the natural course of extracranial cere-bral occlusive disease is not known. Althoughprevious studies have shown that approxi-mately 20% of patients with cerebral infarc-tion develop recurrent cerebral infarctionwithin 2 years, such case studies cannot beused for comparison.5 Many of the patientsincluded in such reports had severe cerebraldamage (as a result of one or more previousinfarcts), and few of them had arteriographicdemonstration of carotid arterial disease.

Circulation, Volume XXXVI, August 1967

218

by guest on June 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

CAROTID ENDARTERECTOMY

Recurrent attacks of cerebral ischemia werealso uncommon following endarterectomyand in most instances the ischemic attackscompletely stopped after the surgical pro-cedure. There is, however, little informationas to the natural course of transient cerebralischemia caused by extracranial vascular dis-ease. The few recent papers on this subjectindicate that patients with frequent attacksof carotid arterial insufficiency may have arelatively benign prognosis, the attacks oftenstopping spontaneously without producing in-farction.6 Few of the patients with transientcerebral ischemia studied for long periodshave had arteriographic evaluation of theextracranial cerebral circulation, and the out-come or natural course of such patients like-wise cannot be compared with that of thesurgically treated patients.The prognosis for patients with extracranial

carotid occlusive disease is extremely variableand depends on numerous factors. The out-come of large numbers of surgically treatedand nontreated patients must therefore becompared before the results of endarterectomycan be considered reliable. Bauer and hisassociates" reported the immediate and lateoutcome of a relatively small group of pa-tients with atherosclerotic cerebral occlusivedisease who were selected for surgical andnonsurgical treatment on a random basis. Intheir study, the percentage of patients whodied after carotid endarterectomy (30.3%)was higher than-that (21.2%) of the controlgroup who were followed for as long as 42months. It is conjectural whether a longerperiod of follow-up will eventually demon-strate a greater survival rate among the surg-ically treated group. Large scale cooperativestudies like those now in progress may providean answer to this problem.8The low incidence of recurrent cerebral

infarction following successful carotid endar-terectomy is generally attributed to the im-provement of the cerebral circulation in thesecases. Conversely, six of the seven patientsin this study who developed recurrent cerebralinfarction had incomplete resection of thecarotid obstructive lesions. This observationCirculation, Volume XXXVI, August 1967

suggests that failure to remove all such surg-ically accessible lesions is the cause of recur-rent cerebrovascular manifestations. It shouldbe noted, however, that 29 other patients inthis study also had unsuccessful or incompleteremoval of their carotid lesions, but these pa-tients had no worsening or recurrence of theirneurological deficit. Other investigators9 havealso shown that unilateral or even bilateralocclusion of the carotid arteries may not indi-cate a serious prognosis. Yates and Hutchin-son'0 found cerebral infarction in only six ofthe 18 autopsied cases in which significantarterial stenosis was confined to one or bothcarotid arteries. In four of the six patients,moreover, the intracranial artery supplyingthe infarcted territory was also obstructed.Considerable other evidence also indicates thatif only one carotid artery is occluded and therest of the cerebrovascular tree is healthy,the patient is often able to survive and befree of neurological dysfunction.The known coexistence of cerebral and

coronary atherosclerosis is confirmed in thisstudy as shown by the fact that a significantnumber of our patients had evidence of myo-cardial infarction before or after endarterec-tomy. Approximately one third of the patientshad electrocardiographic changes indicatingsevere myocardial disease prior to endarterec-tomy, and 11 patients subsequently died as aresult of myocardial infarction. In DeBakey's' follow-up study of surgical treatmentof extracranial cerebrovascular disease, the5-year mortality rate of patients with arteri-osclerotic heart disease was two to threetimes that of patients without hypertensionor heart disease. Other studies of carotidendarterectomy have shown that among pa-tients with normal electrocardiograms, 78%had a favorable outcome in contrast to 53%among patients wi-th electrocardiographic evi-dence of old myocardial infarction.2 It is thusapparent that greater attention should begiven to the recognition and treatment ofcoronary artery disease as well as the thrombo-embolic complications of generalized arterio-sclerosis if the overall mortality following

219

by guest on June 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

HEYMAN ET AL.

carotid endarterectomy is to be significantlyreduced.

