Microemboli in Patients With Vertebrobasilar Ischemia
Association With Vertebrobasilar and Cardiac Lesions
Background and Purpose Microembolic high-intensity transient signals (HITS) on transcranial Doppler ultrasound (TCD) are associated with carotid stenosis, artificial heart valves, and other cardiac conditions. Only one case report describes stroke-related HITS in the posterior cerebral circulation. This study was designed to assess prospectively the prevalence of asymptomatic microemboli in vertebrobasilar ischemia and to determine whether potential cardioembolic sources, vertebral or basilar occlusive disease, and infarct subtypes predict the detection of HITS.
Methods We investigated 52 consecutive patients with acute or recent vertebrobasilar ischemia within 48 hours after admission. TCD monitoring was performed according to established criteria for 20 minutes on each posterior cerebral artery. Fetal origin of the posterior cerebral artery was an exclusion criterion and ruled out by carotid compression.
Results Microembolic signals were detected in 10 patients (19.2%). In a multivariate logistic regression analysis that included all independent variables, potential cardiac sources were significantly associated with HITS (odds ratio [OR], 14.3; 95% confidence interval [CI], 1.6 to 128.4), in particular when more than one cardiac abnormality (OR, 32.7; CI, 4.1 to 259.3) or a high-risk source (OR, 14.0; 95% CI, 2.3 to 84.9) was found. Vertebrobasilar vessel lesions and infarct subtype were not related to the detection of microemboli.
Conclusions Cardiac sources of embolism are a determinant of HITS in posterior cerebral circulation ischemia, suggesting that cardioembolic infarcts in that territory might be more common than suspected. Vertebrobasilar vessel abnormalities are less likely to lodge microemboli, which may indicate histopathological changes different from carotid artery disease.
The detection of asymptomatic HITS on TCD has been reported in patients with ischemic strokes of the carotid territory,1 2 3 4 5 as well as in subgroups of patients with stroke risk factors such as carotid stenosis6 and various cardiac conditions.7 8 However, although approximately 25% of brain infarcts involve the vertebrobasilar territory,9 10 only one case report regarding stroke-related microemboli in the posterior cerebral circulation has been published.11 Therein a patient with acute thromboembolic stenosis of one PCA is described, and HITS were interpreted as detaching particles due to autolysis of the clot. No other data exist about microembolic signals on TCD in patients with posterior circulation stroke.
This study was designed to assess prospectively the prevalence of HITS in patients with acute or recent ischemia in the vertebrobasilar circulation. We furthermore evaluated whether PCSE, stenotic or occlusive lesions of the VA and BA, and nonlacunar infarcts are associated with the detection of microemboli.
Subjects and Methods
Two-hundred eighty consecutive patients with acute or recent cerebral ischemia were screened for this study. Patients were excluded from further evaluation when the anterior cerebral circulation or an undetermined vascular territory was affected (n=202, 72.1%). Of the remaining 78 patients with vertebrobasilar events, 16 (20.5%) were excluded because of bilateral inadequate acoustical window; in addition, a carotid compression test revealed fetal origin of the PCA in 8 subjects, who were also excluded. Two patients with a clinical diagnosis of basilar occlusion died before the monitoring. Informed consent was obtained from all patients. In patients without visualized infarcts, the diagnosis of vertebrobasilar ischemia was based on one or more of the following clinical criteria: nuclear and supranuclear oculomotor abnormalities including nystagmus; nausea and vertigo; Horner’s syndrome; decreased level of consciousness; dysarthria combined with ataxia; truncal or appendicular ataxia without other motor deficits; homonymous visual field defects; crossed sensory or motor syndromes; pure sensory syndrome; and tetraparesis.
Fifty-two patients were finally enrolled in the study (Table 1⇓). The median time between onset of the symptoms and HITS monitoring was 2 days (range, 0.5 to 19 days; patients examined on the event day were assigned a value of 0.5 day). Thirty patients were examined within 2 days, 19 within 3 to 7 days, and 3 patients more than 10 days after the ischemia.
