12 - Surgical Measures to Prevent Stroke

Editors: Norris, John W.; Hachinski, Vladimir

Title: Stroke Prevention, 1st Edition

Copyright 2001 Oxford University Press

> Table of Contents > II - Secondary Prevention > 12 - Surgical Measures to Prevent Stroke


Surgical Measures to Prevent Stroke

H. J.M. Barnett

Heather E. Meldrum

Surgical repair of injured arteries on the battlefields World War II set the stage for a surgical attack on arteries to the limbs, the heart, and the brain narrowed by arteriosclerosis. The cervical portion of the internal carotid artery had been shown by Egaz Moniz, the father of angiography, the pathological studies G.T.J. Hultquist, and by the dedicated studies by C. Miller Fisher of the neck arteries at post-mortem to be a commoner site of obstruction than had previously been accepted.1,2,3 The middle cerebral artery, instead of being seen as the primary site of arteriosclerotic mischief, came to be regarded as the artery most commonly causing stroke from lodgment of cardiac and extracranial carotid embolic material. Except in Asian populations, the middle cerebral artery was no longer accepted as the primary site of obstruction due to arteriosclerosis.

As Goethe remarked, What one knows, sees. Quite quickly, an explosion of reports confirmed Moniz's, Hultquist's, and Fisher's observations. As more arterial lesions were identified in the neck, the phenomena of transient and recurrent ischemic events came to light. The stage was set for attempts forestall, by antithrombotics and surgical methods, the strokes threatened these transient events of cerebral and retinal ischemia. This chapter confines itself to the evolution of understanding the appropriate use of surgical measures to prevent stroke.


Development of Carotid Artery Surgery

The credit for the first carotid surgical attempt to prevent ischemic stroke must go to Eastcott, Pickering, and Rob.4 They were the first to publish a report in a peer-reviewed journal, the gold currency of scientific activity. Carotid endarterectomy (CE) quickly replaced localized carotid resection. Many surgeons learned the technique. Neurologists and other practitioners became aware of the significance of focal hemisphere and retinal events. By 1974, 15,000 carotid procedures had been done in the non-veteran hospitals of United States. A small, randomized trial of 316 patients reported results in 1970 that were far from encouraging.5 Too many patients (42%) had vertebral basilar, rather than carotid, symptoms, loss to follow-up was higher than desirable, and the perioperative complication rate was unacceptable (11%).

Despite this negative trial, enthusiasm persisted. It still appeared to be a logical procedure for preventing stroke. Other applications of the were introduced that led to its expanded use. One application arose from the hypothesis that nonspecific symptoms, such as dizziness, vertigo, drop attacks, episodes of loss of consciousness, transitory memory difficulties, and even specific symptoms of vertebral basilar disease, might benefit from a general improvement in blood flowing to the brain by removing the accessible stenoses from diseased carotid arteries. As many as 10% of the increasing number of endarterectomies may have related to this improbable and unproven hypothesis. Another practice that became common was to remove an asymptomatic stenosing plaque as a prelude to major cardiovascular operations, particularly coronary artery bypass grafting. More importantly, the use of the procedure was extended to asymptomatic individuals with carotid artery stenosis. This custom is discussed in detail later in the chapter.

Another innovation was the introduction of ultrasound as an alternative diagnostic tool to be used instead of or with angiography. Safe and noninvasive, it became widely available and very extensively used. More will be said of this but its complete role in patient selection remains uncertain and in some dispute.

In the 11 years between 1974 and 1985, it was estimated that one million individuals worldwide had carotid endarterectomy.6 Figure 12.1 illustrates that enthusiasm increased dramatically in this 11-year period. With doubts expressed about proof of its suitability for its widespread application, the numbers waned for a few years. They doubled again within two or three years of the completion of published trials because of an optimistic interpretation the trial results. By 1999, it was quite certain that two million carotid endarterectomies had been performed worldwide and that a reasonably large number of these (probably at least one-third) were done without firm substantiating evidence, some in situations which the harm may have exceeded benefit.

FIGURE 12.1. The number of endarterectomies performed annually from 1970 to 1995 indicates a decline after 1985, followed by a doubling after the publication of the first phase of NASCET and ECST the final results of ACAS. Adapted from Pokras R., personal communication with permission.


Later Trials of Carotid Endarterectomy

To fulfill the demands of modern scientific medicine, particularly when the course of a chronic disease is unpredictable and has protean outcomes, therapeutic data must be amassed from randomized trials that are designed, executed, and analyzed to answer the appropriate questions. Three such trials were launched and concluded to evaluate the role of CE in patients with symptomatic disease.7,8,9 Two of these three trials prospectively followed the stroke outcomes relating to the stenosing lesions on the asymptomatic other side. Five were launched for asymptomatic individuals, of which four are concluded and published.10,11,12,13,14 Finally, a study in which symptomatic patients and asymptomatic subjects were randomized to determine the optimum perioperative dose of aspirin allowed some observations to be made about asymptomatic disease.15

Symptomatic Trials

In 1981, Warlow and colleagues took up the challenge of appropriateness carotid endarterectomy and from the United Kingdom launched an ambitious trial. The European Carotid Surgery Trial (ECST)7 required that patients have symptoms within 180 days and have angiographic proof of a stenosis caused by arteriosclerosis. Two patients were randomized to receive endarterectomy for


every patient randomized to treatment with best medical care alone. Between 1981 and 1994, 3024 patients were randomized.

The other large trial, the North American Symptomatic Carotid Endarterectomy Trial (NASCET), 8 randomized 2885 patients between 1987 and 1996. All patients were required to have had recent (120 days for the first 30 months of the trial, 180 days thereafter) focal symptoms appropriate to an angiographically demonstrated stenosing lesion. Surgeons were required to produce evidence of experience and competence in the performance of CE. For each center, randomization


balanced one medical to one surgical patient. The formula for measuring stenosis used the narrowest diameter of the stenosed segment as numerator, and the artery diameter well beyond disease, bulb, and post-bulbar dilatation as the denominator. Selected films were submitted for the central measurements used in all analyses. The formula used to calculate the measurement ECST used a different denominator (Fig. 12.2). The differences have been sufficiently studied and rationalized so easy conversion is possible from the degrees of stenosis depicted in Table 12.1.

