endarterectomy after

Transcription

1 Benefit of carotid prior stroke endarterectomy after Raymond G. Makhoul, MD,* Wesley S. Moore, MD, Michael D. Colburn, MD, William J. Quifiones-Baldrich, MD, and Candace L. Vescera, RN, Los Angeles, Calif. Purpose: The benefit of carotid endarterectomy (CEA) in preventing recurrent stroke in patients who have sustained a previous stroke remains controversial. The purpose of this study was to evaluate the immediate results and long-term benefit of CEA after recovery from a prior ipsilateral stroke. Methods: Between 1980 and 1990, 85 patients underwent CEA for prior stroke with an average follow up of 54 months (range 0.3 to 130). The interval from prior stroke to CEA averaged 19.8 months (range 0.1 to 158.3). Vascular risk factors included hypertension in 70.6%, diabetes in 20.0%, history of smoking in 80.0%, and associated coronary artery disease in 51.8% of the patients. Results: There were no perioperative deaths. Four patients (4.7%) had an ipsilateral stroke within 30 days of operation. During the follow up 31 patients (36.5%) died. The leading cause of death was cardiac (54.8%) followed by cancer (16.1%). By life-table methods, the cumulative incidence of recurrent stroke at 9 years was 14% for strokes in the ipsilateral distribution, for an annual stroke risk of 1.6% per year. The cumulative incidence of stroke in other distributions was 5%, for an annual stroke risk of 2.1% per year for all strokes. When interval to operation, preoperative stroke severity, vascular risk factors, and neurologic symptoms were evaluated, no independent indicator of increased risk of recurrent postoperative stroke could be identified. Conclusion: These results demonstrate a marked improvement over the natural history and best medical therapy for these lesions as reported in the literature. We conclude that CEA is beneficial in preventing recurrent stroke in this group of patients and should be considered the appropriate management in this setting. (J VASC SURG 1993; 18: ) Vascular surgeons have traditionally considered patients with established stroke to be a higher-risk population for carotid endarterectomy. Early experience with emergency endarterectomy for acute stroke generally produced poor outcomes. 1'2 Even when operation is delayed until full neurologic recovery is achieved, the morbidity and mortality of carotid endarterectomy are believed to be higher after stroke than in those patients with lesser degrees of neuro- From the Department of Surgery, Vascular Surgery Section, UCLA School of Medicine. Supported in part by a grant from the JOASH Foundation. Presented at the Eighth Annual Meeting of the Western Vascular Society, Sonoma, Calif., lan. 9-12, Reprint requests: Wesley S. Moore, MD, , CHS, UCLA School of Medicine, Los Angeles, CA *Dr. Makhoul is currently at the Department of Surgery, Division of Vascular Surgery, Medical College of Virginia, Richmond, Va. Copyright 1993 by The Society for Vascular Surgery and International Society for Cardiovascular Surgery, North American Chapter /93/$ /6/48587 logic injury. Although the North American Symptomatic Carotid Endarterectomy Trial unequivocally demonstrated the benefit of carotid endarterectomy in patients with transient ischemic attack, amaurosis fugax, and minor stroke, no outcome data were reported for this important subgroup of patients with established stroke. In addition, the North American Symptomatic Carotid Endarterectomy Trial was limited to patients with recent neurologic events (less than 4 months). 3 To determine the immediate and long-term benefits of carotid endarterectomy in patients with a documented previous stroke, we have reviewed a I0-year clinical experience of carotid endarterectomy for stroke. METHODS Between January 1980 and June 1990, 85 patients with a previous history of cerebral infarction in the ipsilateral carotid distribution underwent elective carotid endarterectomy at UCLA Medical Center. Only patients undergoing elective carotid endarterectomy were included in this series. Patients having 666

2 JOURNAL OF VASCULAR SURGERY Volume 18, Number 4 Makhoul et al. 667 Table I. Neurologic event severity scale ~ Neurologic Neurologic Severity grade Impairment~ symptoms signs 0 None Absent Absent 1 None Present Absent 2 None Absent Present 3 None Present Present 4 Minor, in one or more domains Present Present 5 Major, in only one domain N/A1: N/A 6 Major, in any two domains N/A N/A 7 Major, in any three domains N/A N/A 8 Major, in any four domains N/A N/A 9 Major, in all five domains N/A N/A 10 Reduced level of consciousness N/A N/A NA, Not applicable. ~Adapted from Stroke 1985;16: ?Impairment in the domains of swallowing, self-care, ambulation, communications, and comprehension. If independence is maintained despite the impairment, it is classified as minor; if independence is lost, it is classified as major. SNeurologic