Embolic material in the posterior circulation is very likely to reach the PCA.12 For this reason HITS monitoring was performed through a transtemporal approach on each PCA simultaneously for 20 minutes. Patients were in the supine position, and the P1 segment was identified in a depth of 60 to 70 mm according to established criteria.13 Patients were excluded when compression of the common carotid artery led to a decrease or discontinuation of flow in the insonated artery, thus making fetal origin of the PCA and monitoring of the internal carotid artery by mistake very unlikely. A TC 2020 device (EME/Nicolet) with pulsed-wave 2-MHz probes and a 486 processor employing a 128-point fast Fourier transform was used for all examinations. HITS counting was performed on-line by two investigators, both attending all examinations, who were blinded to the results of all diagnostic procedures. We applied recently published criteria for HITS identification14 but used a higher-amplitude threshold (9 instead of 3 dB). Signals were accepted when both observers were in agreement and all identification criteria were met. To facilitate monitoring, specially designed software (EME/Nicolet, version 1.0) recorded HITS within segments of 1-second duration, thereby allowing determination of direction, amplitude, and duration of microembolic signals. This additional recording on computer hard disk did not include the audio channel. Therefore, post hoc analysis of the few equivocal signals was confined to amplitude, duration, and random occurrence. Gain was set to a low level (6 to 10) with a sample volume of 10 or 12 mm. The pulse repetition frequency ranged from 2 to 7 kHz; sweep time was 3.1 seconds. On the basis of these machine parameters and according to a recent report,15 the fast Fourier transform time-window overlap varied from 67% to 90%, and therefore no HITS were missed as a result of inadequacy of processing speed. Monitoring was confined to one side in 5 patients because of technically difficult acoustical windows.
All patients underwent color-coded duplex sonography of the VA in the neck; pulsed-wave sonography of the VA origins; and TCD studies of the circle of Willis, the VA, and the BA. MR angiography was performed in 29 patients and conventional angiography in 2 subjects. According to the combined findings of this vascular workup, 11 patients (21.2%) had vessel abnormalities: 5 BA stenoses (combined with distal VA stenosis in 2 cases), 3 unilateral or bilateral proximal VA stenoses, 1 distal VA stenosis, and 2 cases of VA occlusion were detected. All patients had an electrocardiogram; 46 transthoracic and 15 transesophageal echocardiographies were performed. PCSE corresponding to published criteria16 were revealed in 27 patients (Table 2⇓), of whom 15 had more than one source and 8 a high-risk source (atrial fibrillation, prosthetic valve, ventricular thrombus). In 7 patients both a vessel lesion and PCSE were diagnosed. Brain imaging studies consisted of MRI (43 patients) or CT scan (9 patients). A new infarct was demonstrated in 28 patients (53.9%). Infarcts in the territory of the PCA or the large circumferential arteries (posterior inferior, anterior inferior, and superior cerebellar artery) were found in 11 patients, while 17 infarcts were classified as lacunar. The latter group included infarcts of penetrating branches in the pons or midbrain (median/paramedian). Imaging studies were negative in 7 patients who clinically suffered an infarct; all but 1 of these patients underwent MRI. Evidence for prior ischemic events was obtained from history and imaging.
In addition, any form of antithrombotic treatment at the time of the HITS monitoring was recorded. Anticoagulation was regarded as therapeutic when the activated partial thromboplastin time was at least two times the patient’s baseline or the international normalized ratio was greater than 1.5. Patients with subtherapeutic anticoagulation and those on low-dose subcutaneous heparin (2×5000 U/d) were considered untreated.
Hypotheses and Statistical Methods
Independent variables were defined as follows: absence or presence of PCSE; normal or diseased (stenotic/occluded) VA or BA; and lacunar or nonlacunar infarct subtype. The prospective part of the study included the following three hypotheses: (1) the presence of microemboli is associated with PCSE; (2) patients with pathological vertebrobasilar vessels are more likely to have HITS than patients with normal vessels; and (3) the prevalence of HITS is higher in patients with nonlacunar infarcts. Univariate analyses were applied with the two-tailed Fisher’s exact test. Multivariate logistic regression analyses were conducted to assess independent effects of the predictor variables.
Asymptomatic microembolic signals were detected in 10 patients (19.2%) or 14 of 99 insonated vessels (14.1%). Bilateral HITS were found in 4 patients, while 5 had unilateral microemboli; 1 patient with HITS was monitored on one side only. The mean rate of microemboli in HITS-positive vessels was 4.7±8.8 (range, 1 to 29) per 20 minutes; when 1 patient with a BA stenosis and multiple HITS (23 right, 29 left) was omitted, the rate dropped to 1.2±0.4/20 minutes (range, 1 to 2), corresponding to a rate of 3.6 per hour.
The prevalence of microemboli was higher among patients with PCSE (9/27, 33.3%) than among patients with normal cardiac findings (1/25, 4%). This distribution was significant (P=.012, Fisher’s exact test). A multivariate logistic regression model (Table 3⇓) including the three independent variables and other confounding variables (age, sex) confirmed PCSE as an independent determinant of HITS. Further analyses in which the same regression model was used demonstrated that the detection of more than one PCSE (OR, 32.7; 95% CI, 4.1 to 259.3; P=.001) and the finding of a high-risk source (OR, 14.0; 95% CI, 2.3 to 84.9; P=.004) were strong and independent predictors for microemboli. The prevalence of HITS was almost equally distributed among patients with (2/11, 18.2%) and without (8/41, 19.5%; P=1.0, Fisher’s exact test) vessel disease. Neither was the type of brain infarction separated by the absence or presence of microemboli (nonlacunar: 2/11, 18.2%; lacunar: 2/17, 11.8%; P=1.0, Fisher’s exact test; patients with nonvisualized infarcts and transient ischemic attacks were excluded from this analysis).
