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


Complications after carotid endarterectomy were not predicted by sociodemographic or hospital factors in elderly patients

ACP J Club. 1991 Mar-Apr;114:56. doi:10.7326/ACPJC-1991-114-2-056

Source Citation

Brook RH, Park RE, Chassin MR, et al. Carotid endarterectomy for elderly patients: predicting complications. Ann Intern Med. 1990;113:747-53.



To determine whether elderly patients' risk for death, stroke, or myocardial infarction after carotid endarterectomy can be predicted from information about their sociodemographic status, their disease, their surgeons, or their hospitals.


Historical cohort study based on hospital medical records and national mortality data.


Hospitals selected for high, average, and low rates of carotid endarterectomy in 3 areas of the United States.


1171 patients had complete data and were analyzed. They constituted 90% of the patients, 65 years of age or older, who had had carotid endarterectomies by participating surgeons, randomly selected from the study areas. (89% of invited surgeons, who performed 89% of carotid endarterectomies, agreed to participate.)

Assessment of prognostic factors

Patient sociodemographic descriptors were age, sex, race, and income. Surgeon descriptors were age, medical school, whether any residency training occurred in the study area, board certification, and number of carotid endarterectomies done in 1981 on patients over 65 years of age. Hospital descriptors were number of beds, occupancy rate, ownership, for-profit status, and teaching status. Disease descriptors were myocardial infarction, premature ventricular contraction, hypertension, congestive heart failure, limited activities of daily living, carotid transient ischemic attack, vertebral-basilar transient attack, old stroke, evolving stroke, before coronary artery bypass surgery, and recent stroke. Multivariate analysis controlled for disease severity, comorbidity, and other factors when each descriptor was studied.

Main outcome measures

Death from any cause within 30 days of surgery; stroke (new, nonfatal neurologic deficit impairing function) present at discharge; myocardial infarction (by electrocardiogram and enzyme criteria) during hospitalization. Only 1 outcome was counted per patient.

Main results

11.3% of patients had an in-hospital adverse event: 6.6% had a postoperative stroke, 1.7% had a myocardial infarction, and 3% died within 30 days of hospitalization. After controlling for disease severity and comorbidity, the only patient demographic physician, or hospital characteristic descriptor (among the 14 studied) of a higher complication rate [19.4% vs 10.6%, P < 0.01] (relative risk [RR] 1.91, 95% CI 1.12 to 3.13) was the surgeon's graduation from a foreign, but not Canadian or Western European, medical school. Severity variables associated with an adverse outcome were carotid transient ischemic attack (RR 74, CI 1.06 to 2.79); recent stroke (RR 2.42, CI 1.10 to 4.84); myocardial infarction within 6 months (RR 2.87, CI 1.09 to 6.41); and evolving stroke (RR 7.58, CI 1.54 to 15.04).


When disease severity and comorbidity were taken into account, the complication rate after carotid endarterectomy could not be predicted from sociodemographic or hospital descriptors. Surgeons trained in the United States, Canada, or Western Europe had lower complication rates.

Sources of funding: The American Association of Retired Persons; The Commonwealth Fund; The John A. Hartford Foundation; the Health Care Financing Administration of the U.S. Department of Health and Human Services; The Pew Memorial Trust; the Robert Wood Johnson Foundation.

Address for article reprint: Dr. R.H. Brook, The Rand Corporation, 1700 Main Street, Santa Monica, CA 90406, USA.


To understand the context of this study, read the article published by this group in March 1988 (1). The Rand study was done to evaluate the "appropriateness" with which carotid endarterectomy was done on the elderly patients. The term "appropriate" was applied to indications where benefit exceeds risk as defined by the literature and a consensus panel of "experts." The conclusion of the study was that 64% of the cases were done for less than "appropriate" indications. This may be the reason why the study failed to predict risk from patient, hospital, and surgeon predictors.

Vascular surgeons agree that carotid endarterectomy should not be done if the combined perioperative stroke and mortality rate exceeds 4% (2). Because the mortality for this procedure is often caused by a cardiac problem, careful patient selection and pre- and perioperative evaluation and care should reduce this rate. Although mortality was 3% in this study, the literature suggests that less than 1% is attainable (3). The complication of stroke should then be the major risk factor and should be kept to less than 3%, as shown in a recently published trial (4). Carotid endarterectomy is a therapeutic option for the right patient (5). There are surgeons and centers with acceptable results, and I agree with the authors' recommendation for careful referrals.

P. L. Ergina, MD, MPH
Royal Victoria HospitalMontreal, Quebec, Canada


1. Winslow CM, Solomon DH, Chassin MR, et al. The appropriateness of carotid endarterectomy. N Engl J Med. 1988; 318:721-7.

2. Quill DS, Sumner DS. What are valid indications for carotid endarterectomy? Adv Surg. 1989;22:277-300.

3. Maini BS, Mullins TF, Catlin J, et al. Carotid endarterectomy: a ten-year analysis of outcome and cost of treatment. J Vasc Surg. 1990;12:732-40.

4. Towne JB, Weiss DG, Hobson RW. First phase report of cooperative Veterans Administration asymptomatic carotid stenosis study—operative morbidity and mortality. J Vasc Surg. 1990;11:252-9.

5. Cebul RD, Whisnant JP. Carotid endarterectomy. Ann Intern Med. 1989;111:660-70.