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


Review: Carotid endarterectomy modestly reduces ipsilateral stroke in asymptomatic stenosis

ACP J Club. 1999 May-June;130:59. doi:10.7326/ACPJC-1999-130-3-059

Source Citation

Benavente O, Moher D, Pham B. Carotid endarterectomy for asymptomatic carotid stenosis: a meta-analysis. BMJ. 1998 Nov 28;317: 1477-80.



Does carotid endarterectomy (CE) reduce stroke in patients with asymptomatic stenosis?

Data sources

Studies were identified by searching MEDLINE (1966 to January 1998) using the terms carotid stenosis; endarterectomy, carotid, asymptomatic; clinical trial; and randomized controlled trial. The Cochrane Controlled Trials Register, Ottawa Stroke Trials Register, Current Contents (1995 to 30 January 1998), 3 journals (New England Journal of Medicine, JAMA, and Stroke), and bibliographies of relevant studies were reviewed, and experts and authors were contacted.

Study selection

Randomized controlled trials were selected if CE was compared with standard medical treatment and patients had confirmed asymptomatic carotid stenosis with no history of cerebrovascular disease; confirmed asymptomatic carotid stenosis with previous stroke or transient ischemic attack in the vertebrobasilar circulation or contralateral carotid territory; or previous contralateral CE.

Data extraction

Data were extracted in duplicate on patient and trial characteristics; study quality; degree of stenosis; use of anti-thrombotic therapy; and outcomes (perioperative complications [stroke or death], stroke ipsilateral to the qualifying stenosis, and all stroke) within 30 d of randomization.

Main results

5 trials (1215 patients in the CE groups and 1225 in the medical care groups) met the inclusion criteria. 74% of patients were men, mean age range was 64 to 74 years, and mean follow-up was 3.1 years. CE increased stroke and death in the 30-day perioperative period (P < 0.001) but overall reduced the combined end point of ipsilateral stroke plus perioperative complications (P = 0.005), all ipsilateral stroke (P < 0.001), and all stroke plus perioperative death (P = 0.007).


For patients with asymptomatic stenosis, carotid endarterectomy transiently increases the risk for stroke and death but overall reduces the combined end points of ipsilateral stroke plus perioperative stroke or death, all ipsilateral stroke, and all stroke plus perioperative death.

Source of funding: No external funding.

For correspondence: Dr. O. Benavente, Department of Medicine, Division of Neurology, University of Texas Health Science Center, San Antonio, TX 78284-7883, USA. FAX 210-567-4659.

Table. Carotid endarterectomy (CE) vs medical care for patients with asymptomatic carotid stenosis*

Outcomes at mean 3.1 y Weighted event rates RRI (95% CI) NNH (CI)
CE Medical care
PO stroke or death 2.5% 0.4% 423% (127 to 1107) 49 (34 to 91)
RRR (95% CI) NNT (CI)
Stroke plus PO complications 5.0% 7.4% 36% (12 to 54) 42 (24 to 185)
IP stroke and PO IP stroke 3.7% 7.1% 53% (33 to 67) 30 (20 to 61)
All stroke plus PO complications 7.9% 10.7% 30% (9 to 45) 36 (20 to 178)

*IP = ipsilateral; PO = perioperative (within 30 d of randomization). Other abbreviations defined in Glossary; RRI, RRR, NNH, NNT, and CI calculated from data in article by using a fixed-effects model.


The benefits of medical intervention should exceed the risks. If the intervention is surgical and the condition is asymptomatic, at least 2 criteria must apply: The risks of not operating must exceed the risks of the procedure (including the work-up) plus the subsequent long-term risk, and the absolute benefits must justify the personal and societal costs, whether financial, psychological (pain, stress, or induced anxiety), or social (difficulty in meeting expectations or ensuring equity).

With almost 8000 patient-years of follow-up data from randomized trials, we can now be reasonably sure that CE, if done at centers with established expertise, can reduce the risk for stroke in some patients with asymptomatic moderate or severe carotid stenosis. However, we still do not know which patients are likely to be helped or harmed.

Before having angiography or CE, patients with asymptomatic carotid stenosis should understand that without surgery the average risk for ipsilateral stroke is only 2%/y, that 40 to 50 patients will need operations to prevent 1 additional stroke at 3 years, that the risks and benefits in terms of disabling stroke are still unknown, and that CE will do more harm than good if perioperative complication rates are > 4%.

Tax or insurance payers should understand that the cost of CE per stroke prevented is at least U.S. $500 000 (1) and many times that of simpler (although still underused) approaches, such as the treatment of hypertension in persons > 70 years of age.

Further research is therefore needed to identify higher-risk patients with asymptomatic stenosis who may benefit substantially from CE (2). Efforts should target such patients, but the public health benefits of surgery for primary stroke prevention will remain small.

David Barer, MSc, DM
Queen Elizabeth HospitalGateshead, England, UK


1. Derdeyn CP, Powers WJ. Stroke. 1996; 27:1944-50.

2. Warlow C. BMJ. 1998;317:1468.