In the present study, the evaluation ofsurgical therapy is based largely on the in-cidence of recurrent cerebral infarction andsurvival following endarterectomy. The de-gree of immediate improvement or gradualrecovery of the neurological manifestationspresent at the time of the surgical procedurehas not been considered in evaluating thisform of treatment, since the degree of naturalrecovery of the neurological dysfunction isknown to vary considerably. We have not ob-served any dramatic postoperative neurologi-cal improvement which could be definitelyattributed to the effect of surgery. Althougha few patients appeared to show considerableimprovement of the hemiplegia or aphasiawithin a few days or a week after endarterec-tomy, we have observed such recovery duringthe natural course of the disease as well as inpatients in whom endarterectomy was un-successful in removing the obstructive lesion.Thus, the reports in which the value of surgicaltherapy is largely based on "improvement"or "recovery" from hemiplegia or other neu-rological deficits are open to criticism.4The overall survival rate in our patients 40

months after the operation, as determined bylife table analysis, was 656 per 1,000, a figureless satisfactory than that of 768 per 1,000found in the study by DeBakey and asso-ciates' of a larger series of patients of thesame mean age. The percentage of deathsin our study (35.8%) observed an averageof 42 months after operation may be com-parable to that observed by Yashon and hisassociates2 who reported that 25% of theirpatients died within an average of 26 monthsafter endarterectomy. The data obtained inthe present study indicate that differences inmortality rates are probably due to factorssuch as the presence of cerebral infarction,the existence of severe electrocardiographicabnormalities, the severity of occlusive le-sions in the carotid arterial system, and theage of the patient. In the past few years,the acute morbidity and mortality associatedwith carotid endarterectomy in our hospital

have shown definite improvement, presum-ably as a result of more careful case selection,greater surgical experience, and perhaps morecomplete pre-surgical arteriographic evalua-tion. Indeed, eight of the 12 deaths associatedwith the endarterectomy occurred during theinitial portion of this study. Many of thesepatients were over the age of 70 years, hadsevere hypertension, and would not be candi-dates for operation at the present time.

Currently, "four vessel" arteriography (thatis, visualization of the extracranial and intra-cranial portions of both vertebral and carotidarteries) is generally carried out to assessthe overall state of the arterial supply tothe brain. In the present report, no mentionhas been made of the presence of stenoticlesions in the vertebral and other extracranialor intracranial vessels which were sometimespresent but were not considered to be di-rectly related to the patient's symptoms. Itmust be admitted, however, that the exact re-lationship of such multiple lesions to the pa-tient's symptoms and to his eventual outcomeis difficult to assess. It is known that symptomsreferable to one area of the brain may resultfrom an occlusive lesion in distant collateralchannels. In the present study, carotid endar-terectomy was carried out only when thereseemed to be a clear-cut relationship betweenthe patient's neurological deficit and the vas-cular lesion. Moreover, in patients with bi-lateral carotid lesions, an attempt was madeto limit the operation to the carotid arterycorresponding to the patient's symptoms. Noneof our patients had endarterectomy of thevertebral arteries, nor was any attempt madeto operate on lesions in the contralateralcarotid artery once the patient's presentingsymptoms had improved. Since the presenceof multiple extracranial occlusive lesions areresponsible for the overall decrease in cere-bral blood flow,'1 many investigators believethat endarterectomy should be carried outon as many such lesions as feasible regardlessof their apparent relationship to the patient'spresenting symptoms. The results obtainedin the present study indicate that multiple

Circulation, Volume XXXVI, August 1967

220

by guest on June 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

CAROTID ENDARTERECTOMY

operations are not usually necessary to pro-duce satisfactory outcome, since relativelyfew of our patients had recurrent cerebralinfarction even though endarterectomy wasgenerally confined to the one lesion whichseemed to be responsible for the acute symp-toms.Not only is there some controversy as to

the number of extracranial vascular lesionswhich should be operated on, but there is thequestion as to the need for resection of asolitary stenotic lesion which appears to beproducing relatively little obstruction of thevascular lumen. Flowmeter studies haveshown that interference with carotid bloodflow occurs only in the presence of verylarge intravascular lesions which must reducethe lumen to approximately 2 sq mm beforeproducing a significant reduction in circula-tion.12 On the other hand, an irregular, friablearteriosclerotic plaque of almost any size mayprovide a site for formation of emboli whichmay be carried to the brain and cause symp-toms of transient cerebral ischemia or in-farction. Thus, the indication for endarter-ectomy of a solitary lesion depends uponwhich of these two factors, that is, a reductionin blood flow or a site for embolic forma-tion, seem to be operable in a given individ-ual-a decision which is often difficult tomake. Further investigation of this problemis needed.

