Multicenter Review of Preoperative Risk Factors for Endarterectomy for Asymptomatic Carotid Artery Stenosis
Background and Purpose—The benefit of carotid endarterectomy is highly dependent on surgical risk. However, little data are available concerning factors affecting the risk of endarterectomy performed for asymptomatic carotid artery stenosis outside the setting of a randomized controlled trial. The purpose of this study was to analyze the impact of potential preoperative risk factors on the frequency of postoperative complications in patients undergoing the operation for asymptomatic disease in academic medical centers.
Methods—Data regarding postoperative complications were systematically abstracted from the medical records of a random sample of patients who underwent carotid endarterectomy at 12 academic medical centers.
Results—Of 1160 procedures reviewed, 463 (40%) were performed for asymptomatic disease. Postoperative stroke or death occurred in 13 (2.8%), and myocardial infarction occurred in 8 (1.7%). The rate of postoperative stroke or death was lower in asymptomatic patients than in those with a history of cerebrovascular symptoms in a different vascular distribution, but the difference was not significant (1.8% versus 4.2%; P=.21). There were no significant differences in these rates based on race, a history of angina, recent myocardial infarction, chronic obstructive pulmonary disease, hypertension, the degree of stenosis of the contralateral or ipsilateral carotid artery, or the presence of angiographically recognized ulceration, intraluminal thrombus, or siphon stenosis in the ipsilateral vessel (χ2; P>.05). Postoperative stroke or death was more frequent in women (5.3% versus 1.6% in men; P=.02), in those aged 75 years or older (7.8% versus 1.8% in those younger than 75 years; P=.01), and in those with a history of congestive heart failure (8.6% versus 2.3% in those without a history of congestive heart failure; P=.03). The risk of stroke or death was higher in the 16 patients who had carotid endarterectomy performed in combination with coronary artery bypass surgery than in those who had only endarterectomy (18.7% versus 2.1%; P<.001).
Conclusions—The overall risk of postoperative stroke or death was nearly twice that reported by Asymptomatic Carotid Atherosclerosis Study (ACAS) investigators in the setting of a clinical trial but was within acceptable guidelines. Women were at higher postoperative risk than men, which supported ACAS findings. Additional high-risk groups were those aged 75 years or older, those with a history of congestive heart failure, and those undergoing prophylactic endarterectomy for asymptomatic stenosis in combination with coronary surgery. Knowledge of these rates may help to better assess an individual’s postoperative risk and therefore the anticipated benefit of surgery.
The benefit of carotid endarterectomy in comparison to medical therapy alone is highly dependent on surgical risk.1 2 3 4 5 A critical feature of the Asymptomatic Carotid Atherosclerosis Study (ACAS), which demonstrated the efficacy of the operation when performed in selected patients with asymptomatic disease, was the very low observed rate of perioperative complications (estimated combined morbidity and mortality rate of 2.7%).6 Central to this low complication rate was the careful selection of surgeons with extensive experience and demonstrated competence.7 8 In addition to the skill of the surgeon, the risk of carotid endarterectomy may also be related to characteristics of the patient undergoing the operation. In ACAS, a history of diabetes, contralateral carotid artery stenosis, and never consuming alcoholic beverages were associated with a higher risk of perioperative complications.7 In addition, although not statistically significant, lower estimates of efficacy in women than in men were attributed to their higher risk of surgical complications.6
Little data are available concerning the impact of specific patient characteristics on the risk of carotid endarterectomy in asymptomatic patients outside the setting of randomized controlled trials. We previously performed a review of carotid endarterectomies performed at 12 academic medical centers and developed an overall risk model based on preoperative factors.9 A second analysis focused on patients undergoing carotid endarterectomy for symptomatic disease.10 The goal of the present study was to identify preoperative clinical factors associated with an increase in the risk of carotid endarterectomy in patients who underwent the operation for asymptomatic stenosis.