TABLE 12.1. Correlation of NASCET and ECST Measurements of % Stenosis

30 65
40 70
50 75
60 80
70 85
80 91
90 97
NASCET found benefit from 50% 99%; greatest benefit 70% 99%. ECST found benefit at and above 80%(NASCET 60%).

FIGURE 12.2. NASCET calculated the percentage of stenosis by using narrowest linear diameter in two (or three) planes as the numerator, and the artery well beyond the carotid bulb (circled) and any poststenotic dilatation as the denominator (closed arrow) (1 N/D 100 = % stenosis). Measurement of the artery too close to bulb produced a misleading denominator (asterisk). The European Carotid Surgery Trial (ECST) used the same numerator but took the imagined site of the wall carotid bulb as denominator. Reproduced with permission from

Barnett et al. Current Critical Problems in Vascular Surgery, St. Louis, MO: Quality Medical Publishing, 1994.

Best medical care included therapy to normalize blood pressure, cholesterol, and blood sugar and advice about the avoidance of tobacco. Enteric-coated aspirin was supplied to all NASCET patients at a recommended dosage of 1300 mg daily. Lesser amounts were discretionary, as was the use of other platelet inhibitors. Excluded were patients with symptoms due to nonarteriosclerotic disease of the carotid arteries, previous endarterectomy on symptomatic side, heart disease likely to be associated with early cerebral embolic events, and organ failure or other morbid illnesses likely to cause death within five years. Outcome events in NASCET were adjudicated by independent clinicians blinded to the treatment arm. Data were gathered at entry and at the time of outcome events to assist the NASCET investigators in determining cause, severity, and course of stroke and particularly in identifying the occurrences of hemorrhagic, large artery, lacunar, or cardioembolic stroke.

In neither ECST nor NASCET was ultrasound evaluation alone acceptable for randomization. It was encouraged in the preliminary screening of potential candidates, and in NASCET was performed at the first follow-up after CE and requested annually in all patients.

The third symptomatic trial, confined to 16 Veterans Affairs (VA) Medical Centers in the United States, randomized 189 symptomatic patients with >50% stenosis


determined by angiography. The numbers were small, and the trial was discontinued after the declaration of benefit from CE for patients with 70% 99% stenosis.

The Results of the Trials for Symptomatic Patients

The benefit of CE was similar in the two large trials. Figure 12.3 demonstrates the five-year Kaplan-Meier survival curves from NASCET for any stroke and death and for disabling stroke death in the patients 70% 99% the 50% 69% stenosis range. It is clear that patients with focal symptoms who have severe (70% 99%) stenosis benefit most from the procedure. In the patients with moderate stenosis, the confidence intervals overlap, and the benefit, although present at a p-value of 0.045, is muted.

Several conditions were identified in post-hoc observations of perioperative risk from CE that led to a doubling, or near doubling, of the likelihood postoperative outcome complications.16 They included a hemispheric TIA compared with a retinal TIA as the qualifying event, left-sided procedure, the presence of contralateral carotid occlusion, an ipsilateral ischemic lesion on the entry CT scan, and irregular or ulcerated plaque detected by angiography on the side of surgery and a lack of collateral circulation in the hemisphere beyond severe (70% 99%) symptomatic stenosis.17 Despite imposing an increase in risk at the time of the procedure, none these conditions contraindicate CE because, in long-term follow-up of patients with both severe and moderate stenosis, these variables did not eliminate the benefit favoring surgery. The most unexpected of these observations was the increased complication rate for left-sided lesions. The initial impression was that this could be due to a statistical spin of chance. This probably is not the explanation, as it has been encountered in two other series of patients submitted to CE.15,18 In patients who had contralateral occlusion (n = 43), the operative risk was tripled (14%) compared to those patients without contralateral occlusion. Treated medically, patients with this condition have a 69% risk of stroke in two years compared to 29% for patients without contralateral occlusion.19

FIGURE 12.3. (A) Kaplan-Meier curves for event-free survival of medical and surgical patients with severe (70% 99%) stenosis. The results favor endarterectomy for any ipsilateral stroke and disabling ipsilateral (p < 0.001). The 95% confidence intervals are shown for each curve and there is no overlap. The number at risk annually are shown out to 8 years, (b) Kaplan-Meier curves for event-free survival of medical and surgical patients with moderate stenosis (50% 69%). The results favor endarterectomy (p = 0.045), but the confidence intervals overlap throughout, muting importance of the statistical significance. The results for disabling stroke showed no benefit. Reproduced by permission of the New England Journal of Medicine.8



In the severe range (70% 90%), all risk subgroups benefited from procedure. In the moderate range (50% 69%), best results were detected in males, patients with tandem lesions (intracranial disease), in patients with hemisphere rather than retinal symptoms, and in patients presenting with a stroke rather than only transient ischemic attack (TIA). Below 50% stenosis, symptomatic patients received no benefit and were best treated with medical care; the ECST study found that these individuals were significantly worse off with CE than with best medical care.

For patients at ranges of stenosis <60%, 60% 69%, and 70% 99%, outcomes were analyzed as to cause of stroke (large artery, lacunar, or cardioembolic in origin). The breakpoint at 60% was used because this was the minimal degree of stenosis in the Asymptomatic Carotid Atherosclerosis Study (ACAS) trial, and above it a benefit had been claimed for asymptomatic individuals from CE. In Figure 12.4, 83.6% of subsequent strokes were large artery origin when the stenosis was severe (70% 99%). The combination of cardioembolic and lacunar ipsilateral strokes together constituted only 16.4% in patients with this severe degree of stenosis. Strokes non-large artery origin rose to 40% in the patients with moderate stenosis. Lacunar strokes as outcome events were commoner


than cardioembolic strokes. In large hospital-based and community-based stroke data banks, these two stroke causes are close to equal. The percentage of cardioembolic strokes in NASCET is lower because the protocol excluded patients who had evidence at entry of cardiac lesions likely to lead cerebral embolization. The age of the patients in NASCET, however, determined that heart disease and small vessel disease became manifest in the long-term follow-up of these patients. The cause of strokes in asymptomatic patients is discussed the next section.

