|
|
| |||||||||||||||||||||||||||||||||
Background Transient monocular blindness associated with internal-carotid-artery stenosis is a risk factor for stroke. The effect of carotid endarterectomy in patients who present with transient monocular blindness has not been determined.
Methods We compared the risk of stroke among patients presenting with transient monocular blindness with the risk among patients presenting with hemispheric transient ischemic attack. The effect of endarterectomy was assessed in patients with transient monocular blindness. The analyses were based on data from the North American Symptomatic Carotid Endarterectomy Trial.
Results A total of 198 medically treated patients with transient monocular blindness had a three-year risk of ipsilateral stroke that was approximately half of that among 417 medically treated patients with hemispheric transient ischemic attack (adjusted hazard ratio, 0.53; 95 percent confidence interval, 0.30 to 0.94). Six factors were associated with a higher risk of stroke in patients with monocular blindness — an age of 75 years or more, male sex, a history of hemispheric transient ischemic attack or stroke, a history of intermittent claudication, stenosis of 80 to 94 percent of the luminal diameter, and the absence of collateral circulation. The three-year risk of stroke with medical treatment for patients with zero or one risk factor was 1.8 percent, with two risk factors 12.3 percent, and with three or more risk factors 24.2 percent (P=0.003). The three-year absolute reduction in the risk of stroke associated with endarterectomy was –2.2 percent (i.e., a 2.2 percent increase in risk) among patients with zero or one risk factor, 4.9 percent among those with two risk factors, and 14.3 percent among those with three or more risk factors (P=0.23 by a test for interaction).
Conclusions Among patients with internal-carotid-artery stenosis, the prognosis was better for those presenting with transient monocular blindness than for those presenting with hemispheric transient ischemic attack. Among patients with transient monocular blindness, carotid endarterectomy may be beneficial when other risk factors for stroke are also present.
Two large clinical trials reported that endarterectomy is effective in reducing the risk of stroke in patients with symptomatic internal-carotid-artery stenosis.15,16 The effect of endarterectomy in patients with only transient monocular blindness is unclear, since both trials reported their results in aggregate form by combining patients who presented with transient monocular blindness with patients who presented with hemispheric transient ischemic attacks or nondisabling stroke.
We undertook a study to compare the risk of stroke among patients presenting with transient monocular blindness with the risk among those presenting with hemispheric transient ischemic attack and to assess the effectiveness of carotid endarterectomy in patients with transient monocular blindness. The present study was a predefined research project of the North American Symptomatic Carotid Endarterectomy Trial.
Methods
Eligibility and Randomization
Patients were enrolled in the North American Symptomatic Carotid Endarterectomy Trial if they had had a transient ischemic attack or nondisabling ischemic stroke associated with carotid-artery stenosis within the previous 180 days. Between December 1987 and December 1996, 2885 patients were randomly assigned to receive optimal medical therapy or optimal medical therapy plus endarterectomy and were followed until the end of December 1997. Patients were not eligible if they had a probable cardiac source of embolism or a serious disease likely to cause death within five years. All patients underwent a detailed evaluation at base line, including electrocardiography, chest radiography, imaging of the brain, and conventional carotid angiography. Follow-up consisted of clinical examinations every four months. The protocol was approved by the appropriate institutional review boards, and all patients gave written informed consent.
Clinical Analysis
Clinical details of the episodes of transient monocular blindness and the hemispheric transient ischemic attacks, including their duration and number, were collected as part of the trial protocol. Transient monocular blindness was defined as a partial or complete visual-field loss in one eye that was of ischemic origin and lasted less than 24 hours with complete recovery. A hemispheric transient ischemic attack was defined as a transient, focal cerebral dysfunction of ischemic origin lasting less than 24 hours. The ischemic event that had occurred most recently before randomization was considered the qualifying (presenting) event for the purposes of the trial.
