Effect of Time and Cerebrovascular Symptoms on the Prevalence of Microembolic Signals in Patients With Cervical Carotid Stenosis
Background and Purpose High-intensity transient signals (HITS) detected by transcranial Doppler ultrasonography correspond to microemboli in intracranial arteries. The aim of this study was to determine the time course of cerebral microembolism in patients with symptomatic internal carotid artery stenosis and to assess its relation to specific symptoms of cerebral ischemia.
Methods On the basis of criteria established a priori, 69 middle cerebral arteries were selected from a series of consecutive studies obtained at our neurovascular laboratory. All patients had radiological evidence of cervical internal carotid artery disease and had corresponding symptoms. A TC-2000 instrument equipped with special software for microembolus detection was used. Accepted signals were unidirectional from baseline, had a chirping sound, were 9 dB higher than the surrounding blood, and lasted 25 milliseconds or more.
Results HITS were identified in 20 of 69 (29%) arteries. The median interval between onset of symptoms and time of testing was 4 days for HITS-positive arteries and 12 days for those that were HITS negative (P=.0046). Fourteen of 32 (44%) arteries with transient ischemic attacks and 6 of 37 (16%) arteries with cerebral infarction were HITS positive (P=.012).
Conclusions In patients with symptomatic carotid stenosis, HITS are detected more frequently when patients are tested soon after symptoms of cerebral ischemia. HITS are also more prevalent in the territories of arteries with transient ischemic attacks rather than cerebral infarction. These findings may have diagnostic and therapeutic implications.
Although cardiac and artery-to-artery embolisms are recognized today as common mechanisms of cerebral infarction, the natural history of cerebral embolization remains relatively unknown. Several recent large studies have provided data regarding the course of clinical events in patients with atrial fibrillation1 2 and cervical ICA stenosis3 who are at risk for embolic events. However, these studies have not permitted the assessment of embolism in vivo. Data regarding the duration, rate, characteristics of emboli, and their relationship with the type of cerebral ischemic symptoms are limited.4 Information of this nature may be useful to determine the risk of first or recurrent stroke and to direct therapeutic options.
HITS detected during TCD testing correspond to microemboli coursing through cerebral arteries. HITS, also called “microembolic signals” by a consensus committee,5 are associated with symptoms of cerebral ischemia.6 7 8 Laboratory studies show that they correspond to particles composed of platelets, fibrinogen, fat, or cholesterol,9 10 findings that are supported by the examination of surgical specimens obtained at carotid endarterectomy.11
The ability to detect HITS provides an opportunity to examine the course of cerebral embolism in vivo under various conditions. The purpose of this study was to assess, in patients with symptomatic cervical ICA stenosis, the effect of time after symptoms of cerebral ischemia on the characteristics of cerebral microembolism and the relationship of the latter with the type of symptoms.
Subjects and Methods
The records of the Neurovascular Laboratory at the Boston Veterans Administration Medical Center were reviewed for consecutive patients who had undergone TCD microembolus detection studies as part of their evaluation between March 1993 and February 1995. One hundred ninety-three MCAs from 146 patients were studied. The following inclusionary and exclusionary criteria were selected a priori in an attempt to ensure the homogeneity of the studied group and to limit the sample to patients with presumed atherosclerotic ICA stenosis.
Selected patients who met the inclusionary criteria for symptomatic ICA stenosis presented with a diagnosis of either TIA, including transient hemispheric attacks and transient monocular blindness, or cerebral infarction. Exclusionary criteria consisted of the following diagnoses: (1) asymptomatic ICA stenosis (n=57); (2) potential cardiac source for embolism such as atrial fibrillation (n=19), prosthetic heart valve (n=17), akinetic ventricular wall with or without intraventricular thrombus (n=10), or aortic valve stenosis and insufficiency (n=2); (3) aortic arch atherosclerotic plaque (n=1); (4) intracranial stenosis detected by conventional cerebral or MR angiography (n=10); and (5) known clotting abnormality (n=1). Seven patients were excluded because of incomplete data. Several patients satisfied more than one of the exclusionary criteria. It is recognized that these selection criteria and the clinical diagnosis of symptomatic ICA stenosis have their inherent limitations.