FIGURE 12.4. Kaplan-Meier proportion (%) of each cause (cardioembolic, lacunar, largeartery) of ipsilateral stroke in the symptomatic arteries and in the asymptomatic other side. The cutpoints at 60% and 70% stenosis correspond to the current American Heart Association guidelines for recommending carotid endarterectomy to asymptomatic and symptomatic patients, respectively.47

The degree of stenosis proved to be the most important risk factor (Figure 12.5). The average annual risk of stroke peaked at 9.4% for those patients with 75% 94% stenosis. In the nearly occluded patients with a collapsing artery (95% 99%), the risk diminished again (6.5%). This decreased risk in the patients with very severe stenosis has been previously reported from NASCET.20

Asymptomatic Carotid Artery Disease

Angiography and noninvasive studies detect atherosclerotic plaque in the extracranial arteries, which increases in amount with the subject's age. From a small number of community surveys, it has been estimated that 2 million people above the age of 50 in the United States have an asymptomatic carotid artery stenosis.21


This means that there are twice as many people living with this condition in the United States at the present time as were submitted to CE worldwide in the first thirty years of its application.

FIGURE 12.5. The average annual risk of stroke in the involved artery territory over a three-year period is depicted for increasing degrees of stenosis. The rear bars relate to the strokes of the symptomatic artery, the front bars to territory asymptomatic artery on the other side. No symptomatic patient had normal arteries, and the NASCET protocol excluded patients with occlusion of symptomatic arteries. Reproduced by permission of REMEDICA.

The question of appropriate treatment in the presence asymptomatic lesions has been the subject of considerable enquiry. Uncertainty continues to surround the application of CE for these subjects, who in this discussion will not be described as patients : they carry on normal lives with freedom from symptoms, and the likelihood is that a majority will never require aggressive therapy. The assumption that they are all patients requiring CE would increase by 10-fold the number currently being performed.

The risk of stroke facing these individuals was subjected to enquiry in large observational case-series. The best data came from approximately 1000 subjects in two studies who came under observation in two vascular ultrasound imaging laboratories and were examined at intervals over a five-year period.22,23 These two studies concluded that the annual risk of stroke with lesions producing stenosis below 75% and 80%, respectively was less than 2% per year and rose to 5% only when the stenosis was above 80%. This risk for stroke decreased to 3% when only events ipsilateral to the lesion were included. Men were at higher risk than women. Some other observations about the risk came to light from Toronto asymptomatic series. In a mean follow-up of 52 months involving 500 persons, Norris and Zhu detected a bi-modal distribution of ischemic events, observing a peak risk in the 75% 90% range of stenosis, and then reduced in subjects with 91% 99% stenosis.24 Fifty-one cerebral ischemic events were reported, most in the 75% 90% range. TIA, a harbinger of potential stroke, was four times as likely to occur (n = 40) as was ischemic stroke 11). From this study and from other reports, evidence indicates that there is an increased likelihood of the development symptoms when the lesion under observation increases rapidly in size. The postulate is that the increase size due either to rapid increase in atheromatous deposit or to intraplaque hemorrhage. More data are needed in this regard. It would appear to be an uncommon occurrence. NASCET has data indicating that the average rate of increase in stenosis is only 4% per year in long-term annual ultrasound observations.25 The question remains unsettled about how often individuals with known stenosis should be reexamined for evidence of rapid increase while still asymptomatic. Studies at less than annual intervals appear to be excessive.

The gold standard of the randomized clinical trial has been applied in evaluating both medical and surgical therapy for asymptomatic subjects. The number of subjects has been smaller than ideal. The one medical trial, involving 372 subjects, was of meticulous design but was unable to determine any benefit from prophylactic aspirin therapy (325 mg daily) compared to placebo.26 Larger studies are required to confirm or deny these observations, possibly using different dosages of aspirin.


Five randomized trials evaluating surgical treatment have been started, and four of these have been completed and published. Table 12.2 sets out the numbers admitted to the relevant trials and contrasts those with symptomatic disease those with no symptoms. It identifies the outcome events of any stroke, vascular death, and other death extracted from the trials in which the information was available. Clearly, the symptomatic patients face an increasing annual risk of stroke and vascular death with the increasing degree of stenosis. The annual stroke risk in NASCET's medically treated symptomatic patients was 9.6% in those with severe disease and 4.7% in those with moderate disease. In ECST, the figures were 6.2% and 2.6% respectively. In asymptomatic subjects the annual risk varied between 2.6% and 2.9%. To improve on these medical risk figures with CE is realistic for the severe patients, as shown in ECST and NASCET, for some moderate patients, as shown in NASCET. The challenge appears to be more daunting and may be impossible when the medical risk figures are below 3%, as they were in the asymptomatic trials and in the moderate patients ECST. The trials involving asymptomatic subjects numbering in aggregate 2584 involved less than half the numbers in the trials of symptomatic patients (n = 6092). The lesser risk of stroke experienced in the asymptomatic artery suggests that larger trials would have been desirable. Perhaps as many 10,000 subjects may be needed to establish efficacy with certainty. Whether seeking benefit by using such large numbers of subjects with such low risk of stroke is worthwhile remains in doubt.27

Two of the asymptomatic trials had sufficient design problems that they will not be further considered.10,11 The Veterans Administration Cooperative Study trial randomized 444 men, and had it not been accompanied by a high perioperative risk, modest benefit in stroke reduction might have been shown. With a 4.4% complication rate, neither immediate nor long-term stroke-free survival resulted.

The AC AS study randomized 1662 individuals, 3 of whom in the surgical arm were not followed. Approximately 45% had previous symptoms or surgical procedures. The final publication reported a relative risk reduction of 53% favoring stroke-free survival for the subjects in the surgical arm compared to those medical arm. Unhappily, the absolute risk reduction was only 5.9% at five years, yielding little more than an annual risk reduction of 1%. The difference in the impact of relative and absolute risk reductions are seldom better demonstrated than in this differential of 53% and 1%. The relative risk reduction may be unduly optimistic. A statistically significant difference may not amount to an important clinical difference. For example, a relative risk reduction of 50% is obtained when the absolute risk is reduced from 50% to 25%. A 50% relative reduction also ensues when the absolute risk is reduced from 2% to 1%. In such low-risk situations, as in subjects who have asymptomatic carotid stenosis, attention should concentrate on the absolute reduction of risk.

The benefit claimed for CE based on the ACAS reports has not received universal acceptance.6,28,29,30,31,32,33 The controversies of interpretation arise from a combination



of what was shown, what was not shown, what has not been extrapolated lacking good data, and what has been observed from other databases.