Computed tomographic scans of the brain were evaluated for the presence of infarcts and to rule out other causes of hemispheric symptoms. The degree of internal-carotid-artery stenosis was calculated from the angiogram with the use of strict criteria.17,18 Near-occlusion of the carotid artery was defined by the presence of severe stenosis that was associated with one or more of the following: a narrowed distal internal carotid artery beyond the diseased portion (considered narrowed if it was only slightly wider than, the same width as, or narrower than the external carotid artery or if it was narrower than the contralateral internal carotid artery); evidence of intracranial collateral supply (a visible collateral flow or evidence of the dilution of intracranial vessels by unopacified blood from collateral sources); or slower progress of contrast material from the internal carotid artery into the cranium than into the distal branches of the external carotid artery.
All strokes were evaluated by the outcomes committee to establish the type, cause, territory, and level of disability. Strokes were considered to be disabling if they received a modified Rankin score of 3 or more (on a scale on which 0 indicates normal and independent functioning and 6 indicates death)19 90 days after the onset of symptoms. Further details of the trial methods have been published previously.20
Statistical Analysis
The base-line characteristics of the patients were compared with the use of the chi-square test. The three-year risk of stroke ipsilateral to the symptomatic internal-carotid-artery stenosis was analyzed according to the intention-to-treat principle. The estimates of risk were derived from Kaplan–Meier event-free survival curves and were evaluated for statistical significance by means of a log-rank test. The Kaplan–Meier analyses included all deaths and any strokes (regardless of location) that occurred during the 30-day perioperative period among patients who underwent endarterectomy and during the 32 days after randomization among the patients who were treated medically. A Cox proportional-hazards regression model was used to adjust the analyses of risk for the base-line characteristics of the patients and to test for interactions.
To identify base-line characteristics that could increase the risk of an ipsilateral stroke and influence the effectiveness of endarterectomy, we analyzed the risk factors among the medically treated patients with transient monocular blindness and stenosis of at least 50 percent of the diameter of the internal carotid artery. Patients with stenosis of less than 50 percent of the arterial diameter were not included in the analysis of risk factors because they do not benefit from endarterectomy.15 Univariate Kaplan–Meier analyses were used to identify important risk factors among the base-line characteristics we analyzed, and patients were then grouped into three categories of risk — those with zero or one risk factor (low risk), those with two risk factors (moderate risk), and those with three or more risk factors (high risk). The three-year risk of ipsilateral stroke among the medically treated patients with transient monocular blindness, as determined by the Kaplan–Meier analysis, was compared with the three-year risk among the surgically treated patients with transient monocular blindness in an analysis stratified according to the category of risk. We defined a variable with scores of 0, 1, and 2, corresponding to the three categories of risk, and performed a test of interaction (with 1 df) between the treatment-group assignment and the number of risk factors using a Cox regression model.
Results
A total of 2885 patients were enrolled in the trial. Transient ischemic attack was the qualifying event in 1583 patients (54.9 percent), and the remaining 1302 patients had a nondisabling stroke. A total of 496, or approximately one third, of the qualifying transient ischemic attacks were episodes of transient monocular blindness, and 397 of the patients with transient monocular blindness (80.0 percent) had no history of hemispheric transient ischemic attack or stroke in the vascular territory ipsilateral to the transient monocular blindness.
In the first analysis, we compared two clinically homogeneous groups. One group consisted of the 397 patients who had only ever had transient monocular blindness, and the other consisted of 829 patients who had only ever had hemispheric transient ischemic attack. Patients who had had a previous ipsilateral stroke or who had ever had a transient ischemic attack different from the qualifying event were excluded from the analysis.