Of the original 193 MCAs insonated during the study period, 69 arteries in 66 patients met the inclusionary and exclusionary criteria. These 66 men had a mean age of 67 years (range, 37 to 82 years). Of the 69 selected MCAs, 37 supplied hemispheres with cerebral infarction and 32 supplied territories with symptoms of TIA.
All patients were examined by one of us, and conditions that may mimic symptoms of cerebral or ocular ischemia were carefully excluded. The diagnoses of TIA and cerebral infarction were based on conventional clinical criteria and were supported by the laboratory investigations and brain CT (n=42) or MR imaging (n=44) studies. Thirty arteries were insonated while patients were receiving heparin, 23 during aspirin treatment, 5 during warfarin treatment, 1 during ticlopidine treatment, and 8 while patients were receiving no antiplatelet or anticoagulant therapy. One artery was studied while the patient was receiving both heparin and warfarin and another while the patient was taking heparin and aspirin. Analyses regarding the potential influence of medications on HITS occurrence were performed after exclusion of subjects receiving more than one of the preceding medications and of those receiving no antiplatelet or anticoagulant therapy.
TCD studies were performed according to previously described methods.12 A TC-2000 instrument (Eden Medical Electronics/Nicolet) equipped with a 2-MHz probe was used for all studies, and embolus detection was performed with specially designed software (5.3002 E Beta and 5.40 E Beta 01A). The recording time was 30 minutes. In patients who had been monitored more than once, only the first study results were considered in our analyses to ensure uniformity of data acquisition.
A headband was used to immobilize the probe against the temporal bone window at the beginning of each test. An experienced technician performed all studies, monitoring the patient being evaluated and the instrument screen. The technician was asked to note all events, such as patient blinking and probe slippage, that could potentially cause recording artifacts. In addition to handwritten log sheets describing the specifics of each study, signals were saved on floppy disk by either the technician or the automatic embolus-detection program that was preset to identify signals with specific features. With the exception of one case, floppy disks were reviewed as part of this study, and microemboli counts as well as measurements of signal amplitude and duration were performed. In the case with a corrupted disk, emboli characteristics originally recorded on the log sheets were used.
Accepted microembolic signals were unidirectional from baseline, lasted 25 milliseconds or more, had intensities of 9 dB or higher than the surrounding blood, and were associated with a “chirping” sound on the audio output. These criteria are more strict than the ones proposed by the Consensus Committee of the Ninth International Cerebral Hemodynamics Symposium5 and presumably led to selection of larger emboli. On the basis of their ultrasonic characteristics, we also divided microemboli into “large” particles, ie, those more than 12 dB in intensity and 50 milliseconds in duration, and “small” particles with intensities and durations of 9 to 12 dB and 25 to 50 milliseconds, respectively.
Vascular Imaging Studies
The severity of cervical ICA stenosis was assessed on the basis of conventional cerebral angiography in 38 of 69 (55.1%) cases, by color duplex imaging in 29 of 69 (42.0%) arteries, and by MR angiography in 2 of 69 (2.9%).
Duplex studies were performed on an Ultramark 9-HDI instrument (Advanced Technology Laboratories). They were performed by one of three technicians and interpreted by the same radiologist. MR imaging was performed on a 1.5-T General Electric Signa unit. All radiographic films were reviewed by us. The only exception was the MR angiography study of one patient, which could not be located; the original report of the neuroradiologist was used in that case.
Group comparisons were made with the χ2 test, Fisher’s exact test (two-tailed), Wilcoxon’s rank-sum test, and Spearman’s rank-order correlation. In addition, multivariate analysis was performed using logistic regression. All statistical analyses were performed at the Data Coordinating Center of the Boston University School of Public Health.
