Online newspaper of professor Yasser Metwally

The author: Professor Yasser Metwally

http://yassermetwally.com

INTRODUCTION

Carotid endarterectomy appeared and rapidly expanded as a treatment for patients with carotid artery disease after Eastcott reported the successful excision of a diseased segment of the carotid artery in 1954 (Eastcott et al, 1954). Patients with and without symptoms related to arteriosclerotic and other diseases of the carotid artery then became the focus of surgical attention. Attempts were later made to evaluate the benefits of carotid endarterectomy in two multicenter trials, one conducted in the United States, the other in Britain (Fields et al, 1970; Shaw et al, 1984). The first was unsuccessful because of the trial design in that 42% of the patients had vertebral-basilar rather than carotid symptoms. The initial published analysis overlooked the complications which occurred during the period of hospitalization. This gave a false impression of benefit. When the perioperative morbidity and mortality was restored to the analysis there was no apparent benefit for endarterectomy. The second trial entered only 41 patients and was stopped because of an excessive perioperative complication rate.

It was felt by many observers that these trials inaccurately portrayed the capability of endarterectomy to prevent stroke in properly selected individuals operated on by skilled surgeons. The procedure continued to grow in its usage and by 1985 was carried out 107 000 times in the non-Veterans hospitals of the United States. It was estimated that up until that time one million endarterectomies had been performed worldwide. Since then its popularity declined and recently has moved upward. Surveys were conducted which indicated that there were some individual series in which a very acceptable morbidity and mortality was obtained in the perioperative period but that the average in the United States was above 9% (Fode et al, 1986; Winslow et al, 1988). These surveys indicated that approximately 40% of the procedures were carried out in asymptomatic patients and about 60% in those with appropriate symptoms. The extent of the disease and the degree of narrowing was incompletely specified or ignored in a large number of the reports. Some of the reports contained a scattering of patients who did not have arteriosclerotic disease but had kinks, coils and fibromuscular dysplasia. Despite the growing popularity of the procedure there was confusion as to which patients would benefit and for whom investigation was appropriate as a prelude to endarterectomy (Barnett et al, 1984; Warlow, 1984). The therapy exposes patients to the particular risk (stroke) which endarterectomy is intended to prevent. Under these conditions of uncertainty, the launching of randomized clinical trials was appropriate.

ASYMPTOMATIC CAROTID ARTERY STENOSIS

Moniz et al (1937) by introducing cerebral angiography and Fisher (1951) by careful pathological studies, revealed that the extracranial portion of the internal carotid artery was a common cause of stroke. The earlier concept was inaccurate. The most common arteriosclerotic lesion producing cerebral ischaemia was not in the middle cerebral artery. The middle cerebral artery is recognized now as the usual site for emboli, not for thrombosis-in-situ. Following Moniz’ and Fisher’s observations, transient ischaemic attacks of the brain or retina were recognized as the harbingers of stroke. The spread of this knowledge, the recognition of the significance of bruits at the carotid bifurcation, improved surgical skill in the hands of an increasing number of surgeons dealing with symptomatic carotid arteries, and the introduction of non-invasive methods of imaging the cervical arteries set the stage for an increased interest in performing carotid endarterectomy in symptomatic as well as asymptomatic individuals. It was performed and used as a prophylaxis against stroke or carotid occlusion.

The greatest enigma facing all who would prevent stroke is the management of these symptomless patients. The solution has been sought by population studies and by randomized trials testing endarterectomy against medical care. Long-term observational studies have been carried out in which the carotid lesion in asymptomatic patients has been followed to determine the risk of cerebral ischaemia (Autret et al, 1987; Hennerici et al, 1987; O’Holleran et al, 1987; Norris et al, 1991; Shanik et al, 1992). The best known have been those of Norris et al and of Hennerici et al. Norris et al followed 696 individuals with Doppler studies for 8 years and reported that when the stenosis reached 75% the risk of stroke increased from less than 1% per year to 2.5%. Hennerici et al followed 500 asymptomatic individuals with carotid stenosis identified by Doppler. At the critical point of 80% stenosis the annual risk of stroke rose from I to 2.4%. These observational studies have indicated that if there is to be a benefit from endarterectomy, the procedure must be done with exceptional skill. Otherwise the probability of improving stroke-free survival is remote, with the possible exception of patients with the most severely stenosing lesions. When the degree of stenosis is in the higher deciles the observational studies have found a high annual rate of fatal myocardial infarction (6.5% in the Norris study) in addition to an increased stroke rate.