signs and symptoms are integrated into the higher grades of impairment. an emergency procedure, simultaneous endarterectomy combined with coronary artery bypass grafting, nonatherosclerotic carotid disease, recurrent carotid disease, or operations such as carotid-subclavian artery bypass or carotid aneurysm repair were excluded. The patients' medical records were reviewed for details regarding risk factors; location, severity, and time of stroke; angiographic and noninvasive studies; preoperative neurologic status; details of the operation; and postoperative course. Long-term follow-up data were obtained from outpatient clinic records or telephone contact with a referring physician, the patient, or the patient's family. The preoperative, postoperative, and late neurologic status of each patient was classified on a neurologic event severity scale (Table I). The event rates and survival rates were analyzed by life-table analysis. 4 The univariate and joint relationships between various potential risk factors and incidence of stroke were evaluated by logistic regression analysis. RESULTS There were 85 carotid endarterectomies performed on 85 patients who had a previous stroke in the ipsilateral carotid artery distribution. The mean age was 68 years (range 35 to 86 years). There were 50 men and 35 women. Seventy-one percent of patients had a history of hypertension, 52% had significant coronary artery disease, 20% had diabetes mellitus, and 80% gave a history of cigarette smoking. The mean neurologic event severity scale score was 2.3 (range 0.0 to 5.0). Twenty-two patients (26%) who had had a previous ipsilateral stroke were having transient ischemic attacks before operation. Table II. Preoperative strokes Patient Cause Neurologic event severity scale (preop/postop/long-term) A Embofic* 4/4/1 B Thrombosis? 4/4/4 C Embolic* 4/4/4 D Thrombosis? 3/4/1 *Patent by duplex scan, therefore presumed embolic.?thrombosis at time of reexploration. The degree of stenosis of the operated artery was greater than 70% in 56 patients (66%) and between 30% and 70% in 20 cases (23%). Nine patients (11%) had less than 30% stenosis but had significant ulcerated plaques on angiography. The mean time interval from stroke to operation was 19.8 months (range 0.1 to 158 months). Twenty-eight (33%) patients underwent carotid endarterectomy within 6 weeks of the stroke; the remaining 57 (67%) had operation after 6 weeks. An indwelling shunt was used in 62 (73%) of cases. The carotid artery was closed primarily in 77 (91%) of patients, with patch closure used in the remaining 9%. There were no perioperative deaths. Four patients had a stroke in the ipsilateral carotid artery distribution within 30 days of operation, for a perioperative stroke rate of 4.7%. The details of these events are shown in Table II. One (3.6%) of 28 patients who underwent operation within 6 weeks of the initial stroke had a perioperative stroke. Three (5.2%) of 57 patients who received operation at a time more than 6 weeks after the initial stroke

3 JOURNAL OF VASCULAR SURGERY 668 _Makhoul et al. October 1993 A g ~oo a m L. 14 > '.~ 60 ~a 60 o~ E,-,1 o 20 2O 0 I r i I I I I I I i I Fig. 1. Cumulative survival ectomy. Months after carotid endarter- 0 i 0 I I I I I I I I i I I I I I I I I Months Fig. 3. Cumulative stroke-free rate after carotid endarterectomy for strokes in all distributions. == I.I O L ] 0 I I I I I I I I I ~ 1 ~ I I I I I i Months Fig. 2. Cumulative ipsilateral stroke-free rate after carotid endarterectomy. had a perioperative stroke. One patient had amaurosis fugax on the first postoperative day and was returned to the operating room and found to have a thrombosed external carotid artery. This was opened, and the patient had no postoperative sequelae. Three patients (3.5%) had cranial nerve palsies after carotid endarterectomy. One patient had a seventh nerve paresis, which was permanent; one a temporary twelfth nerve paresis, which resolved after 6 weeks; and the third had injury to the glossopharyngeal nerve and had dysphagia, which requires a gastrostomy tube 2 years after operation. There was an 8.5% incidence of associated postoperative morbidity including four neck hematomas that required operative drainage, one intraoperative cardiac arrest, and two cases of pulmonary complications that resolved. At the time of this review, follow-up information was complete in 73 patients (86%). The mean follow-up interval in this group of patients was 53.9 months (range 0.3 to months). During the follow-up interval 31 (36.5%) patients are known to have died. Cardiac disease was the most common cause of death (55%), followed by cancer (16%), stroke (6%), and pulmonary insufficiency (6%); 16% of patients died of a variety of other causes. The overall survival rate calculated by the life-table method was 77% at 3 years, 71% at 5 years, and 54% at 10 years (Fig. 1). Thus the average mortality rate was 4.6% per year. Six patients had a late stroke in the vascular distribution of the operated carotid artery. These strokes occurred at 2, 3, 7, 24, 36, and 42 months after operation. Two of these patients died as a result of the stroke. The occurrence of late strokes in the ipsilateral distribution is analyzed by life tables in Fig. 2. At 9 years the cumulative stroke-free rate was 86%. Therefore the annualized ipsilateral stroke risk was 1.6% per year. There were an additional three strokes that occurred in other distributions during the period of follow up, which yields an annual stroke risk of 2.1% per year for all strokes (Fig. 3). When the variables interval to operation, preoperative stroke