HITS were detected in 20 of 69 (29%) arteries. The Figure⇓ presents the temporal distribution of the interval between onset of symptoms and TCD testing. The median interval between onset of neurological symptoms and time of TCD testing was 4 days (range, 0.16 to 45 days) for arteries that were HITS positive and 12 days (range, 1 to 1080 days) for those that were HITS negative. The difference between the two groups was significant (P=.0046).
In HITS-positive studies, the median number of HITS during the 30-minute recording time was 2.5 (range, 1 to 20). There was no significant association between the HITS rate and the interval between onset of symptoms and time of TCD testing. Large microemboli were detected in 11 arteries and small ones in 9. The median interval was 2 days (range, 0.16 to 11 days) for large HITS and 4 days (range, 2 to 45 days) for small ones. The difference between the two groups did not reach statistical significance (P=.127).
The effect of the qualifying event was studied by comparing patients with TIAs to those with completed cerebral infarcts. Among the 32 subjects in the first group, 14 (44%) were HITS positive, whereas 6 of the 37 patients (16%) with cerebral infarction were HITS positive. The difference between the two groups was significant (P=.012). Multivariate logistic regression analysis showed that the associations of HITS with TIAs and with shorter time intervals after onset of symptoms were independent of each other.
Nineteen (45%) of the HITS-negative and 11 (65%) of the HITS-positive arteries were in patients receiving heparin at the time of TCD testing. The difference between the two groups did not reach statistical significance. There was also no significant difference between the two groups (18 [43%] HITS-negative cases, 5 [29%] HITS-positive cases) with regard to aspirin intake.
In this study, we found that microembolic signals are more frequently detected when patients are evaluated soon after symptoms of cerebral ischemia. We also show that these signals are more frequent in the territories of arteries with transient hemispheric or retinal ischemia than those with cerebral infarction.
The clustering of clinically detected recurrent cerebral and systemic embolism close to an initial embolic event has previously been recognized. In patients with acute myocardial infarction, 86% of recurrent emboli occur within 12 weeks.13 In individuals with acute cardioembolic cerebral infarction, recurrent cerebral embolism occurs in 10% to 20% during the first 2 weeks after the onset of original symptoms14 15 16 ; the majority of these events occur within the first 7 days of that period.17 Recent investigations have also studied the recurrence rate in relation to the etiology of cardioembolism. The rate is higher in patients with rheumatic heart disease and intracardiac thrombi and slightly lower in those with atrial fibrillation.15 However, even in the latter group, stroke recurrence also seems clustered during the early months.17
In patients with asymptomatic carotid stenosis treated medically, the 2.3% annual risk of ipsilateral stroke is distributed throughout the follow-up period.18 In symptomatic severe cervical carotid stenosis, the 2-year risk of subsequent ipsilateral stroke is 17% if the original event is a retinal TIA and 44% if the qualifying event is a hemispheric TIA.3 Moreover, in both groups, approximately 50% of recurrent episodes happen during the 2-month period following the original event.3 Because the major mechanisms of cerebral ischemia in acute carotid disease are embolism and impaired arterial perfusion,19 with embolism being the more common of the two,19 this difference between asymptomatic and symptomatic stenoses identifies the latter as particularly malignant lesions that act as sources of emboli during the weeks and months following initial symptoms of ipsilateral retinal or cerebral ischemia. Our finding of a clustering of HITS-positive studies close to symptoms of cerebral or retinal ischemia constitutes in vivo evidence that supports the preceding consideration. This finding suggests that microembolism is a frequent phenomenon during the days that follow symptoms of cerebral ischemia and that its prevalence drops afterward.