Subjects and Methods
Details of the primary data collection methodology have been reported previously.9 Briefly, 100 carotid endarterectomies were selected randomly from all those performed during the calendar years 1988 to 1990 at each of 11 of the 12 member hospitals of the Academic Medical Centers Consortium (see the Appendix). The cases were identified from administrative records without additional knowledge of the patients’ preoperative clinical status or subsequent course. Because of a lower operative volume, the 100 procedures were randomly selected from those performed during the calendar years 1987 to 1990 in the 12th hospital. Forty cases were subsequently excluded from further analysis because the operative procedure was miscoded in administrative records. Records from the remaining 1160 carotid endarterectomies constitute the primary data set.
Hospital charts for the admission during which carotid endarterectomy was performed were systematically reviewed by abstractors using a defined protocol. Data included patient demographics, comorbid diseases, neurological symptoms (type, location, and frequency of symptoms), data from the radiologists’ reading of preoperative carotid angiograms, clinical details from the operative report, and postoperative complications. When sufficient data were available, the patient’s preoperative functional abilities were rated (impaired or unimpaired in activities of daily living). Angiographic risk factors were considered “absent” unless specifically recorded in the radiologist’s report. Recorded operative factors included the use of a shunt, the use of a patch graft, and whether prophylactic carotid endarterectomy was performed as a combined procedure with coronary artery bypass graft (CABG) surgery. Abstracted data forms (along with xerographic copies of discharge summaries, angiogram, and operative reports) were reviewed by a study investigator. Adverse outcomes included postoperative in-hospital strokes, myocardial infarctions, and deaths occurring during the hospitalization for carotid endarterectomy.
Univariate Fisher’s exact tests were used to identify potentially important predictors, and χ2 statistics were used for comparisons of categorical data as indicated. Logistic regression modeling was not appropriate because the number of outcome events was relatively small and the number of potential risk factors was relatively large (regression analysis is unreliable when there are fewer than 10 times as many outcome events as there are candidate variables).11
An initial analysis of the data failed to reveal any significant differences in adverse outcomes among the 12 hospitals constituting the Academic Medical Center Consortium.9 Therefore, hospital-level variables were not further considered, and the data were collapsed for further analysis.
Forty percent of the 1160 patients undergoing carotid endarterectomy were asymptomatic or had symptoms in a vascular territory other than that of the operated artery (n=463). We previously found that the rate of postoperative complications (either stroke, myocardial infarction, or death) was lower in patients undergoing endarterectomy for asymptomatic stenosis than in those undergoing the operation for symptomatic disease (4.8% versus 8.5%; P<.01).10 This difference was largely related to a lower risk of postoperative stroke in the patients undergoing the operation for asymptomatic carotid artery stenosis (2.2% versus 5.7%; P<.004). The rates of postoperative myocardial infarction (2.4% versus 3.2%; P=.40) and death (1.1% versus 1.6%; P=.50) were also lower in these patients, but the differences were not significant compared with patients with ipsilateral symptoms. The combined risk of postoperative stroke or death in patients undergoing carotid endarterectomy for asymptomatic disease was 2.8% compared with 6.2% for patients undergoing the operation for symptomatic disease (P=.01). A total of 189 of the 463 patients (41%) who had endarterectomies performed for asymptomatic stenosis had cerebrovascular symptoms in other vascular distributions. Although these patients were at higher risk of complications than asymptomatic patients (4.2% versus 1.8%), the difference was not significant (P=.21). These subgroups were combined for further analyses.
The Table⇓ presents a list of candidate variables with the frequencies of postoperative stroke and death and associated univariate probabilities. Significant medical and demographic variables included sex, age 75 years or older, and a history of congestive heart failure. There were no relationships between a variety of angiographically defined potential risk factors and postoperative complications. Stroke or death was more common among patients in whom the surgeon used a patch graft (3.9% versus 1.1%), but the difference was not significant (P=.08). Patients who had prophylactic endarterectomy in association with CABG were at particularly high risk of postoperative stroke or death (18.7% versus 2.1%; P<.001).