4 JOURNAL OF VASCULAR SURGERY Volume 18, Number 4 3/iakhoul et al. 669 severity, vascular risk factors, and neurologic symptoms were evaluated, no independent predictor of increased risk of recurrent postoperative stroke could be identified (2 > 0.05). DISCUSSION It is estimated that one half of strokes may be related to disease at the carotid bifurcation and thus be potentially amenable to surgical intervention by carotid endarterectomy, s Patients with carotid disease who have had a minor stroke and are left with minimal or no deficit are at risk for further cerebral infarction. Studies of the natural history of such patients show that their risk of recurrent stroke is 5% to 20% per year, with an average recurrence of 50% at 5 years. 6"9 It should be noted that these are old studies that predate the diagnostic techniques currently available and relate to the incidence of strokes in all distributions. As such, they may not exactly mirror the group of patients described in this and other surgical series. Medical management has included the administration of antiplatelet agents or oral anticoagulants (warfarin). But a prospective, randomized trial of aspirin versus placebo in patients with stroke and carotid disease failed to demonstrate any reduction in subsequent strokes with aspirin. At 2 years the recurrence rates for the aspirin and placebo groups were 12% and 13%, respectively) More recently, ticlopidine, a new antiplatelet agent, has been evaluated for the prevention of recurrent stroke in patients with a completed stroke. These results showed that ticlopidine was somewhat more effective than aspirin for reducing the risk of recurrent stroke, but there was still a 5% incidence of stroke in all distributions at 1 year. H Operation for the treatment of carotid disease after stroke has been controversial, both with regard to timing of operation and its benefit in preventing recurrent stroke. The reported outcome of emergency endarterectomy for acute stroke was generally poor with significant morbidity and mortality. 1'2 Often, this was because of the conversion of a bland to a hemorrhagic infarct. On the basis of these results, an arbitrary delay of 6 weeks before the performance of carotid endarterectomy was recommended. Subsequent studies by Whittemore and Mannick 12 and Piotrowski et al)3 have demonstrated that this delay is not required to prevent perioperative neurologic complications and that operation can be safely performed when neurologic recovery has reached a plateau. In these and other series, the perioperative stroke rate ranges from 0% to 8%, with a perioper- ative death rate of 0% to 6%) 1"16 The results of the present series compare favorably with these results, with a combined perioperative mortality and stroke rate of 4.7%. In addition, no difference was found in the incidence of perioperative stroke between patients operated on early (less than 6 weeks after stroke) versus late (more than 6 weeks). However, these results must be viewed in the context of the selection process to determine the timing of intervention. It has been the practice at UCLA to follow the clinical course of the stroke patient, and once the patient has achieved maximum recovery from the initial stroke, an assessment is made as to the amount of cerebral function that is left to lose. If the patient has considerable function left at risk, work-up leading to possible carotid endarterectomy is undertaken and operation is done without delay when appropriate carotid pathologic conditions are uncovered. Whether endarterectomy is beneficial in preventing recurrent stroke has been retrospectively studied by others. Bardin et al) 6 reviewed their results from 127 carotid endarterectomies in 107 patients with completed stroke. The combined perioperative morbidity and mortality rate was 7%, with a cumulative 5-year late stroke incidence of 20%. They concluded that operation for the prevention of recurrent stroke carries a high risk and that it may not be superior to a nonoperative approach. Subsequent reviews have demonstrated a beneficial effect of operation, with recurrent stroke rates ranging from 0% to 3% per year) 7z With the use of life-table