To our knowledge, the prevalence of HITS as a function of time has not been studied previously in individuals with cervical ICA stenosis. Sitzer et al8 showed that the rate of microembolism decreases over a period of approximately 30 days in symptomatic patients. This seemingly transient nature of microembolism in patients with carotid disease differs from its course in individuals with prosthetic cardiac valves. In the latter, the incidence of HITS increases over a period of 1 to 5 years,20 and the rate remains stable21 or increases20 22 with the duration since valve implantation. Given the known association of HITS with symptoms of cerebral ischemia,6 7 8 23 this information could potentially be useful to clinicians making decisions about the duration of anticoagulant or antiplatelet therapy. In conjunction with other variables,6 24 25 the time factor should also be taken into account by ultrasonographers reporting the results of HITS detection studies.
In patients with symptomatic carotid disease, hemodynamic and hemostatic factors at the level of the stenotic lesion may change over time and from one patient to the other, thus leading to variable clinical manifestations of initial or recurrent cerebral ischemia. The difference between hemispheric and retinal TIAs with regard to the increased risk of subsequent ipsilateral hemispheric infarction3 supports the notion that intravascular events leading to these conditions may not be the same. Similarly, the observation that a large proportion of patients have multiple TIAs preceding a single stroke21 26 suggests that there are at present unidentified intravascular factors that lead to transient rather than permanent tissue damage. Currently available techniques used to assess the severity of carotid stenosis, lesion characteristics, and extent of collateral flow do not permit a satisfactory explanation of this variability in tissue damage and clinical symptoms. The difference between TIAs and cerebral infarcts with regard to HITS positivity, as detected in this study, should be viewed in this perspective. It may identify some of the intravascular hemostatic changes associated with these conditions. It certainly raises a question regarding the thromboembolic etiology of cerebral infarction in some patients.
Several technical limitations of this investigation should be taken into consideration when interpreting its findings. First, our monitoring time of only 30 minutes per study may be too short. Because embolism may be intermittent, a HITS-negative study may actually represent a false-negative, and our analyses may have been based on a nonrepresentative selection of cases. Second, HITS were selected on the basis of criteria set a priori, and those less than 9 dB in intensity were not registered, thus introducing a bias in our selection of signals, with a shift toward larger HITS. As a result, our analyses regarding HITS rate and size have considerable limitations. Third, although medications such as anticoagulants and antiplatelet agents have not been shown to have an impact on HITS occurrence24 27 and we detected no significant difference between the HITS-positive and HITS-negative groups with regard to heparin and aspirin intake, we suspect that the relationship between these medications and HITS occurrence has not been fully elucidated. Prospective studies assessing individual patients longitudinally, over longer study periods, and with slightly different selection criteria should address some of these issues.
Selected Abbreviations and Acronyms
|HITS||=||high-intensity transient signals|
|ICA||=||internal carotid artery|
|MCA||=||middle cerebral artery|
|TCD||=||transcranial Doppler ultrasonography|
|TIA||=||transient ischemic attack|
We wish to thank Leticia Fuentes-Sanz and Miriam Marie Guanche for their help in the preparation of this manuscript. We also would like to thank Dr W.G. Bradley for his critical review of the manuscript.
- Received September 19, 1995.
- Revision received December 18, 1995.
- Accepted January 15, 1996.
- Copyright © 1996 by American Heart Association
Streifler JY, Eliasziw M, Benavente OR, Harbison JW, Hachinski VC, Barnett HJM, Simard D, for the North American Symptomatic Carotid Endarterectomy Trial. The risk of stroke in patients with first-ever retinal vs hemispheric transient ischemic attacks and high-grade carotid stenosis. Arch Neurol. 1995;52:246-249.
Timsit SG, Sacco RL, Mohr JP, Foulkes MA, Tatemichi TK, Wolf PA, Price TR, Hier DB. Brain infarction severity differs according to cardiac or arterial embolic source. Neurology. 1993;43:728-733.
Consensus Committee of the Ninth International Cerebral Hemodynamics Symposium. Basic identification criteria of Doppler microembolic signals. Stroke. 1995;26:1123.