AddendumFollowing submission of this paper for pub-

lication, our attention was called to a relatedreport by Bradshaw and Casey.'3 This study of47 patients with carotid obstructive lesions, fol-lowed for 6 months to 10 years, found that sevenpatients died of related disease and that morethan 50% of the survivors had an overall satis-factory neurological outcome.

AcknowledgmentWe are indebted to Mrs. Jane Sayer and Mrs.

Margaret Rossett for their valuable assistance.

References1. DEBAKEY, M. E., CRAWFORD, E. S., COOLEY,

D. A., MoRms, G. C. J., GARRECrr, H. E.,AND FIELDS, W. S.: Cerebral arterial insuffi-ciency: One to 11-year results followingarterial reconstructive operation. Ann Surg 161:921, 1965.

2. YASHON, D., JANE, J. A., AND JAVID, H.: Longterm results of carotid bifurcation endarterec-tomy. Surg Gynec Obstet 122: 517, 1966.

3. THOMPSON, J. E., KARTCHNER, M. M., AusTIN,D. J., WHEELER, C. G., AND PATMAN, R. D.:Carotid endarterectomy for cerebrovascularinsufficiency (stroke): Follow up of 359 cases.Ann Surg 163: 751, 1966.

4. GuRDJIAN, E. S., DARMODRY, W. R., LINDNER,D. W., AND THOMAS, L. M.: Fate of patientswith carotid and vertebral artery surgery forstenosis or occlusion. Surg Gynec Obstet 121:326, 1965.

5. DAVID, N. J., AND HEYMAN, A.: Factors in-fluencing the prognosis of cerebral throm-bosis and infarction due to atherosclerosis. JChronic Dis 2: 394, 1960.

6. ACHESON, J., AND HUTCHINSON, E. C.: Ob-servations on the natural history of transientcerebral ischaemia. Lancet 2: 871, 1964.

7. BAUER, R. B., ET AL: A controlled study ofsurgical treatment of cerebrovascular disease:42 Months experience with 183 cases, cerebralvascular diseases. In Cerebral Vascular Diseases,Transactions of the Fifth Conference, editedby R. G. Siekert and J. P. Whisnant. NewYork, Grune & Stratton, Inc., 1966.

8. FIELDS, W. S.: Progress report of the joint studyof extracranial arterial occlusion, cerebral vas-cular diseases. In Cerebral Vascular Diseases,Transactions of the Fifth Conference, editedby R. G. Siekert and J. P. Whisnant. NewYork, Grune & Stratton, Inc., 1966.

9. WHISNANT, J. P., MARTIN, M. J., AND SAYRE,G. P.: Atherosclerotic stenosis of cervicalarteries. Arch Neurol (Chicago) 5: 429, 1961.

10. YATES, P. O., AND HUTCHINSON, E. C.: Cerebralinfarction: Role of stenosis of the extracranialcerebral arteries. Med Res Coun Spec Rep(London) 300: 1, 1961.

11. ADAMS, J. E., SMITH, M. C., AND WYLIE, E. J.:Cerebral blood flow and haemodynamics inextracranial vascular disease: Effect of en-darterectomy. Surgery 53: 449, 1963.

12. BRICE, J. G., DOWSETT, D. J., AND LOWE, R. D.:Haemodynamic effects of carotid artery stenosis.Brit Med J 2: 1363, 1964.

13. BRADSHAW, PETER, AND CASEY, EoIN: Outcomeof medically treated stroke associated withstenosis or occlusion of the internal carotidartery. Brit Med J 1: 201, 1967.

Circulaton, Volume XXXVI, August 1967

221

by guest on June 14, 2018http://circ.ahajournals.org/

Dow

nloaded from

S. GRIMSONALBERT HEYMAN, W. GLENN YOUNG, JR., IVAN W. BROWN, JR. and KEITH

Cerebral Ischemia and InfarctionLong-Term Results of Endarterectomy of the Internal Carotid Artery for

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1967 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.36.2.212

1967;36:212-221Circulation. 

http://circ.ahajournals.org/content/36/2/212located on the World Wide Web at:

The online version of this article, along with updated information and services, is

  http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. and Rights Question and Answer

Permissionsthe Web page under Services. Further information about this process is available in thewhich permission is being requested is located, click Request Permissions in the middle column ofClearance Center, not the Editorial Office. Once the online version of the published article for

can be obtained via RightsLink, a service of the CopyrightCirculationoriginally published in Requests for permissions to reproduce figures, tables, or portions of articlesPermissions:

by guest on June 14, 2018http://circ.ahajournals.org/

Dow

nloaded from