TABLE 12.2. Number of Patients and Outcomes in the Symptomatic and Asymptomatic Randomized Trials

Symptomatic Trials ECST7
   Moderate 1057 1533    6 yrs. 163 234 122 219 282 418 4.5 2.6
   Severe 154 274    6.6 yrs. 63 40 32 55 40 80 9.3 6.2
   Moderate 1118 1108    5 yrs. 264 233 123 130 108 88 6.9 4.7
   Severe 331 328    2 yrs. 64 34 14 10 7 5 11.8 9.6
VA9* 98 91    lyr. 7 3
Total 2758 3334 551 585 371 494 203 237
Asymptomatic Trials ACAS13 834 825    2.7 yrs. 66 44 60 46 29 37 5.5 2.9
VA Cooperative12 233 211    4 yrs. 25 17 51 45 27 25    8 2.6
CASANOVA10 204 206 3 yrs.
MACE11 35 36    2 yrs. 0 3 0 0 0 0    0    0
Total 1306 1278 91 64 111 91 56 62
*Data unavailable for detailed breakdown of outcome events.
Additional data supplied from ECST by PM Rothwell using the NASCET Measurement Method, personal communication, 1999.

A major problem with ACAS was that too few patients were entered into the study to produce enough outcome events answer pertinent questions. Neither the numbers entered nor the reliance on Doppler examinations alone for the analyses allowed a distinction in benefit among any deciles of stenosis. It is counterintuitive to believe that a person without symptoms, shown have 60% narrowing by ultrasound alone, can be given a green light to have CE, while symptomatic patient with a 60% angiographic lesion is only to be considered for CE under certain circumstances.

Subgroup analyses were unable to declare benefit or lack of for disabling stroke. This could be misleading again because of the small numbers. Women were not shown to benefit. This may be due to the better prognosis for recurrent ischemic events in women compared to men, or solely due to the higher perioperative complication rate in ACAS for women (3.3% compared to 1.7% for men). The reported combined perioperative complication rate of 2.3% for men and women would be a challenge to most institutions surgical departments. The surgical complication rate is reduced to a daunting 1.1% when the inclusion of an unexpectedly high complication rate 1.2% from angiography is subtracted from the 2.3%.

Large Medicare surveys report hospital mortality alone. They do not provide data about stroke morbidity nor distinguish between CEs performed for symptomatic patients or asymptomatic individuals. The latest mortality figures from the Medicare source indicate that 1.8% of patients submitted to CE under Medicare leave the hospital dead (Robert Pokras, personal communication, 1998). Because morbidity is 3 to 4 times this figure, no comfort can be taken from the possibility of a complication rate stroke and death approaching 8%.

The databases of NASCET and ECST have been able to provide some useful information about asymptomatic arteries. Of 2377 NASCET patients from whom bilateral angiograms were available and who had never had symptoms on the other (asymptomatic) side, 509 had a stenosis of 50%, and 1330 had <50%, but a detectable amount of stenosis. The average annual risk of stroke over three-year period from these asymptomatic arteries was calculated, and a comparison by different degrees of stenosis is depicted in Figure 12.5. At all of stenosis, the annual risk is one-third, or at most one-half, that in the territory of the symptomatic artery. Only at the stenotic range of 75% 94% does risk exceed what could be improved upon by good surgeons. The risk for subjects with nearly occluded arteries was less than for the two deciles below this. This seeming paradox was observed by Norris and Zhu in their case-series of asymptomatic subjects.24 It appears that the flow beyond stenosis in the severely narrowed artery is diminished to the point where the hemisphere no longer dependent on it and emboli travel less frequently through the severely narrowed


segment. Comparable observations have been reported from the symptomatic side of NASCET.20

NASCET prospectively recorded all strokes in the asymptomatic arteries. The causes of strokes were identified as related either to the large artery lesion or to be of cardioembolic or lacunar origin. Either cardioembolic or cause was ascribed to more than 40% of these strokes, and the high percentage of strokes not of large artery origin was observed at all degrees of stenosis (Figure 12.4). A figure of 50% is probably the accurate number in the series endarterectomies not restricted by protocol to exclude threatening cardiac lesions. In any population of asymptomatic individuals, a recalculation the annual risk of stroke must be made, accounting for the fact that approximately 50% of the risk is not subject to reduction by a carotid CE. For example in the 75% 94% stenosis category, the highest risk group, only a 1.9% annual large-artery stroke would be predicted. This is not a risk that would improved upon by CE.

The comparable ECST database contained 843 randomized patients with an asymptomatic (other side) artery with 30% 69% stenosis by ECST measurement, and 127 with 70% 99% stenosis.34 During follow-up, 69 patients had a first stroke on the previously asymptomatic side lasting longer than 7 days, giving an estimated Kaplan-Meier stroke risk at three years of 2.1%. The risk for patients with 70% 99% stenosis was 5.7%. By degrees of there a sharp gradient of risk when the stenosis reached 80%. The three-year risk was 9.8% for patients with 80% 89% stenosis and 14.4% with 90% 99% stenosis. Data were not available identifying stroke by cause. There is no reason to believe that the distribution of large artery, cardioembolic, and lacunar strokes differed from the percentages found in NASCET.

Additional data have become available from the Aspirin and Carotid Endarterectomy trial (ACE), which ran parallel to NASCET in the centers. Altogether, 2804 CEs were performed, 1521 for asymptomatic subjects, and 1283 for symptomatic patients. The complication rate at 30 days was 6.4% for the symptomatic patients and 4.6% for the subjects without symptoms. Superimposing the 4.6% obtained from ACE upon Kaplan-Meier medical-surgical curves from ACAS, as Figure 12.6 demonstrates, no benefit was achieved CE out to six years. This is of particular concern because the ACE trial submitted twice as many asymptomatic subjects to CE as did the ACAS trial. Harm was done, and medical therapy should become the preferred treatment.