Four randomized trials evaluating the possible benefit of endarterectomy in asymptomatic patients have now been reported (CASANOVA Study Group, 1991; MACE Study Group, 1992; Hobson et al, 1993; Executive Committee for the Asymptomatic Carotid Atherosclerosis Study, 1995). The first of these, the Carotid Artery Stenosis with Asymptomatic Narrowing: Operation versus Aspirin (CASANOVA), was very clearly a negative study. The surgical group did not fare as well as the medical group. Unfortunately the patients with 90% stenosis were excluded from the trial. Because many believe these patients with the highest grade of stenosis are those who are likely to demonstrate a benefit, their exclusion prevented the possibility of obtaining this answer. The surgical morbidity was higher than desired at 6.9% and one-half of the patients assigned to medical care eventually had some surgical intervention. The Mayo Clinic carried out an asymptomatic carotid endarterectomy trial. It was discontinued after 71 patients had been enrolled. The reason for discontinuing the study related to the fact that eight instances of myocardial infarction and three of cerebral ischaemia occurred in the group assigned to endarterectomy while none of these outcomes occurred in the aspirin group. The design of the study discouraged patients in the surgical group from taking aspirin.

The third trial was conducted in the department of Veterans Affairs and it enrolled 444 patients. Of the 203 patients that underwent endarterectomy, the perioperative stroke rate was 4.4%. Counting perioperative deaths and stroke, the stroke-free survival curves for the two treatment groups were identical, not only overlapping during the early weeks of follow-up but also throughout the 4 year mean follow-up. This publication reported a reduction of ischaemic episodes if one counted transient events in the surgical group compared with the medical group. To rid patients of the harbinger of stroke without improving stroke-free survival is not considered by most investigators to be a justification for the risk of endarterectomy.

The fourth trial, most recently reported, is the Asymptomatic Carotid Atherosclerosis Study (ACAS). The results stated a benefit from endarterectomy as compared with medical treatment in 1662 patients followed for an average of 2.7 years. They reported a relative risk reduction of 53% for stroke in patients randomized to surgery compared with those randomized to medicine. The absolute risk reduction for disabling stroke was small. No difference in the surgical benefit between the higher and lower degrees of stenosis was established.

None of these trials give clear answers as to which patients, if any, will have fewer strokes as a result of endarterectomy carried out prior to the onset of symptoms. There is a fifth underway in Europe which proposes to randomize as many patients as in the other four trials combined (Halliday AW for the Steering Committee, 1994).

CAROTID ENDARTERECTOMY TRIALS IN SYMPTOMATIC PATIENTS

Three trials have been initiated in symptomatic patients. The first of these, The European Carotid Surgery Trial (ECST), was launched in 1981 (ECST Collaborative Group, 1991); the second, the North American Symptomatic Carotid Endarterectomy Trial (NASCET), randomized its first patients in December 1987 (NASCET Collaborators, 1991); and the third trial, The Veterans Administrations Trial, was launched in July 1988 (Mayberg et al, I 99 1).

• North American Symptomatic Carotid Endarterectomy Trial (NASCET)