methods, our series demonstrated a cumulative incidence of recurrent stroke in the ipsilateral distribution at 9 years of 14%, which yields an annual stroke risk of 1.6% per year. For all strokes, the cumulative incidence was 19%, for an annual stroke risk of 2.1% per year. These results represent a marked improvement over the natural history of the disease as reported in the literature. In addition, these are superior to reported results with use of the best medical management. Thus our results, with a mean follow up of 4.5 years, strongly favor the efficacy of endarterectomy in the prevention of recurrent stroke. One concern about carotid endarterectomy has been that patients treated in this fashion might have a shortened life expectancy, thus negating the benefits of prophylaxis from recurrent stroke. In this report, there were no perioperative deaths or myocardial infarctions, and although the most common cause of late death was cardiac, the patient survival rate was

5 670 A/Iakhoul et al. JOURNAL OF VASCULAR SURGERY October % at 5 years. Therefore it seems that these patients do survive, further justifying an aggressive approach to prevent recurrent stroke. On the basis of these results we conclude that carotid endarterectomy after prior stroke can be performed safely and effectively. In addition, operation affords significant protection from recurrent ipsilateral stroke in the long term and should be considered the treatment of choice. REFERENCES 1. Wylie EJ, Hein MF, Adams JE. Intracranial hemorrhage following surgical revascularization for treatment of acute strokes. J Neurosurg 1964;21: Rob CG. Operation for acute completed stroke due to thrombosis of the internal carotid artery. Surgery 1969;65: North American Symptomatic Carotid Endarterectomy Trial Collaborators. Beneficial effect of carotid endarterectomy in symptomatic patients with high grade stenosis. N Engl J Med 1991;325: Baker JD, Rutherford RB, Bernstein EF, et al. Suggested standards for reports dealing with cerebrovascular disease. J VASC SURG 1988;8: Mohr JP, Caplan LR, Melski JW, et al. The Harvard Cooperative Stroke Registry: a prospective registry. Neurology 1978;28: Robinson RW, Dimirel M, LeBeau RJ. Natural history of cerebral thrombosis: 9-19 year follow-up. J Chron Dis 1968;21: Enger E, Boyesen S. Long-term anticoagulant therapy in patients with cerebral infarction: a controlled clinical study. Acta Med Scand 1965 ;178: Baker RN, Schwart WS, Ranseyer JC. Prognosis among survivors of ischemic stroke. Neurology 1968; 18: Sacco RL, Wolf BA, Kannel WB, McNamara BM. Survival and recurrence following stroke: the Framingham study. Stroke 1982;13: Britton M, Helmers C, Samuelson K, et al. High-dose acetylsalicylic acid after cerebral infarction-a Swedish cooperative study. Stroke 1987;18: Harbison JW. Ticlopidine versus aspirin for the prevention of recurrent stroke. Stroke 1992;23: Whittemore AD, Mannick JA. Surgical treatment of carotid disease in patients with neurologic deficits. J VASC SFRG 1987;5: Piotrowski JJ, Bernhard VM, Rubin JR, et al. Timing of carotid endarterectomy after acute stroke. J VASC SURG 1990;11: Giordano JM, Trout HH III, Kozloff L, DePalma RG. Timing of carotid endarterectomy after stroke. ~ VAsc SURG 1985;2: Takolander RJ, Bergentz SE, Ericsson BF. Carotid artery surgery in patients with minor stroke. Br J Surg 1983;70: Bardin JA, Bernstein EF, Humber PB, et al. Is carotid endarterectomy beneficial in prevention of recurrent stroke? Arch Surg 1982;117: Rubin JR, Goldstone J, Mclntyre KE, Malone JM, Bernhard VM. The value of carotid endarterectomy in reducing the morbidity and mortality of recurrent stroke. J VASC SURG 1986;4: McCullough JL, Mentzer RM, Harman PK, Kaiser DL, Kron IL, Crosby IK. Carotid endarterectomy after a completed stroke: reduction in long-term neurologic deterioration. J VASC SURG 1985;2: Rosenthal D, Borreno E, Clark MD, Lamis PA, Daniel WW. Carotid endarterectomy after reversible ischemic neurologic deficit or stroke: Is it of value? J VASC SURG 1988;8: Hertzer NA, Arison R. Cumulative stroke and survival ten years after carotid endarterectomy. J VAsc SURG 1985;2: Submitted Feb. 23, 1993; accepted May 7, DISCUSSION Dr. Eugene F. Bernstein (La Jolla, Calif.). We should be grateful to Dr. Makhoul and his associates for providing us with additional documentation regarding the relative safety of carotid surgery in the post-stroke patient, particularly when the selection process has been careful, and the surgical group is skillful. Certainly, the data from NASCET, the VA and ECST have provided sound and convincing data regarding the superiority of surgical therapy for patients with transient symptoms associated with high-grade carotid artery stenoses. However, no such data exist regarding the benefits of surgery after a stroke. Did all the patients have severe degrees of carotid artery stenosis? Many surgeons believe