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


Risk for stroke increased with degree of carotid stenosis but was low in asymptomatic carotid arteries

ACP J Club. 1995 July-Aug;123:3. doi:10.7326/ACPJC-1995-123-1-003

Source Citation

The European Carotid Surgery Trialists Collaborative Group. Risk of stroke in the distribution of an asymptomatic carotid artery. Lancet. 1995 Jan 28; 345:209-12.



To determine the risk for stroke in the distribution of asymptomatic carotid arteries in patients who received standard medical therapy.


Follow-up for a mean of 4.5 years in patients randomly assigned to medical therapy in the European Carotid Surgery Trial (ECST).


100 centers in 14 European countries.


2295 patients (mean age 62 y, 70% men) who had had a carotid-distribution transient ischemic attack, a minor ischemic stroke, a nondisabling major ischemic stroke, or a retinal infarction within the previous 6 months. Patients received medical care, which included aspirin or another antiplatelet drug, treatment of hypertension, and advice to stop smoking. Exclusion criteria were history of transient ischemic attacks or strokes in the territory of both carotid arteries, bilateral endarterectomy shortly after randomization, or absent or inadequate angiographic view of the asymptomatic carotid artery.

Assessment of prognostic factors

Data pertaining to previous angina or myocardial infarction (MI), presence of peripheral vascular disease or diabetes, blood pressure, mean cholesterol level, current smoking status, mean obesity index, and degree of stenosis in both internal carotid arteries were recorded at baseline.

Main outcome measures

First stroke, hemorrhage, or infarction in the distribution of the asymptomatic carotid artery that was fatal or lasted > 7 days.

Main results

During follow-up, 69 carotid territory strokes occurred, 9 of which were fatal. The overall Kaplan-Meier estimate of stroke risk at 3 years was 2.1% (95% CI 1.2% to 2.8%). The stroke risk was almost identical in patients with mild (0% to 29%) and moderate (30% to 69%) stenosis (1.8% vs 2.1%). The stroke risk at 3 years was 5.7% (CI 1.5% to 9.8%) in patients with 70% to 99% stenosis and, among these patients, 9.8% (CI 0.6% to 21.0%) in those with 80% to 89% stenosis and 14.4% (CI 5.0% to 38.0%) in patients with 90% to 99% stenosis. These risks were not statistically greater than the stroke risk in the rest of the group. The 3-year risk for fatal stroke was 0.3% (CI 0.06% to 0.56%) and for disabling or fatal stroke was 1.0% (CI 0.3% to 1.8%).


The risk for stroke in the distribution of the asymptomatic carotid artery in patients receiving standard medical care increased with the degree of stenosis, but was so low over-all that population screening cannot be justified.

Source of funding: Medical Research Council.

For article reprint: Dr. P.M. Rothwell, Department of Clinical Neurosciences, Western General Hospital, Crewe Road, Edinburgh EH4 2XU, United Kingdom. FAX 44-1865-790493.


Endarterectomy reduced the risk for stroke in asymptomatic carotid artery stenosis

At first glance, these 2 large prospective studies seem to be in conflict because they arrive at different overall conclusions. This is probably because the conclusions are based on different experimental designs and methods of analysis. Critical differences between the Asymptomatic Carotid Atherosclerosis Study (ACAS) and ECST include the following: 1) The ACAS was a controlled treatment study that included some patients who had had surgery and some who had not. 2) The ACAS was smaller; it included 825 patients treated surgically and 834 patients treated medically, whereas ECST included 2295 medically treated patients. 3) The ACAS had a median follow-up of 2.7 years, and ECST had a mean follow-up of 4.5 years. 4) Kaplan-Meier estimates were for 3 years in ECST and for 5 years in ACAS. 5) All the arteries opposite to the asymptomatic artery were symptomatic in ECST compared with only 25% in the ACAS. 6) In the ECST, the degree of stenosis was determined by arteriography. In the ACAS, the initial study for randomization could be ultrasonic or angiographic, but all patients assigned to surgery and 38% of the medical group had angiography. 7) The degree of stenosis was measured differently on arteriography. 8) Possibly the most important difference is that in the ECST a stroke was defined as either being fatal or with deficits persisting longer than 7 days, whereas ACAS defined stroke as any deficit persisting longer than 24 hours.

The ECST, because of the low stroke risk in the distribution of asymptomatic arteries and the small potential benefit of CEA, concluded that population screening is not justified. The ACAS, on the basis of a projected 5-year aggregate risk reduction of 53%, concluded that for patients in good health with CAS of ≥ 60%, CEA in the hands of a surgeon with < 3% perioperative morbidity and mortality is beneficial. Although CEA significantly reduced the estimated 5-year risk for ipsilateral stroke by 1% per year, major ipsilateral stroke or perioperative death was not significantly decreased (24 events in the medical group and 21 in the surgical).

In the ECST, Kaplan-Meier 3-year estimates of risk for stroke by deciles of stenosis were very low for all deciles until 80% to 89%. At this point, the risk increases from < 2% to 9.8%. In the ACAS, only the patients who had arteriography in the 6 months preceding randomization were analyzed by decile of stenosis. These 313 medical patients and the 329 surgical patients had a total of 16 and 11 events in 2.7 years of follow-up, respectively (8 in the medical group and 7 in the surgical group in the 60% to 69% decile, 5 and 2 in the 70% to 89% decile, and 3 and 2 in the 90% to 99% decile). These numbers are too small to draw any conclusions concerning differences between these subgroups. Thus, the ACAS data do not refute the ECST conclusion that CEA would probably not be beneficial for a patient with an asymptomatic artery with < 80% stenosis.

With the key assumption that the surgeon has an established low complication rate, the combined data from ECST and ACAS lead this reviewer to conclude that surgery should be done in cases of asymptomatic CAS of 80% to 99%, that surgery should not be done in cases of CAS < 70%, and that the issue of surgery in cases of CAS of 70% to 79% cannot be resolved until further data are available. Although overall conclusions drawn in ECST have not been refuted, it would seem to be appropriate to screen patients who have other major risk factors for stroke. A randomized European Asymptomatic Carotid Surgery trial is ongoing (1). Along with these 2 studies, it may give us enough data to determine how best to treat the patients with asymptomatic stenosis < 80%.

Mark L. Dyken, MD
Indiana University School of MedicineIndianapolis, Indiana, USA


1. Halliday AW, Thomas D, Mansfield A. The Asymptomatic Carotid Surgery Trial (ACST). Rationale and design. Steering Committee. Eur J Vasc Surg. 1994;8:703-10.