This trial, in its first phase, required that patients be admitted within 4 months of a cerebral or retinal ischaemic event. To be eligible, the symptoms could be transient, persistent but not disabling, and within the territory of an arteriographically-proven carotid artery lesion appropriate to the symptoms. Patients were not eligible who had overt cardiac disease which could cause cerebral emboli, nor if they were over 80 years of age, nor if they had any evidence of organ failure or illness likely to deny them 5 years of further life. After obtaining informed consent they were randomized by a computerized method to either best medical care or best medical care plus endarterectomy. The best medical care consisted of a recommendation for aspirin therapy at a dose not exceeding 1300 mg nor above a dose that the patient could tolerate. Other anti-thrombotics were acceptable if aspirin was not tolerated or ischaemic events occurred despite its use. The management of hypertension, diabetes, and blood cholesterol was strongly recommended but the specifics of such therapy were left to the discretion of the attending physician. Cigarette smoking was discouraged. Ultrasound was used as a screening test to identify who might be appropriate for carotid arteriography and ultimately found to be eligible for the trial. Ultrasound was requested also at the 30 day follow-up in both the medical and the surgical patients and was recommended after an outcome event. The protocol required that conventional arteriography be performed in each patient with bilateral selective images, including the intracranial circulation. This ensured against intracranial disease of greater significance than the putative extracranial cause and guarded against the randomization of patients with unruptured but vulnerable cerebral aneurysms. The reports and hard copies of the ultrasound examinations and arteriograms were submitted to the Central Office for evaluation and, in the case of the arteriograms, for final inclusion in the database.

The arteriograms were measured using, as the numerator, the narrowest segment of the artery in two or three planes and using, as the denominator, the distal part of the internal carotid artery well beyond the disease, the siphon, and any post-stenotic dilatation . Intra- and inter-observer reliability studies were carried out at regular intervals to ensure the consistency of the method of measurement of the arteriographic stenotic diameters.

Fifty centres began to randomize patients in December 1987. They were selected because the neurologists and surgeons were willing to submit appropriate patients for randomization and because the surgical audit of perioperative morbidity and mortality was at or below 6% in 50 successive patients. Stopping Rules were introduced so that after I year the outcome events of stroke or perioperative death were analyzed secretly and thereafter on a regular basis by one of the study biostatisticians. If a benefit or harm from surgical or medical therapy was identified and this benefit persisted over a 6 month period, the study was to be stopped.

Patients were stratified into those at or above 70% stenosis and those below 70% stenosis. When 659 patients had been randomly assigned to best medical care alone (N = 3 3 1) or best medical care plus endarterectomy (N = 328) the study was stopped in February 1991 for patients with 70% stenosis or greater. The results are quoted as they were reported:

Life-table estimates of the cumulative risk of any ipsilateral stroke at two years were 26 percent in the 331 medical patients and 9 percent in the 328 surgical patients an absolute risk reduction (+/-SE) of 17 +/-3.5 percent (P <0.001). For a major or fatal ipsilateral stroke, the corresponding estimates were 13.1 percent and 2.5 percent an absolute risk reduction of 10.6 to 2.6 percent (P < 0.001). Carotid endarterectomy was still found to be beneficial when all strokes and deaths were included in the analysis (P < 0.001) (NAS CET Collaborators, 1991).

After February 1991 the patients who were in the severe group but assigned to medical therapy have been encouraged to have carotid endarterectomy. More than half have had carotid endarterectomy and those who have not have followed a nonsurgical course for a variety of reasons including occlusion, no recent events, death, poor medical condition, patient refusal, and major stroke. The patients with 70% stenosis randomized to receive endarterectomy during the first phase of the trial have all been followed and the operation has proved to be durable (Barnett et al, 1995)

Sub-group analyses have been done from the NASCET database and the following observations have been made:

1. A risk profile of 15 vascular and carotid artery risk factors were identified (NASCET Collaborators, 1991). The patients in the medical arm with a low risk profile (less than six risk factors) survived free of stroke substantially better than did those with a median or a high risk profile.

2. Patients with retinal ischaemic events as the presenting symptom had a much more favourable outlook than those presenting with hemispheric events (Streifler et al, 1995) . With a 17% risk of stroke at 2 years, patients with amaurosis fugax should still be considered candidates for endarterectomy if the symptom is appropriate to the severe carotid lesion.

3. Computerized tomographic (CT) images of one-third of the patients in the severe group presenting only with transient events had lesions suggesting ischaemic infarctions in the appropriate hemisphere. From initial analysis of the medical group it appeared that these patients were at greater risk for subsequent stroke because of this lesion than those without these lesions (Streifler et al, 1992a). Multivariate analysis determined that this was not accurate. As a single risk factor, a CT lesion not accompanied by persisting clinical symptoms or signs is merely a marker for vascular disease (Eliasziw et al, 1995b).