The overall rate of postoperative nonfatal stroke or death (2.8%) for patients undergoing carotid endarterectomy for asymptomatic stenosis was within the guidelines recommended by the American Heart Association.12 This rate of postoperative adverse events for endarterectomy performed for asymptomatic disease is similar to that reported in a recent systematic literature review (3.35% rate of perioperative stroke or death).13 However, it should be noted that because the present study was retrospective, the traditional definition of perioperative complications as those occurring within 30 days of the surgical procedure was not employed. Only those events recorded during the index hospitalization were recorded. Therefore, the 2.8% rate of postoperative stroke or death may represent an underestimate of the true complication rate at the time the study data were collected. This rate is nearly twice that later obtained by ACAS surgeons (1.5% risk of stroke or death within 30 days of carotid endarterectomy) 6 but somewhat less than the 5.3% reported by Fode et al14 based on a community survey of complications of carotid endarterectomies performed for asymptomatic bruit carried out more than a decade previously. Although the difference between this latter rate and that found in the present review is not statistically significant (χ2; P=.10), the decrease is consistent with other data suggesting that the safety of the operation has improved over the last decade.15
Female sex, age 75 years or older, and a history of congestive heart failure were each associated with a higher risk of postoperative stroke or death. These factors were identified based on univariate tests of significance, and there may be significant intercorrelations among the selected variables. Because of the small number of outcome events, the data were not amenable to formal regression modeling, and we cannot comment on the independent contributions of the individual variables. In addition, the power to detect a difference in some variables may be limited.
The validity of factors identified through analyses performed in a given study such as ours can be supported if similar results are found in an independent data set. Although approximately one third of the centers participated in ACAS, the number of ACAS patients in the present sample was negligible. The ACAS investigators found a lower point estimate for the efficacy of carotid endarterectomy in women, which was attributed to a higher, albeit nonsignificant, sex-based complication rate (3.6% in women versus 1.7% in men; P=.08).6 Although the potential reasons for this increased perioperative risk in women remain speculative, similar results in these two very different types of studies suggest that the finding is not spurious. Age 75 years or older and a history of congestive heart failure were also associated with an increased risk of postoperative complications in the present analysis. In ACAS, those older than 75 years had a 3.8% incidence of postoperative stroke or death compared with 2.2% of those younger than 75 years (P=.25).7 Those with a history of congestive heart failure had a 5.6% postoperative risk versus a 2.3% risk in patients without such a history (P=.39).7 Therefore, as with sex, these variables were also associated with a somewhat higher (but not statistically significant) perioperative risk. However, the magnitude of these differences was greater in the present analysis based on patients having the operation outside the setting of a randomized trial.
Analysis of ACAS data revealed that a prior history of stroke, contralateral stenosis greater than 60%, contralateral siphon stenosis, and never consuming alcoholic beverages were associated with a higher risk of perioperative events.7 Those with a history of transient ischemic attack or stroke in a vascular distribution other than that of the operated artery were also at somewhat higher risk in the present study (4.2% versus 1.8%), but the difference was not significant. There was also a trend toward higher complication rates in patients with higher degrees of contralateral stenosis (3.6% to 4.0% versus 1.4% to 1.6%; Table⇑), but again, the difference was not significant. These potential risk factors should be further examined in future analyses. We did not collect data concerning the patient’s history of alcohol consumption or the degree of contralateral siphon stenosis.
The few patients undergoing prophylactic endarterectomy for asymptomatic stenosis in combination with CABG surgery were at particularly high risk for postoperative complications. Because there were no comparison control data, it is not possible to determine how a similar group of patients having only CABG would have fared. The approach to these types of patients remains controversial16 17 ; however, given the apparent perioperative risk of complications in those having combined procedures compared with the perioperative incidence of stroke in patients undergoing CABG alone, extreme caution should be exercised.18 19
Several important caveats should again be stressed. First, this was a retrospective survey with all of its inherent limitations. Unless a factor was noted in the medical record, it was considered to be absent. This is particularly important with respect to the angiographic data and for the identification of mild, nondisabling strokes. Second, because there were relatively few postoperative complications, the use of multivariate analysis was precluded. Such analyses would require much larger data sets than are currently available. Finally, the present analyses were performed on the basis of data collected several years ago. As noted, the overall safety of carotid endarterectomy has likely improved. However, there is no reason to believe there should be a change in the relative impact of individual risk factors.