The base-line characteristics of the patients are shown in Table 1. The patients with transient monocular blindness were more likely to have smoked in the year before they entered the study, were roughly twice as likely to have stenosis of at least 70 percent of the diameter of the internal carotid artery or near-occlusion of the carotid artery, and were more than three times as likely to have evidence of collateral circulation. The patients with hemispheric transient ischemic attack tended to be older and were more likely to have a history of hypertension and diabetes mellitus and to have evidence of brain infarcts on imaging. Among those with cerebral infarcts on brain imaging, the patients with hemispheric transient ischemic attack were twice as likely as those with transient monocular blindness to have infarcts of 1 cm or larger. All other vascular risk factors, including the presence of irregular or ulcerated plaque in the ipsilateral internal carotid artery and the presence of ipsilateral intracranial artery disease, were evenly distributed between the two groups.
|
|
|
Among the 397 patients who had only ever had transient monocular blindness (including patients in both the medically and surgically treated groups), the median number of episodes of transient monocular blindness that occurred within the 180 days before randomization was 3 (interquartile range, 1 to 7; 5 percent of these patients had 45 or more episodes within that period). The median duration of the qualifying episode of transient monocular blindness was 4 minutes (interquartile range, 1 to 10 minutes; 5 percent had had episodes lasting 60 minutes or more). There were no significant differences according to the degree of stenosis in the number or duration of the episodes of transient monocular blindness. In the 198 medically treated patients who had presented with transient monocular blindness, the three-year risk of ipsilateral stroke was similar regardless of whether the patient had had a single episode of transient monocular blindness or two or more such episodes (risks of 10.4 percent and 8.2 percent, respectively; P=0.44). Similarly, the duration of the episode of transient monocular blindness — less than five minutes or five or more minutes — did not influence the risk of ipsilateral stroke (which was 8.6 percent and 9.3 percent, respectively; P=0.70).
In a subgroup analysis, we examined the relation between the higher prevalence of angiographically visualized collateral vessels in patients with transient monocular blindness and the risk of ipsilateral stroke. Because collateral vessels are seldom present in patients with stenosis of less than 70 percent of the diameter of the internal carotid artery,21 the analysis was restricted to medically treated patients with stenosis of 70 to 94 percent of the diameter or near-occlusion of the internal carotid artery (both patients with transient monocular blindness and patients with hemispheric transient ischemic attacks). The three-year risk of ipsilateral stroke among the 35 patients with transient monocular blindness and evidence of collaterals was lower (2.9 percent) than the risk among the 30 patients with hemispheric transient ischemic attack and evidence of collaterals (16.7 percent; P=0.06). The three-year risk of ipsilateral stroke among the 44 patients with transient monocular blindness and no collaterals was 16.0 percent, as compared with 34.4 percent among the 69 patients with hemispheric transient ischemic attack and no collaterals (P=0.03). A P value of 0.42 was derived from a Cox regression model for the test of the interaction between the type of transient ischemic attack and the presence or absence of collaterals.
To identify the risk factors that were predictive of ipsilateral stroke in patients with transient monocular blindness, we performed an analysis that included all patients with stenosis of at least 50 percent of the diameter of the internal carotid artery, including those with a history of stroke or hemispheric transient ischemic attack. With the use of univariate Kaplan–Meier analyses, all the variables listed in Table 1 were examined. Six of these variables were strongly associated with the occurrence of stroke in medically treated patients, each more than doubling the risk (Table 2). Patients with stenosis of 70 to 79 percent of the diameter of the internal carotid artery were grouped together with those who had stenosis of 50 to 69 percent of the diameter or near-occlusion of the artery, since the three-year risk of ipsilateral stroke was similar in these three subgroups (10.1 percent, 9.8 percent, and 9.1 percent, respectively). According to the Kaplan–Meier analysis, the risk of ipsilateral stroke among medically treated patients classified as at low risk, moderate risk, and high risk was 1.8 percent, 12.3 percent, and 24.2 percent, respectively (P=0.003 for the comparison of all three groups) (Table 3). The absolute reduction in the three-year risk of stroke associated with endarterectomy was –2.2 percent in the low-risk group (indicating a 2.2 percent increase in risk), 4.9 percent in the moderate-risk group, and 14.3 percent in the high-risk group (P=0.23 for the test of interaction between the treatment group and the category of risk) (Table 3).