4. Ulceration, when it was unequivocally present, carried an increased risk of stroke and this risk increased progressively as the degree of stenosis increased . Patients without obvious ulceration whether in the eighth, ninth or tenth deciles of stenosis had a prognosis that was better than those with ulceration at any degree of stenosis at or above 70% (Eliasziw et al for NASCET, 1994).

5. Thrombi located in the lumen of the artery carried a poor prognosis that was not mitigated by endarterectomy. This was unique because the other risk factors did not increase the risk of endarterectomy, nor did they influence the good outlook observed during the follow-up of these patients. Several other series are on record where intraluminal thrombi added to the seriousness of the prognosis and which affirm the fact that there is a high post-operative risk of stroke in such patients (Biller et al, 1986; Pelz et al, 1986; McCrory et al, 1993; Gertier et al, 1994). Recommendations have been formulated from anecdotal data that when a thrombus is identified within the lumen of the artery the patient should be given 30 days of anticoagulation therapy prior to endarterectomy (Buchan et al, 1988). At the conclusion of 30 days an ultrasound is recommended to ensure that the artery remains patent. As judged by the few patients in the case-series, the endarterectomy may then be performed with a reduced risk.

6. Other observations have emerged from the database of NASCET including the fact that age alone was not a deterrent to endarterectomy (Streifler et al for the NASCET Group, 1992b); women fared as well as men in the perioperative period (NASCET Collaborators, 1991); women in the medical an-n of the trial had an outlook comparable to men in long-term follow-up (Sharpe et al, 1995); and there was no correlation between the existence of severe stenosis and the location and extent of leukoaraiosis (Streifler et al for NASCET, 1995a).

• European Carotid Surgery Trial (ECST)

This trial was launched in 1981. It varied in a few details from NASCET. patients were randomized to surgical or medical therapy in a ratio of 2: 1; a longer period since the last event (I 80 days) was allowed as a prelude to randomization; centres were allowed to decide at the onset of their collaboration with the trial not to submit to randomization those patients whom they were comfortable sending for endarterectomy; patients judged by experience to be unlikely to benefit from the procedure because of the modest degree of stenosis were left out of the study. Arteriograms were required and the degree of stenosis was measured by utilizing the imagined position of the lumen of the diseased carotid bulb as the denominator in the equation.

In February 1991 there was notification to the investigators of this trial that a significant benefit was detected in the patients with greater than 70% stenosis by the ECST method of measurement and for these patients the trial was halted. The benefit in the ECST for endarterectomy was less than in NASCET even though their perioperative morbidity and mortality (counting any minor event persisting more than 7 days as a stroke) was slightly greater than in NASCET. Patients in the ECST medical group had fewer strokes and deaths than in the NASCET. These differences were explained by the different methods of measurement. More patients had been allowed into the ECST trial in the severe group than were recognized as severe by NASCET. When the analysis was repeated to include only the patients in ECST who corresponded to 70% or greater by NASCET measurement, the Kaplan-Meier survival curves for the two studies became virtually identical (Barnett and Warlow, 1993).

• Veterans Affairs (VA) Cooperative Studies Program 309

The third study was launched in the US Veterans institutions in July 1988. Patients were randomized with arteriographic evidence of at least 50% stenosis measured by linear diameter and using the same measurement formula as in NASCET. A small number (N= 189) of patients had been randomized by the time ECST and NASCET declared benefit for patients with severe disease. This Veterans study was terminated. A trend toward benefit in stroke reduction was beginning to emerge.

• Non-hemispheric events

No randomized trials have included patients with non-specific, so-called ,non-hemispheric symptoms. There are no Convincing data that either these poorly-defined symptoms, nor symptoms that are clearly identifiable as vertebral-basilar in origin, will benefit from carotid endarterectomy.

SUMMARY OF CURRENT KNOWLEDGE ABOUT CAROTID ENDARTERECTOMY

Current knowledge about the application of endarterectomy may be summarized as follows.

1. Endarterectomy is the best treatment for patients who have symptoms clearly causing focal, retinal or hemispheric symptoms related to a 70% or greater carotid stenosis which is accessible to surgery.

