Current issues of ACP Journal Club are published in Annals of Internal Medicine


Carotid endarterectomy reduced cardiovascular events in symptomatic carotid stenosis

ACP J Club. 1992 Mar-April;116:42. doi:10.7326/ACPJC-1992-116-2-042

Source Citation

Mayberg MR, Wilson SE, Yatsu F, et al. Carotid endarterectomy and prevention of cerebral ischemia in symptomatic carotid stenosis. JAMA. 1991 Dec 18;266:3289-94.



To determine whether carotid endarterectomy reduces the risk for cerebral infarction or crescendo transient ischemic attacks (TIA) among men with recent ischemic cerebrovascular events and carotid stenosis.


Randomized controlled trial.


16 university-affiliated Veterans Affairs medical centers.


5000 men were screened; 189 (80%) of 237 eligible patients were selected who had TIA, transient monocular blindness, or small completed strokes within the previous 120 days, and ipsilateral, angiographically defined carotid stenosis > 50% without distal, more severe stenosis or occlusion. Patients were excluded if duplex ultrasound screening had shown < 50% ipsilateral stenosis or if they had global ischemia, focal seizures, vertebrobasilar ischemia, potential cardiac source of emboli, major strokes, previous carotid surgery, or other serious illness. 1 patient was lost to follow-up.


All patients were prescribed enteric-coated aspirin, 325 mg/d. Ninety patients had carotid endarterectomy (average delay 2.2 d after entry). Surgery was not standardized.

Main outcome measures

Ipsilateral cerebral or retinal infarction, crescendo TIA (TIAs of increased frequency, duration, or severity), or death from any cause within 30 d of randomization.

Main results

The trial was stopped early because similar studies showed benefit of carotid endarterectomy for high-grade stenosis. Compared with angiography, duplex ultrasound underestimated the category of stenosis in some vessels {19%}* and overestimated stenosis in others {16%}*. Angiography led to transient complications in 9% and to lasting complications in none. After a mean follow-up of 11.9 months, 7.7% of surgical patients and 19.4% of nonsurgical patients had primary events (P = 0.01). {This absolute risk reduction of 11.7% means that 9 patients would need to be treated with surgery (rather than no surgery) to prevent 1 additonal primary event, 95% CI 5 to 50; the relative risk reduction was 60%, CI 13% to 82%.}* This absolute risk reduction for the combined primary end points of 17.7% (P = 0.004) for 129 patients with > 70% stenosis, and -0.4% (P > 0.2) for 58 patients with 50% to 69% stenosis. 3 perioperative deaths occurred in the surgical group compared with no deaths in the comparable period for the nonsurgical group; 12 of 19 primary end points in the nonsurgical group were crescendo TIAs compared with none in the surgical group.


Carotid endarterectomy was beneficial for selected men with recent hemispheric transient ischemic attacks or nondisabling strokes and ipsilateral internal carotid artery stenosis.

Source of funding: Department of Veterans Affairs Cooperative Studies Program.

Address for article reprint: Dr. M.R. Mayberg, Department of Neurosurgery, RI-20, University of Washington, Seattle, WA 98195, USA.

*Numbers calculated from data in article.


This is the report of the third randomized trial declaring benefit for carotid endarterectomy in symptomatic patients. Its early termination was necessitated by the positive results achieved for patients with severe stenosis in the North American (NASCET) (1) and European (ECST) (2) trials. Unlike NASCET and ECST, no benefit was demonstrated until crescendo TIA (a prognostic indicator for stroke and death) was added as an outcome event; reduction of TIAs by surgery indicates that the results of this trial were going in the right direction. 3 reasons may account for the failure to detect benefits in stroke and death end points: The numbers were small (n = 193); follow-up was short (mean, 11.9 months); and 29% of surgical and 30% of medical patients were randomized with only 50% to 70% stenosis.

Analyses in NASCET show decreasing benefit of endarterectomy with diminishing stenosis. Benefit for patients with stenosis below 70% remains unknown. Because of this, NASCET and ECST continue to randomize symptomatic patients with stenosis between 30% and 69%.

In Mayberg and colleagues' study, 2104 of the 5000 patients studied as possible candidates were rejected because of a carotid duplex indicating < 50% stenosis. However, comparing the noninvasive studies with arteriography among included patients, ultrasound frequently overestimated or underestimated the stenosis. NASCET data confirm this poor correlation.

Physicians must now consider carotid endarterectomy to be a proven treatment strategy for patients with 70% to 99% stenosis. Because of the poor performance of ultrasound, physicians need to ensure that the hazards of arteriography are within an acceptable range in their radiology suites (3). They will be obliged as well to know that surgeons in their institutions have comparable skills to those shown in these trials.

H. J. Barnett, MD
Robarts Research InstituteLondon, Ontario, Canada

H. J. Barnett, MD
Robarts Research Institute
London, Ontario, Canada


1. NASCET Collaborators. N Engl J Med. 1991;325:445-53.

2. European Carotid Surgery Trialists' Collaborative Group. Lancet. 1991;337:1235-43.

3. Dion JE, Gates PC, Fox AJ, Barnett HJ, Blom R. J. Stroke. 1987;18:997-1004.