A fifth asymptomatic trial is ongoing with a goal of 3200 individuals to be submitted to best medical therapy or to CE.14 The larger numbers may confirm or deny the ACAS observations. This study may find a subgroup with convincing evidence of benefit. A large, long-term observational study risk factors is also being carried out that may identify the risk profile of asymptomatic individuals more likely to benefit from CE than the best contemporary medical care.35

FIGURE 12.6. The medical and surgical five-year Kaplan-Meier freedom from ipsilateral stroke and perioperative stroke and death curves indicate a 5.9% difference (11.0% minus 5.1%) favoring endarterectomy at five years in the AC AS study. A marginal but definite statistical significance by Greenwood's formula (p < 0.05) can be claimed for this difference at five years, which time only 149 patients were available for analysis. Superimposed on the curves is the outcome projected to five years for 1521 individuals receiving carotid endarterectomy for asymptomatic lesions in the ACE trial, based upon an actual 4.6% stroke and death rate at 30 days. The outcomes were then extrapolated beyond 30 days from the ACAS data. Using projected 7.4% rate of outcomes at five years, the difference between surgical and the medical group is 3.6% (11.0% minus 7.4%), only a 0.7% risk reduction per year (p [approximate, equals] 0.10). Adapted from ACAS.13 Reproduced by permission of

Barnett HIM, Meldrum HE. Carotid Endarterectomy: A Neurotherapeutic Advance. Arch Neurol 2000;57:40 45.


In two other circumstances, a practice has developed to perform CE in subjects without carotid symptoms. The first has been based on the unsubstantiated hypothesis that nonhemisphere symptoms, including symptoms clearly of vertebral basilar origin, will benefit from CE. This has never been convincingly established. The second hypothesis was that the risk of stroke in the performance of major cardiovascular and aortic surgery would be reduced by performing CE on stenosing lesions incidentally found in the carotid arteries of these patients. Later studies have indicated, first that the strokes complicating procedure occur as often in the territory of nonstenotic artery or in the posterior circulation as in the territory of the stenosed artery. Second, the cause strokes complicating coronary artery bypass grafting (CABG) is most commonly related


to the insertion of the shunt into the diseased aortic site. Finally, it has been recognized that these patients subjected to both CABG and CE have a morbidity and mortality equal to the sum of the two individual risks.36 Neither of these applications of CE can be substantiated on the basis scientifically evaluated evidence. Their use in these circumstances can be justified only if they are being done in a properly designed clinical trial.

Causes of Stroke in the Presence Carotid Stenosis

Therapeutic trials in stroke prevention have been justifiably criticized for regarding stroke as a generic phenomenon and not attempting to distinguish among stroke outcome events by cause.37,38 A striking example was the huge (19,000- patients) International Stroke Trial (1ST), which evaluated anticoagulants compared with platelet-inhibiting therapy.39 Concern was expressed that a different type of protocol, designed to distinguish among outcomes stroke by cause, may have found different results in the treatment for strokes of cardioembolic origin, lacunes, or those of large artery origin.40 It requires a leap faith to accept the contention that all of these causes of stroke will have comparable responses to identical medical therapy. To identify the cause of stroke outcomes is equally important in evaluating the benefit of CE. NASCET identified stroke events by cause. In patients with moderate disease and in strokes that occurred in the territory of the arteries asymptomatic other side, combination of lacunar and cardioembolic strokes was approximately 40% and occurred regardless of the degree stenosis (Figure 12.4).41

Lacunar strokes were not excluded as entry phenomena into NASCET. The contention was that data did not exist at the time of writing the protocol for this trial to state with assurance that patients with lacunar stroke would or would not benefit from CE. At the conclusion of the trial, the nonlacunar patients were compared with the patients who entered with lacunar syndromes. Clear benefit was seen from CE for the nonlacunar patients, but less benefit was seen in the patients who entered the trial after a lacunar stroke.42 Because number of patients who came into NASCET with probable lacunar syndromes was small, further evaluation in other trials is desirable.

Decisions About Ce Based on the Number-Needed-to-Treat Calculations

For symptomatic patients and asymptomatic subjects, it is possible from published randomized trials and from the observations made in the ACE trial to calculate the number needed to treat (NNT) by CE to prevent one additional stroke (in the case of symptomatic patients) or a first stroke (in asymptomatic subjects). These calculations, set out in Tables 12.3 and 12.4, are dependent particularly on




the degree of stenosis and the presence or absence symptoms. It was observed that only six patients in NASCET and eight patients ECST needed to be treated by CE for symptomatic disease with a 70% 99% stenosis. The number reached 19 in NASCET for patients with 50% 69% stenosis. A higher postoperative stroke rate for patients in ECST with moderate disease did not permit an NNT calculation. The numbers in the Veterans Affairs (VA) asymptomatic trial and in ACAS were 48 and 83, respectively. In the ACE study a negative impact was shown and no NNT is calculable. These calculations for asymptomatic subjects do not take into account the fact that medical risk should be cut in half to consider only likely strokes of large artery origin. When this was done, the NNT became incalculable for the VA and ACAS trials.

TABLE 12.3. Number Needed to Treat (NNT) by Endarterectomy: Symptomatic Patients

Symptomatic 70 99% North American   659 24.5   8.6 15.9   65 6 5.8
   Symptomatic Carotid Endarterectomy Trial
Symptomatic 70 99% European Carotid 501 19.9   7.0 12.9   65   8 5.6
   Surgery Trial2 (by NASCET Measurement)
Symptomatic 50 69% North American 858 14.6   9.3 5.3   36 19 6.9
   Symptomatic Carotid Endarterectomy Trial
Symptomatic 50 69% European Carotid 684 9.7 11.1 -1.4   -14 - 9.8
   Surgery Trial (by NASCET Measurement)
Symptomatic <50% North American Symptomatic 1368 11.7 10.2 1.5   13 67 6.5
   Carotid Endarterectomy Trial
Symptomatic <50% European Carotid Surgery 1822 4.3 9.5 -5.2 -109 - 6.1
   Trial (by NASCET Measurement)
1NNT = Number of patients needed to treat by endarterectomy to prevent one additional ipsilateral stroke for 2 years after the procedure, compared to medical therapy alone.
2Additional data supplied by Dr. P. Rothwell.
Reproduced by permission of Barnett HJM, Eliasziw M, Meldrum HE. Prevention of ischaemic stroke. BMJ 1999;318:1539 1543.