2. The perioperative morbidity and mortality for endarterectomy must be as low as possible. Benefit is negated if perioperative complications reach a level of 10% or higher. Ideally the perioperative stroke or death rate should be as much below 6% as skilled individuals can achieve. Hospitals and surgical departments are obliged to conduct independent audits to ensure this level of competence is maintained.

3. The trials for symptomatic disease have evaluated the degree of stenosis by conventional arteriography. Because it is difficult to correlate arteriography and ultrasound, the use of ultrasound alone in determining who should receive endarterectomy remains uncertain. Ultrasound misses some of those with very tight stenoses and too often mismeasures in the moderate range (Eliasziw et al, 1995a; Blakely et al, 1995). The intracranial vasculature cannot be evaluated sufficiently by ultrasound. In the future improved ultrasound and better resolution of magnetic resonance angiography will obviate the need for carotid arteriography. This level of expertise has not been attained yet by single or combined non-invasive methods.

4. Trials of symptomatic patients with stenosis below 70% continue to evaluate the benefit of endarterectomy. The two large ongoing trials will provide the necessary data over the next 2-3 years. The ECST will continue to follow its 1800 patients and NASCET to randomize until it has 2100 patients with an average follow-up of 5 years. It may prove that patients with symptomatic disease below 70% will benefit from endarterectomy only in conjunction with a higher vascular risk profile. The investigators in NASCET plan to determine if this demarcation line will be altered by variations in the risk profile.

5. The European Carotid Surgery Trial has published observations about the risk of stroke in patients with asymptomatic lesions on the other side. These lesions have been identified by arteriography in patients randomized because of appropriate symptoms in the ipsilateral artery (The ECST Collaborative Group, 1995). In this population of asymptomatic carotid lesions, the 3 year risk of stroke in patients with 70-99% stenosis was 5.7%, less than 2% risk per year, and not at the level of statistical significance. The risk of stroke at 3 years was 9.8% in those with 80-89% stenosis and-was 14.4% in those with 90-99% stenosis. Again this was not statistically significant because of the paucity of outcome events in a small number of patients. Similar observations as yet unpublished have been observed in the asymptomatic arteries on the other side in the NASCET database.

6. Acceptance of the importance of the results of The Asymptomatic Carotid Atherosclerosis Study (ACAS) must be tempered with caution. The 5-year risk for the surgical patients was 5.1 % and for the medical patients was 11.0%. Although this is a 53% relative risk reduction at 5 years, the absolute risk reduction is only 5.9%, or less than 1.5% risk reduction per year. This is hard to improve upon with anything other than a nil risk of surgery. Even in the best of hands, the procedure carries with it a 1% risk of death and an equal or greater risk of stroke. The balance between the risk of the disease and of this treatment strategy is too close to be confident that benefit will be achieved.

7. ACAS states that disabling stroke was not reduced by endarterectomy.

8. ACAS was not able to distinguish a differential benefit in the higher (above 80%) compared with the lower (below 80%) levels of stenosis. This was apparently related to the restricted use of carotid arteriography as the definitive imaging test. A total of, 642 patients had carotid angiography and were randomized equally to medicine and surgery. There were only 27 outcome events in this group. Five of these occurred in patients with severe (80-99%) stenosis, three in the medical and two in the surgical arm of the trial. Evidence from NASCET, ECST, the observational case-studies, and the contralateral lesions in ECST and NASCET have strongly suggested that any benefit from endarterectomy would be expected in those at highest risk which were those with the greatest stenosis. Without more patients shown by arteriography to have severe disease and therefore with so few outcome events ACAS could not confirm the importance of severe stenosis.

9. Further studies may determine that there is a group of asymptomatic patients with a severe stenosis and a high vascular risk profile who will benefit from endarterectomy.

10. No data have been accumulated which allows decisions to be made about the performance of carotid endarterectomy in the presence of severe stenosis found in the investigation of a patient requiring a coronary by-pass procedure.