A European study on the efficacy of carotid endarterectomy continues to enroll patients. A careful analysis of preoperative risk factors from this study compared with those identified through ACAS and through studies such as ours will provide further data to aid the risk stratification of potential candidates for this prophylactic procedure.
Member institutions of the Academic Medical Center Consortium are as follows: Alton Ochsner Medical Foundation, New Orleans, La; Brigham and Women’s Hospital, Boston, Mass; Dartmouth-Hitchcock Medical Center, Lebanon, NH; Duke University Medical Center, Durham, NC; Johns Hopkins Hospital, Baltimore, Md; Massachusetts General Hospital, Boston, Mass; Mayo Clinic Foundation, Rochester, Minn; New England Medical Center Hospitals, Boston, Mass; UCLA Medical Center, Los Angeles, Calif; University of Iowa Hospitals and Clinics, Iowa City, Iowa; University of Pennsylvania Medical Center, Philadelphia, Pa; and University of Rochester Medical Center, Rochester, NY.
This study was supported by the Commonwealth Fund, the John A. Hartford Foundation, American Medical Association, the Academic Medical Center Consortium, the RAND Corporation, and the Agency for Health Care Policy and Research, contract No. 282-91-0028.
- Received November 20, 1997.
- Revision received January 26, 1998.
- Accepted January 26, 1998.
- Copyright © 1998 by American Heart Association
Barnett HJM, Warlow CP. Carotid endarterectomy and the measurement of stenosis. Stroke. 1993;24:1281–1284.
Young B, Moore WS, Robertson JT, Toole JF, Ernst CB, Cohen SN, Broderick JP, Dempsey RJ, Hosking JD. An analysis of perioperative surgical mortality and morbidity in the asymptomatic carotid atherosclerosis study. Stroke. 1996;27:2216–2224.
McCrory DC, Goldstein LB, Samsa GP, Oddone EZ, Landsman PB, Moore WS, Matchar DB. Predicting complications of carotid endarterectomy. Stroke. 1993;24:1285–1291.
Goldstein LB, McCrory DC, Landsman PB, Samsa GP, Ancukiewicz M, Oddone EZ, Matchar DB. Multicenter review of preoperative risk factors for carotid endarterectomy in patients with ipsilateral symptoms. Stroke. 1994;25:1116–1121.
Moore WS, Barnett HJM, Beebe HG, Bernstein EF, Brener BJ, Brott T, Caplan LR, Day A, Goldstone J, Hobson RW II, Kempczinski RF, Matchar DB, Mayberg MR, Nicolaides AN, Norris JW, Ricotta JJ, Robertson JT, Rutherford RB, Thomas D, Toole JF, Trout HH III, Wiebers DO. Guidelines for carotid endarterectomy: a multidisciplinary consensus statement from the Ad Hoc Committee, American Heart Association. Stroke. 1995;26:188–201.
Rothwell PM, Slattery J, Warlow CP. Systematic comparison of the risks of stroke and death due to carotid endarterectomy for symptomatic and asymptomatic stenosis. Stroke. 1996;27:266–269.
Fode NC, Sundt TM Jr, Robertson JT, Peerless SJ, Shields CB. Multicenter retrospective review of results and complications of carotid endarterectomy in 1981. Stroke. 1986;17:370–376.
Dyken ML. Controversies in stroke: past and present: the Willis Lecture. Stroke. 1993;24:1251–1258.
Mora CT, Murkin JM. The central nervous system: responses to cardiopulmonary bypass. In: Mora CT, ed. Cardiopulmonary Bypass: Principles and Techniques of Extracorporeal Circulation. New York, NY: Springer-Verlag; 1995:114–146.