TABLE 12.4. Number Needed to Treat (NNT)by Endarterectomy Asymptomatic Individuals

Asymptomatic 50%, Veterans Affairs 444 7.72 5.62 2.1 27 48 4.4
   Cooperative Study, men only
Asymptomatic, Asymptomatic Carotid 1662 5.0 3.83 1.2 24 83 2.6
   Atherosclerosis Study     (actual)
Asymptomatic, Aspirin and Carotid 1521 5.04 5.8 -0.8 4.6
   Endarterectomy Trial     (assumed)
1NNT = Number of subjects needed to treat by endarterectomy to prevent one ipsilateral stroke for two years after the procedure, compared to medical therapy alone.
2Extrapolated from results. The perioperative risk is calculated on the 203 subjects who actually received surgery.
3Assignin g a perioperative risk of 2.6% based on 724 of 825 subjects who actually received endarterectomy in the surgical arm of ACAS, and utilizing a 0.6%risk of stroke in each of the two years after endarterectomy. The same 1.2% risk is assumed for the ACE patients and VA patients.
4No medical arm assumed from ACAS data.
Reproduced by permission of Barnett HJM, Eliasziw M, Meldrum HE. Prevention of ischaemic stroke. BMJ 1999;318:1539 1543.

Requirements for an Institution Performing Ce

The appropriateness of CE in certain circumstances has been established. Expert surgeons, trained and capable of performing CE, and knowledgeable about how to deal with brain ischemia, working with a team of experts in intensive care, and whose anesthetists are familiar with the best neuroanesthetic methods for a brain with compromised circulation, must function as a cohesive team. This cooperation is required to keep the 30-day perioperative morbidity and mortality at a maximum of 7% for death and any stroke (regardless site severity). When the degree of disability is evaluated at 90 days, a death and disabling stroke rate of no more than 2% must be achieved. The team must aware that the first 48 hours are more critical than all the rest of days making up 30-day postoperative period combined (Figure 12.7). The surgeons and intensive care staff must be prepared to cope with wound hematoma, wound infections, and the rare life-threatening injury to the vagus nerve that may cause acute dysphagia and aspiration.

Ideally, the team will include a stroke neurologist involved from early selection of appropriate patients to the conclusion long-term follow-up. In evaluation of stroke occurrence and in the assessment its severity, neurologists provide the most reliable data.43 Such data are necessary to obtain an accurate complication rate for the institution.

The team must be alert to the fact that the common nonstroke medical problems that arise in the postoperative period are of cardiac and respiratory origin.44 Of 1415 patients treated with CE in NASCET, postoperative fatal and nonfatal myocardial infarction combined with sudden death occurred in only 1.1%. The figure will be higher in the routine practice of CE, where cardiac conditions threatening emboli are not necessarily excluded, as they were in NASCET. Rhythm disorders in this population occurred 22 patients, of which 11 were nonvalvular atrial fibrillation. New or recurrent angina pectoris occurred in 19 patients. Respiratory complications were the next most common cause of medical


complications (N = 11) and most were pneumonia. Pulmonary embolism is a rare event as a sequel to CE. Patients are quickly mobilized. An alert for possible deep vein thrombosis must continue during the hospital period but will be expected in few post-CE patients.

FIGURE 12.7. The time of onset 92 perioperative outcome events in NASCET. Most strokes occurred during the endarterectomy and within succeeding 24 hours. Reproduced by permission of Stroke.16

Carotid Artery Imaging Methods

The symptomatic patients who entered into the randomized trials were studied by conventional angiography, and diameter stenoses were assigned. It is against these well-defined levels of stenosis that decisions must be made about the potential benefit or lack of from CE. Because no more powerful predictor of prognosis has been identified, and because the differences in are so striking for different degrees of stenosis, its accurate evaluation is of paramount importance.

Some progress has been made in efforts to determine which of the parameters of ultrasound measurements correspond to the levels of stenosis known from angiography to be within the range of benefit from CE. The correlations will never be exact, because angiography is a measurement to map the lumen, any intrusion into it, or any ulceration within the plaque, while ultrasound is a measurement of the velocity and volume blood flowing through the artery. Approximations are achievable, but at two important levels of angiographicallymeasured stenosis, difficulties in correlation between ultrasound and angiography


are still encountered: The first is in arteries narrowed by a substantial amount beyond severe stenosis and through which very little blood flows, the second is in arteries with middle ranges of stenosis in which a close correlation between what is 70% or 50% by angiography and what assessed by ultrasound would be desirable. In a patient with a severely narrowed artery and collapse of the distal segment, ultrasound may be erroneously misinterpreted because of modest velocity and volume, the stenosis characterized as moderate. Such a patient would benefit from CE but may be denied the procedure on the basis of these measurements. This mistake is made less often with the additional use of B-mode and color, but it still occurs. In the second situation, the mid-range of stenosis, overreading frequently occurs, and the degree of disease is misinterpreted as sufficiently severe to benefit from CE. This mistake leads a procedure being applied when the risk posed by the lesion is not sufficient to justify the risk of the operation.

Several reasons are advanced by some practitioners to persuade themselves and their colleagues that angiography can now be replaced by noninvasive studies. Poor angiograms are displayed that have failed to detect a severe stenosis because of failure to take oblique views the site lesion. Such poor imaging is inexcusable, and expert neuroradiologists will not tolerate it. High risk of stroke from angiography is reported.45 This too is unacceptable and denotes the need for proper training. The reports disclaiming the need for angiography make much of the complications from procedure. They fail to report fact that majority (about 80%) of patients who suffer this complicating stroke make good to excellent recovery, and only one in five patients is left with a disabling stroke.46 NASCET recorded nondisabling strokes throughout the trial from 2885 angiograms. The nondisabling stroke rate was 0.6%, from which may be extrapolated a risk of disabling stroke of 0.1%. This low risk is a small price to ask patients to pay so that none without an appropriate lesion be submitted to the 20-fold increase of risk from disabling stroke (2%) that confronts patients who undergo CE, even in the best centers.

Cost has been an argument supporting ultrasound as a replacement for conventional angiography. In some centers the cost attached to the performance of angiography and film interpretation is as much as the cost of CE. Moderate savings are not a reason to settle for the second-best in imaging methods. The economic cost of permanent disability resulting from stroke after CE exceed the savings by a wide margin.

Other methods may make conventional angiography obsolete in the future. Furthur improvements may occur in ultrasonography, so that the over-reading in the mid-ranges of stenosis is eliminated and the under-reading in the highest range is overcome. Angiography can then be replaced. To date, this remains a hope. Magnetic resonance imaging over-reads stenosis. When and if the problem of signal void is overcome, another possible replacement for angiography


will have emerged. Meanwhile, ultrasound and magnetic resonance imaging are excellent screening tools, but conventional angiography remains desirable as a prelude to CE.