that routine shunting is appropriate for carotid artery surgery in patients who had had a stroke. Because shunts were not used in all of these operations, was the operative stroke rate different in those patients who were not shunted? What was the relationship of operative and late morbidity and mortality rates to the severity of preoperative neurologic deficit? Are symptoms and signs the best criteria for the selection of such patients for carotid artery surgery? What about the infarct size data from CT/MRI scanning, and the distribution and extent of the cerebrovascular arterial disease as factors in predicting the likelihood of operative or late postoperative stroke or death? What is the importance of contralateral disease in this equation? The NASCET data show no effect of timing of carotid endarterectomy on morbidity or mortality rates after stroke. Have you reviewed this question, and do you have a recommendation regarding the appropriate interval to wait after a patient with a stroke has become neurologically stable? This presentation highlights the need to perform a more careful analysis of the important factors that may

6 IOURNAL OF VASCULAR SURGERY Volume 18, Number 4 Makhoul et al. 671 contribute to the decision for or against carotid artery surgery. Dr. Raymond Makhoul. Most of these patients had a greater than 70% stenosis. Some patients underwent operation for lesser degrees of stenosis or for ulceration. In terms of shunting, shunts were not used in all of these patients. There was no difference between the shunted and unshunted groups with respect to perioperative stroke rate. It is the practice of the senior author to shunt all patients with previous stroke. However, some of the other surgeons used selective shunting. With regard to the relationship of operative and late morbidity rates to the severity of the preoperative neurologic deficit, we studied the Neurologic Event Severity Scale score as an indicator of recurrent stroke and found it not to be a significant indicator of that. We did not examine it in relationship to survival rates. With regard to the timing of surgery after stroke, there was no difference when we stratified patients undergoing surgery within 6 weeks and greater than 6 weeks after the initial stroke. It is our practice to perform surgery at the plateau of the neurologic recovery after the initial stroke. Dr. Victor Bernhard (Tucson, Ariz.). We have reported on 129 patients who underwent operation after stroke, most of whom underwent operation within 6 weeks of the ischemic episode after neurologic recovery had reached a plateau. Similar to our experience, the operative deficit rate in this report is not significantly different from that in patients who have not had a previous stroke. This leads me to conclude that perioperative stroke-related problems are primarily related to operative technique. They are most likely thromboembolic phenomena and probably not a result of some unexplained pathophysiologic intracerebral process related to the previous stroke. In review of our concurrent experiences, what do you think is the most likely cause of deficits in patients undergoing carotid endarterectomy after recovery from a stroke! Dr. Makhoul. In this series of patients, there were four perioperafive events. Two of these were due to thrombosis of the vessel and two were likely thromboembolic. Therefore I agree with your comments. Dr. Hugh Beebe (Toledo, Ohio). Patients that I've seen have had a stroke fairly recently, and I'm not sure how much to apply these data to that problem. A mean of 19 months since the time of stroke when you perform a carotid endarterectomy indicates to me that there is a selection process. The patients who eventually found their way to you a year and a half after their stroke were a selected set of pretty good patients. Can we apply your data to the patient who had a stroke last week? Dr. Makhoul. I certainly agree that there's a spectrum in terms of the timing of surgery. However, a third of our patients did have surgery within six weeks of their stroke. Dr. Ka i ]ohansen (Seattle, Wash.). Although the leading cause of death in patients with cerebrovascular disease is coronary artery disease, it's critical for us to recall that, for at least 5 years after a stroke, the leading cause of death in a stroke victim is another stroke. In a series of 58 patients who underwent operation with prior ipsilateral stroke, contralateral occlusion, and other high-risk criteria with local or regional anesthesia, the use of a shunt was reduced to 4.5 %, and the perioperative stroke plus death rate to 1.9%. Would you comment on the use of "awake" anesthesia, which eliminates the requirement for any other means of monitoring the adequacy of cerebral peffusion? Dr. Makhoul. I can't comment with regard to this series because all of these patients underwent operation receiving general anesthetic.