|
|
Attacks of transient monocular blindness represent a common manifestation of internal-carotid-artery stenosis and are warnings of impending stroke. They were the presenting symptoms in 20 percent of the patients who underwent randomization in three large trials of carotid endarterectomy.15,16,22
The results of the present study were based on data from a large sample of patients with transient monocular blindness associated with carotid-artery stenosis, consisting of a total of 1057 person-years of prospective follow-up. The three-year risk of ipsilateral stroke was lower and the strokes less disabling among patients who presented with transient monocular blindness than among those who presented with hemispheric attacks. The perioperative rate of stroke and death was also lower among patients with monocular blindness. The frequency and duration of monocular ischemic attacks bore no relation to the risk of stroke.
Despite their having, on average, a higher degree of stenosis (Table 1), patients with transient monocular blindness had, paradoxically, a lower risk of hemispheric stroke, and the strokes that did occur were less disabling. One possible explanation is that there was a higher prevalence of collateral circulation in these patients, which has been shown to be associated with a better prognosis.21 Good collateral circulation was identified in almost one quarter of the patients with transient monocular blindness but in less than 7 percent of the patients with hemispheric transient ischemic attack. Another explanation for the difference in prognosis between patients with transient monocular blindness and those with hemispheric transient ischemic attack is that each small group of cells in the retina is more sensitive than a group of cells of similar size in the brain. This means that a small platelet–fibrin embolus will more readily become clinically manifest in the retina than it would in the brain and that transient monocular blindness may frequently result from small emboli that would be less likely to cause brain infarction. The patients with transient monocular blindness were more likely to have retinal strokes than were those who presented with hemispheric transient ischemic attacks, probably as a consequence of the phenomenon of intraluminal streaming.23,24
Approximately 30 percent of the patients with transient monocular blindness associated with stenosis of at least 50 percent of the diameter of the internal carotid artery were in the high-risk group (had three or more risk factors). Nearly half of these patients had stenosis of 80 to 94 percent of the diameter of the internal carotid artery. In the high-risk group, only seven patients would have to undergo endarterectomy in order to prevent one additional stroke over a three-year period. In the moderate-risk group (with two risk factors), which included approximately 40 percent of the patients, there was a moderate three-year risk of ipsilateral stroke in medically treated patients, and endarterectomy appeared to be less effective (number needed to treat, 20). The remaining 30 percent of the patients had zero or one risk factor; endarterectomy was ineffective in these patients. Although the test of interaction did not find a statistically significant association, there was a clinically important relation between the number of risk factors and the effectiveness of endarterectomy in patients with transient monocular blindness.
We conclude that, as compared with hemispheric transient ischemic attack, transient monocular blindness associated with carotid-artery stenosis carries a better prognosis with respect to subsequent stroke. Among patients with monocular blindness, carotid endarterectomy may improve the outcome in those with other risk factors for stroke.
Supported by a grant (R01-NS-24456) from the National Institute of Neurological Disorders and Stroke.
We are indebted to all the participants in the North American Symptomatic Carotid Endarterectomy Trial and to SmithKline Beecham for providing Ecotrin.
Source Information
From the Division of Neurology, University of Texas at San Antonio, San Antonio (O.B.); the John P. Robarts Research Institute, London, Ont., Canada (M.E., H.J.M.B., H.M.); the Departments of Epidemiology and Biostatistics (M.E.), Clinical Neurological Sciences (M.E., A.J.F., H.J.M.B.), and Diagnostic Radiology (A.J.F.), University of Western Ontario, London, Ont., Canada; and the Rabin Medical Center, Petach Tikva, Israel (J.Y.S.).
Address reprint requests to Dr. Benavente at the Department of Medicine, Division of Neurology, University of Texas Health Science Center, Mail Code 7883, 7703 Floyd Curl Dr., San Antonio, TX 78229-3900, or at benavente{at}uthscsa.edu.
References
| |||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2010 Massachusetts Medical Society. All rights reserved. |
Previous