Symptomatic Patients

  • CE prevents stroke in the territory of arteries with symptoms related to stenoses of 70% 99%, validated angiographically by the NASCET method of measurement.

  • Surgeons who perform the procedure with 7% perioperative complication of stroke and death will achieve benefit in the presence of severely stenosed symptomatic lesions.

  • Patients with <70% stenosis and appropriate symptoms cannot be assigned automatically to surgical therapy. Careful consideration must be given all risk factors, as well to the patients identified likely to benefit the most.

  • Conventional angiography must be seriously considered as a prelude to CE for most patients.

Asymptomatic Subjects

  • Individuals with asymptomatic stenosis have not been shown convincingly to benefit from CE.

  • The risk with medical therapy in asymptomatic individuals is less than onethird that for symptomatic patients with severe stenosis and one-half that symptomatic patients with moderate stenosis.

  • Approximately half the stroke risk facing asymptomatic individual is for lacunar or cardioembolic stroke, neither of which are responsive to CE.

  • The consensus statement provided by the American Heart Association interprets the data supporting CE for asymptomatic subjects in an optimistic fashion. This statement recommends that all patients with >60% asymptomatic stenosis be considered for CE, provided the surgical risk is <3%.47

  • The authors regard CE for asymptomatic lesions as an overutilized therapeutic strategy. Future investigations of the highest risk subgroups may identify some patients who will benefit convincingly from CE.


1. Moniz E, Lima A, de l'Acerda R. Hemiplegies Par Thrombose de la Carotide Intere. Presse Med 1937; 1:977 980.

2. Hultquist GTJ. Uber Thrombose und Embolie der Arteria Carotis. Stockholm: Norstedt, 1942.


3. Fisher CM. Occlusion of the internal carotid artery. Arch Neural Psychiatr Chicago 1951;65:346 377.

4. Eastcott HHG, Pickering GW, Rob CG. Reconstruction of internal carotid artery in a patient with intermittent attacks of hemiplegia. Lancet 1954;2:994 996.

5. Fields WS, Maslenikov V, Meyer JS, Hass WK, Remington RD, Macdonald M. Joint study of extracranial occlusion. V: Progress report prognosis following surgery or nonsurgical treatment for transient cerebral ischemic attacks and cervical carotid artery lesions. JAMA 1970;211:1993 2003.

6. Barnett HJM, Eliasziw M, Meldrum HE, Taylor DW. Do the facts and figures warrant a 10-fold increase in the performance of carotid endarterectomy on asymptomatic patients? Neurology 1996;46:603 608.

7. European Carotid Surgery Trialists' Group. Randomised trial of endarterectomy for recently symptomatic carotid stenosis: Final results of the MRC European Carotid Surgery Trial (ECST). Lancet 1998;351:1379 1387.

8. Barnett HJM, Taylor DW, Eliasziw M. et al., for the North American Symptomatic Carotid Endarterectomy Trial Collaborators. Benefit of carotid endarterectomy in symptomatic patients with moderate and severe stenosis. N Engl J Med 1998:339: 1415 1425.

9. Mayberg MR, Wilson SE, Yatsu F, et al. Carotid endarterectomy and prevention of cerebral ischemia in symptomatic carotid stenosis. JAMA 1991;266:3289 3294.

10. The CASANOVA Study Group. Carotid surgery versus medical therapy in asymptomatic carotid stenosis. Stroke 1991;22:1229 1235.

11. Mayo Asymptomatic Carotid Endarterectomy Study Group. Results of a randomized controlled trial of carotid endarterectomy for asymptomatic carotid stenosis. Mayo Clin Proc 1992;67:513 518.

12. Hobson RW II, Weiss DG, Fields WS, Goldstone J, Moore Towne JB, Wright CB and the Veterans Affairs Cooperative Study Group: Efficacy of carotid endarterectomy for asymptomatic carotid stenosis. N Engl J Med 1993,328:221 227.

13. Executive Committee for the Asymptomatic Carotid Atherosclerosis Study: Endarterectomy for asymptomatic carotid artery stenosis. JAMA 1995;273:1421 1428.

14. Halliday AW, Thomas D, Mansfield A. The asymptomatic carotid surgery trial (ACST) rationale and design. Eur J Vase Surg 1994;8:703 710.

15. Taylor DW, Barnett HJM, Haynes RB, Ferguson GG, Sackett DL, Thorpe KE, Simard D, Silver FL, Hachinski V, Clagett GP, Barnes RW, Spence JD, for the ASA and Carotid Endarterectomy (ACE) Trial Collaborators. Low-dose and high-dose acetylsalicylic acid for patients undergoing carotid endarterectomy: A randomised controlled trial. Lancet 1999;353:2179 2184.

16. Ferguson GG, Eliasziw M, Barr HWK, Clagett GP, Barnes RW, Wallace MC, Taylor DW, Haynes RB, Finan JW, Hachinski VC, Barnett HJM, for the North American Symptomatic Carotid Endarterectomy Trial (NASCET) Collaborators. The North American Symptomatic Carotid Endarterectomy Trial: Surgical results in 1415 patients. Stroke 1999 30: 1751 1758.

17. Henderson R, Eliasziw M, Fox AJ, Rothwell P, Barnett HJM for the North American Symptomatic Carotid Endarterectomy Trial (NASCET) Group. AngiographicallyDefined Collateral Circulation and the Risk of Stroke in Patients with Severe Carotid Artery Stenosis. Stroke 2000;31:128 132.

18. Kucey DS, Bowyer B, Iron K, Austin P, Anderson G, Tu JV, for the University of Toronto Carotid Study Group. Determinants of outcome after carotid endarterectomy. J Vase Surg 1998;28:1051 1058.


19. Gasecki AP, Eliasziw M, Ferguson GG, Hachinski V, Barnett HJM, for the North American Symptomatic Carotid Endarterectomy Trial (NASCET) Group. Long-term prognosis and effect of endarterectomy in patients with symptomatic severe carotid stenosis and contralateral carotid stenosis or occlusion: Results from NASCET. JNeurosurg 1995;83:778 782.

20. Morgenstern LB, Fox AJ, Sharpe BL, Eliasziw M, Barnett HJM, Grotta JC, for the North American Symptomatic Carotid Endarterectomy Trial (NASCET) Group. The risks and benefits of carotid endarterectomy in patients with near occlusion the carotid artery. Neurology 1997;48:911 915.