SUMMARY

Several randomized trials have recently reported on the benefit of carotid endarterectomy for symptomatic or asymptomatic patients. A positive benefit has been found for symptomatic patients with 70% stenosis measured by angiogram with a particular formula and performed with high surgical skill. Symptomatic patients with <70% stenosis continue to be randomized and followed. The answer for this group is expected in the next two years. The randomized trials of asymptomatic patients have not established the benefit for carotid endarterectomy clearly. The most recent trial shows the absolute risk reduction at 5 years is only 5.9% or less than 1.5% risk reduction per year. A fifth trial continues in Europe.

---

REFERENCES

Autret A, Pourcelot L, Saudeau D et al (1987) Stroke risk in patients with carotid stenosis. Lancet I: 888-890.

Barnett HJM & Warlow CP (1993) Carotid endarterectomy and the measurement of stenosis. Stroke 24: 1281-1284.

Barnett HJM, Plum F & Walton JN (1984) Carotid endarterectomy: an expression of concern. Stroke 15: 941-943.

Barnett HJM, Meldrum HE & Eliasziw M (1994) The North American Symptomatic Carotid Endarterectomy Trial: further observations. In Veith FJ (ed.) Current Critical Problems in Vascular Surgery, pp 290-294. St Louis, MO: Quality Medical Publishing.

Barnett HJM, Meldrum HE & Eliasziw M (1995) Drugs and surgery in the prevention of ischemic stroke. New England Journal of Medicine 332: 238-248.

Biller J, Adams HP Jr, Boarini D et al (1986) Intraluminal clot of the carotid artery. A clinical-angiographic correlation of nine patients and literature review. Surgical Neurology 25: 467-477.

Blakely DD, Oddone EZ, Hasselblad V et al (1995) Noninvasive carotid artery testing. Annals of Internal Medicine 122: 360-367.

Buchan A, Gates P, Pelz D & Barnett HJM (1988) Intraluminal thrombus in the cerebral circulation. Stroke 19: 681-687.

CASANOVA Study Group (1991) Carotid surgery versus medical therapy in asymptomatic carotid stenosis. Stroke 22: 1229-1235.

Eastcott HHG, Pickering GW & Rob CG (1954) Reconstruction of internal carotid artery in a patient with intermittent attacks of hemiplegia. Lancet II: 994-996.

Eliasziw M, Streifler JY, Fox AJ et al for the North American Symptomatic Carotid Endarterectomy Trial (1994) Significance of plaque ulceration in symptomatic patients with high-grade carotid stenosis. Stroke 25: 304-308.

Eliasziw M, Rankin RN, Fox AJ et al (1995a) The accuracy and prognostic consequences of ultrasound in identifying a severe carotid artery stenosis. Stroke (in press).

Eliasziw M, Streifler JY, Spence JD et al (1995b) Prognosis for patients following a transient ischemic attack with and without a cerebral infarction on a brain CT scan. Neurology 45: 428-431.

European Carotid Surgery Trialists Collaborative Group (1991) MRC European Carotid Surgery Trial: interim results for symptomatic patients with severe (70-99%) or with mild (0-29%) carotid stenosis. Lancet 337: 1235-1243.

European Carotid Surgery Trialists Collaborative Group (1995) Risk of stroke in the distribution of an asymptomatic carotid artery. Lancet 345: 209-212.

Executive Committee for the Asymptomatic Carotid Atherosclerosis Study (1995) Endarterectomy for asymptomatic carotid artery stenosis. Journal of the American Medical Association 273: 1421-1428.

Fields WS, Maslenikov V, Meyer JS et al (1970) Joint study of extracranial arterial occlusion. V. Progress report of prognosis following surgery or nonsurgical treatment for transient cerebral ischemic attacks and cervical carotid artery lesions. Journal of the American Medical Association 211: 1993-2003.

Fisher CM (195 1) Occlusion of the internal carotid artery. Archives of Neurology and Psychiatry 65: 346-377.

Fode NC, Sundt TM Jr, Robertson JT et al (1986) Multicenter retrospective review of results and complications of carotid endarterectomy in 1981. Stroke 17: 370-376.

Gertler JP, Blankensteijn JD, Brewster DC et al (1994) Carotid endarterectomy for unstable and compelling neurologic conditions: Do results justify an aggressive approach? Journal of Vascular Surgery 19: 32-42.

