The role of surgery in symptomatic carotid stenosis is controversial. Long-term benefits, particularly the prevention of stroke, may not outweigh the risks of surgery. There is also uncertainty regarding the medical treatment of carotid stenosis, the role of comorbidities and difficulties in defining the degree of stenosis, significant subgroups of patients and the experience/skill of individual surgeons. In the largest randomized clinical trial to date, the European Carotid Surgery Trial compared the outcome of early surgical intervention with delaying surgery as long as possible after an ischemic event attributable to carotid artery stenosis.

Patients were eligible for the study if within the previous six months they had experienced an event suspicious for carotid ischemia not resulting in serious permanent disability. All patients underwent contrast angiography of the symptomatic artery. The percentage of stenosis in the symptomatic artery was estimated by a single researcher for all patients. Patients were then randomly allocated to either the surgical treatment group or the control group. All patients received optimal medical treatment as determined by their physician. This included advice about the risk factors of smoking, treatment of hypertension and antiplatelet medication. Patients were followed at four and 12 months during the first year and then annually until the end of 1995. The mean duration of follow-up was 6.1 years.

The two groups had some small baseline differences: the surgical group had a slightly higher prevalence of both hypertension and ischemic heart disease. However, these differences were unlikely to have been clinically relevant. There was a nonsignificant tendency for patients in the control group to be treated more aggressively with antiplatelet and lipid-lowering drugs during the follow-up period.

In the group undergoing immediate surgery, 669 patients (37 percent) died or experienced a major stroke during the follow-up period. The risk of stroke or death was highest within 30 days of surgery. Of the 1,745 patients assigned to the surgery group, the risk of non-fatal major stroke or death within 30 days of surgery was 7 percent. In the control group, 442 patients (36.5 percent) died or experienced a major stroke during the follow-up period. The risk of major stroke was clearly related to the age of the patient and to the severity of carotid stenosis, but only during the first two to three years following randomization. The authors compiled survival curves to estimate the balance of benefit and harm at different ages and with different degrees of carotid stenosis. These models show that in general the advantages of surgery at three years outweighed the risks only when stenosis was greater than 80 percent of the diameter of the artery. The cutoff points for benefit varied with the age and sex of the patient. In general, men benefited slightly more from surgery than did women.

The authors conclude that decisions about endarterectomy must be based on factors such as patient characteristics (age and sex) and the physician's surgical expertise, but, in general, the balance of risk and benefit does not favor surgery when less than approximately 80 percent of the artery is occluded following a symptomatic episode of carotid ischemia.

editor's note: Applying the results of this study to one's practice is challenging in two respects. First, the results are not quite as clear-cut as described above. The survival diagrams illustrate large zones of “no clear result” and delineate hazard or benefit from surgery by the age of the patient and the degree of stenosis. For a 70-year-old man, the zone of “no clear hazard or benefit” starts around stenosis of 20 percent and extends to approximately 80 percent. For women, clinical decisions may be somewhat easier, as the corresponding area of uncertainty extends from about 70 to 90 percent. In perhaps the majority of patients, however, there can be no confidence that surgery will prevent the very serious consequences of cerebral ischemia. Although we are accustomed to making clinical decisions based on incomplete information, the second challenge—balancing the instinct to actively intervene to help a patient at risk of stroke with the need for objectivity and prudence in decision-making—appears particularly vexing in this situation.—a.d.w.