21. Bornstein NM, Norris JW. Management of patients with asymptomatic neck bruits and carotid stenosis. In: Barnett HJM, Hachinski VC, eds. Neurologic Clinics Cerebral Ischemia: Treatment and Prevention. Philadelphia: W.B. Saunders Company 1992:269 280.

22. Chambers BR, Norris JW. Outcome in patients with asymptomatic neck bruits. N Engl J Med 1986;315:860 865.

23. Hennerici M, Hulsbomer HB, Hefter H, Lammerts D, Rautenberg W. Natural history of asymptomatic extracranial arterial disease. Brain 1987; 10:777 791.

24. Norris JW, Zhu CZ. Stroke risk and critical carotid stenosis. Neural Neurosurg Psychiatry 1990;53:235 237.

25. Cheung RTF, Chan RKT, Eliasziw M, Rankin RN, Hachinski VC, Barnett HJM, for the NASCET Group. Recurrent stenosis after carotid endarterectomy: How often should follow-up ultrasound be performed? Experience from the North American Symptomatic Carotid Endarterectomy Trial (NASCET). Neurology 1996;46(2):A281.

26. Cote R, Battista RN, Abrahamowicz M, Langlois Y, Bourque F, Mackey A. Lack of effect of aspirin in asymptomatic patients with carotid bruits and substantial carotid narrowing: The Asymptomatic Cervical Bruit Study Group. Ann Intern Med 1995; 123(9):649 655.

27. Chambers BR, Norris JW. The case against surgery for asymptomatic carotid stenosis. Stroke 1984; 15:964 967.

28. Foster DS. Endarterectomy for asymptomatic carotid artery stenosis (letter to the editor). JAMA 1995;274:1505.

29. Goldstein MR. Endarterectomy for asymptomatic carotid artery stenosis, (letter to the editor). JAMA 1995;274:1505 1506.

30. Oddone E. Endarterectomy for asymptomatic carotid artery stenosis (letter to the editor). JAMA 1995;274:1506.

31. Seiden SW. Endarterectomy for asymptomatic carotid artery stenosis (letter to the Editor). JAMA 1995;274:1506.

32. Gorelick PB, Sacco RL, Smith DB, Alberts M, Mustone-Alexander L, Rader D, Ross JL, Raps E, Ozer MN, Brass LM, Malone ME, Goldberg S, Booss J, Hanley DF, Toole JF, Greengold NL, Rhew DC. Prevention of a first stroke: A review guidelines and a multidisciplinary consensus statement from the National Stroke Association. JAMA 1999;281:1112 1120.

33. Warlow C. Endarterectomy for asymptomatic carotid stenosis? Lancet 1995;345: 1254 1255.

34. The European Carotid Surgery Trialists Collaborative Group. Risk of stroke in the distribution of an asymptomatic carotid artery. Lancet 1995;345:209 212.

35. Nicolaides AN. Asymptomatic carotid stenosis and risk of stroke. Identification of a high risk group (ACSRS). A natural history study. Int Angiol 1995;14:21 23.


36. Hertzer NR, O'Hara PJ, Mascha EJ, Krajewski LP, Sullivan TM, Beven EG. Early outcome assessment for 2228 consecutive carotid endarterectomy procedures: The Cleveland Clinic experience from 1989 to 1995. J Vase Surg 1997;26:1 10.

37. Caplan LR. Diagnosis and treatment of ischemic stroke. JAMA 1991;266:2413 2418.

38. Caplan LR. TIAs: We need to return the question What is wrong with Mr. Jones? Neurology 1988;38:791 793.

39. International Stroke Trial Collaborative Group. The International Stroke Trial (1ST): A randomised trial of aspirin, subcutaneous heparin, both, or neither among 19,435 patients with acute ischaemic stroke. Lancet 1997;349:1569 1581.

40. Bousser MG. Aspirin or heparin immediately after a stroke? Lancet 1997;349:1564 1565.

41. Barnett HIM, Gunton RW, Eliasziw M, Fleming L, Sharpe B, Meldrum H, for the North American Symptomatic Carotid Endarterectomy Trial (NASCET) Group. The causes and severity of ischemic stroke in patients with internal carotid artery stenosis. JAMA 2000;283:1429 1436.

42. Inzitari D, Eliasziw M, Sharpe BL, Fox AJ, Barnett HIM, for the North American Symptomatic Carotid Endarterectomy Trial (NASCET) Group. Risk Factors and Outcome of Patients with Carotid Artery Stenosis Presenting Lacunar Stroke. Neurology 2000;54:660 666.

43. Rothwell PM, Slattery J, Warlow CP. A systematic review of the risks of stroke and death due to endarterectomy for symptomatic carotid stenosis. Stroke 1996;27:260 265.

44. Paciaroni M, Eliasziw Kappelle LJ, Finan JW, Ferguson GG, Barnett HJM, for the North American Symptomatic Carotid Endarterectomy Trial (NASCET) Collaborators. Medical complications associated with carotid endarterectomy. Stroke 1999; 30:1759 1763.

45. Davies KN, Humphrey PR. Complications of cerebral angiography in patients with symptomatic carotid territory ischaemia screened by carotid ultrasound. JNeurol Neurosurg Psychiatry 1993;56:967 972.

46. Hankey GJ, Warlow CP, Sellar RJ. Cerebral angiographic risk in mild cerebrovascular disease. Stroke 1990;21:209 222.

47. Biller J., Feinberg WM, Castaldo JE, Whittemore AD, Harbaugh RE, Dempsey RJ, Caplan LR, Kresowik TF, Matchar DB, Toole JF, Easton JD, Adams HP, Brass LM, Hobson RW, Brott TG, Sternau L. Guidelines for carotid endarterectomy: A statement for healthcare professionals from a special writing group of the Stroke Council, American Heart Association. Stroke 1998;29:554 562.

48. North American Symptomatic Carotid Endarterectomy Trial Collaborators. Beneficial effect of carotid endarterectomy in symptomatic patients with high-grade stenosis. N Engl J Med 1991;325:445 453.

Stroke Prevention
A Primer on Stroke Prevention and Treatment: An overview based on AHA/ASA Guidelines
ISBN: 1405186518
EAN: 2147483647
Year: 2001
Pages: 23

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net