Halliday AW for the Steering Committee and for the Collaborators (1994) The Asymptomatic Carotid Surgery Trial (ACST) rationale and design. European Journal of Vascular Surgery 8: 703-710.

Haynes RB, Taylor DW, Sackett DL et al (1994) Prevention of functional impairment by endarterectomy for symptomatic high-grade carotid stenosis. Journal of the American Medical Association 271:1256-1259.

Hernerici M, Hulsbomer HB, Hefter H et al (1987) Natural history of asymptomatic extracranial arterial disease. Brain 110: 777-791.

Hobson RW 11, Weiss DG, Fields WS et al (1993) Efficacy of carotid endarterectomy for asymptomatic carotid stenosis. New England Journal of Medicine 328: 221-227.

MACE (Mayo Asymptomatic Carotid Endarterectomy) Study Group (1992) Results of a randomized controlled trial of carotid endarterectomy for asymptomatic carotid stenosis. Mayo Clinic Proceedings 67: 513-518.

McCrory DC, Goldstein LB, Samsa GP et al (1993) Predicting complications of carotid endarterectomy. Stroke 24: 1285-1291.

Mayberg MR, Wilson SE, Yatsu F et al (1991) Carotid Endarterectomy and prevention of cerebral ischemia in symptomatic carotid stenosis. Journal of the American Medical Association 266: 3289-3294.

Moniz E, Lima A & de I’Acerda R (1937) Hemiplegies par thrombose de la carotide inteme. Presse Medicale 1: 977-980.

Norris JW, Zhu CZ, Bornstein NM & Chambers BR (1991) Vascular risks of asymptomatic carotid stenosis. Stroke 22: 1485-1490.

North American Symptomatic Carotid Endarterectomy Trial Collaborators (1991) Beneficial effect of carotid endarterectomy in symptomatic patients with high-grade carotid stenosis. New England Journal of Medicine 325: 445-453.

O’Holleran LW, Kennelly MM, McClurken M & Johnson JM (I 987) Natural history of asymptomatic carotid plaque. Five year follow-up study. The American Journal of Surgery 154: 659-661.

Pelz DM, Buchan A, Fox AJ et al (1986) Intraluminal thrombus of the internal carotid arteries: angiographic demonstration of resolution with anticoagulant therapy alone. Radiology 160: 369-373.

Shanik GD, Moore DJ, Leahy A et al (1992) Asymptomatic carotid stenosis: a benign lesion? European Journal of Vascular Surgery 6: 10-15.

Sharpe BL, Finan JW, Richardson DJ et al (1995) No difference in the risks of stroke between women and men who have symptomatic high grade carotid stenosis. Stroke 26: 184.

Shaw DA, Venables GS, Cartlidge NEF et al (1984) Carotid endarterectomy in patients with transient cerebral ischemia. The Journal of the Neurological Sciences 64: 45-53.

Streifler JY, Eliasziw M, Hachinski VC et al for the NASCET Group (1992a) Symptomatic high grade carotid stenosis in the elderly poor prognosis, high response to surgery: observations from NASCET (Abstract). Journal of Stroke and Cerebrovascular Diseases 2 (supplement 1): 548.

Streifler JY, Fox AJ, Wong CJ et al for the NASCET Group (1992b) Importance of silent brain infarctions in TIA patients with high grade carotid stenosis: results from NASCET. 1992 AAN Scientific Program, San Diego, CA. Neurology 42 (supplement 3): 2265.

Streifler JY, Eliasziw M, Benavente OR et al for the North American Symptomatic Carotid Endarterectomy Trial (1995a) Lack of relationship between leukoaraiosis and carotid artery disease. Archives of Neurology 52: 21-24.

Streifler JY, Eliasziw M, Benavente OR et al (1995b) The risk of stroke in patients with first ever retinal versus hemispheric transient ischemic attacks and high grade carotid stenosis. Archives of Neurology 52: 246-249.

Warlow CP (1984) Carotid endarterectomy: does it work? Stroke 15: 1068-1076.

Winslow CM, Solomon DH, Chassin MR et al (1988) The appropriateness of carotid endarterectomy. The New England Journal of Medicine 318: 721-727.