Babikian VL, Hyde C, Pochay V, Winter MR. Clinical correlates of high-intensity transient signals detected on transcranial Doppler sonography in patients with cerebrovascular disease. Stroke. 1994;25:1570-1573.
Khaffaf N, Karnik R, Winkler WB, Valentin A, Slany J. Embolic stroke by compression maneuver during transcranial Doppler sonography. Stroke. 1994;25:1056-1057.
Sitzer M, Siebler M, Steinmetz H. Silent emboli and their relation to clinical symptoms in extracranial carotid artery disease. Cerebrovasc Dis. 1995;5:121-123.
Markus HS, Brown M. Differentiation between different pathological cerebral embolic materials using transcranial Doppler in an in vivo model. Stroke. 1993;24:1-5.
Russell D, Madden KP, Clark WM, Sandset PM, Zivin JA. Detection of arterial emboli using Doppler ultrasound in rabbits. Stroke. 1991;22:253-258.
Babikian VL, Rosales R, Pochay V. Composition of particles associated with embolic signals on transcranial Doppler ultrasonography. J Stroke Cerebrovasc Dis. 1994;4:86-90.
Saver JL, Feldmann E. Basic transcranial Doppler examination: technique and anatomy. In: Babikian VL, Wechsler LR, eds. Transcranial Doppler Ultrasonography. Saint Louis, Mo: Mosby-Year Book Inc; 1993:11-28.
Geraud G, Bes A. Is anticoagulant therapy too frequently used in ischemic stroke? Cerebrovasc Dis. 1991;1(suppl 1):120-123.
Yasaka M, Yamaguchi T, Oita J, Sawada T, Shichiri M, Omae T. Clinical features of recurrent embolization in acute cardioembolic stroke. Stroke. 1993;24:1681-1685.
Cerebral Embolism Study Group. Immediate anticoagulation of embolic stroke: brain hemorrhage and management options. Stroke. 1984;15:779-789.
Wolf PA, Kannel WB, McGee DL, Meeks SL, Bharucha NE, McNamara PM. Duration of atrial fibrillation and imminence of stroke: the Framingham Study. Stroke. 1983;14:664-667.
Braekken SK, Russell D, Brucher R, Svennevig J. Incidence and frequency of cerebral embolic signals in patients with a similar bileaflet mechanical heart valve. Stroke. 1995;26:1225-1230.
Georgiadis D, Grosset DG, Kelman A, Faichney A, Lees KR. Prevalence and characteristics of intracranial microemboli signals in patients with different types of prosthetic cardiac valves. Stroke. 1994;25:587-592.
Sliwka U, Diehl RR, Meyer B, Schondube F, Noth J. Transcranial Doppler “high intensity signals” in the acute phase and long-term follow-up of mechanical valve implantation. J Stroke Cerebrovasc Dis. 1995;5:139-146.
Tegeler CH, Knappertz VA, Nagaraja D, Mooney M, Dalley GM. Relationship of common carotid artery high intensity transient signals in patients with ischemic stroke to white matter versus territorial infarct pattern on brain CT scan. Cerebrovasc Dis. 1995;5:128-132.
Sitzer M, Muller W, Siebler M, Hort W, Kniemeyer HW, Jancke L, Steinmetz H. Plaque ulceration and lumen thrombus are the main sources of cerebral microemboli in high-grade internal carotid artery stenosis. Stroke. 1995;26:1231-1233.
Valton L, Larrue V, Arrue P, Geraud G, Bes A. Asymptomatic cerebral embolic signals in patients with carotid stenosis. Stroke. 1995;26:813-815.
Fisher CM. Occlusion of the internal carotid artery. AMA Arch Neurol Psychiatry. 1951;69:346-377.
Siebler M, Sitzer M, Rose G, Bendfeldt D, Steinmetz H. Silent cerebral embolism caused by neurologically symptomatic high grade carotid stenosis. Brain. 1993;116:1005-1015.