According to a post hoc analysis, the prevalence of HITS was not different among treated or untreated patients (P=.49, Fisher’s exact test). Subanalyses of patients on anticoagulation or antiplatelet therapy did not reveal an association with the detection of HITS, although there was a trend that patients treated with aspirin were less likely to have microemboli (P=.096, Fisher’s exact test).
Further retrospective analyses were conducted to assess the effect of a prior stroke on the prevalence of HITS. Neither a previous ischemic event in any territory (29 patients; P=.48, Fisher’s exact test) nor in the posterior circulation (18 patients; P=.72, Fisher’s exact test) predicted the detection of microemboli.
This prospective study has demonstrated for the first time that microembolic signals on TCD can be detected in the PCA of consecutive patients with acute or recent vertebrobasilar ischemia. The results also indicate that HITS in acute vertebrobasilar stroke are associated with PCSE but, unlike those in the anterior circulation, they are not related to occlusive vessel disease. In particular, the presence of two or more cardiac abnormalities and cardiac findings carrying a high risk of embolism predict the detection of microembolic signals.
Embolism, either artery-to-artery or cardiogenic, is a common phenomenon in posterior circulation ischemia,12 and PCSE have been reported in 22% to 70% of patients with infarcts in the vertebrobasilar territory17 18 or unselected patients with cerebral ischemia.19 More than 50% of our patients had a PCSE. Inclusion of more cardiac findings presumed to be potential sources such as valvular strands, spontaneous echo contrast, and aortic atheroma in our study most likely explains the difference from the earlier report.17 Thus far, few data exist regarding PCSE and HITS in the posterior circulation. Studies featuring the vascular distribution of artificial particles or microbubbles in patients with cardiac right-to-left shunt detected HITS in the PCA or BA in 75% to 80% of the cases.20 21 According to Grosset et al,22 almost 30% of cardiogenic emboli reach the VA, which is more than the 20% expected from the relative proportion of cerebral blood flow to the posterior circulation. These findings, together with our results, suggest that cardiogenic emboli enter the vertebrobasilar system more often than clinically suspected, and that cardiogenic ischemic strokes in this territory might be more common than currently assumed, in particular in patients with multiple PCSE or high-risk sources.
The detection of HITS was not related to vessel abnormalities in our study, although atherosclerosis is known to be the most common cause of vertebrobasilar ischemia.23 The sensitivity of ultrasonic tests for VA and BA lesions, especially the VA origin, is relatively low.24 Because approximately 50% of our patients had merely Doppler studies to examine the vessel status, the prevalence of stenoses or occlusions might have been underestimated. This is further indicated by a previous study based on MR angiography, in which more than 60% of patients with posterior circulation infarcts were found to have arterial lesions.25 One could argue that the detection of more vascular lesions would have revealed an association of HITS with occlusive vessel disease. However, in our study not even a trend indicating a possible association of microemboli and vessel abnormalities was found. Furthermore, autopsy-based studies indicate that vertebrobasilar atherosclerotic lesions are less ulcerative than in carotid atheromatous disease,26 27 but plaque ulceration is significantly related to microembolic signals in carotid artery stenosis.28 29 We therefore conclude that VA and BA lesions are less likely to lodge microemboli than atherosclerotic carotid lesions, which suggests different histopathological changes among these vessels. Nevertheless, certain lesions within the vertebrobasilar arteries are associated with increased HITS production, as our patient with a BA stenosis indicates. Thus, the detection of HITS in the PCA without an underlying cardiac source may indicate an unrevealed vessel abnormality and prompt further diagnostic tests in these patients.
No association with HITS was found among patients with specific types of infarction or treatment in our study, which is in concurrence with our findings in carotid territory stroke4 and other reports.7 8 However, Grosset et al2 associated HITS with presumed thromboembolic or cardioembolic stroke in the anterior circulation, while no microemboli were detected in their lacunar cases. This study differed from ours methodologically in that stroke subgroups were classified based on etiologic criteria, and accordingly patients with clinical/radiological lacunar syndromes were treated as having cardioembolic strokes when a concomitant PCSE was found. Our infarct classification was exclusively based on imaging criteria, which may explain the differing results. Further studies with more patients are needed to clarify these discrepancies.
Selected Abbreviations and Acronyms
|HITS||=||high-intensity transient signals|
|PCA||=||posterior cerebral artery|
|PCSE||=||potential cardiac source(s) of embolism|
|TCD||=||transcranial Doppler ultrasound|
This study was supported by the Horace W. Goldsmith Foundation. The authors are indebted to Weidong Ma, MS, for her assistance in preparing the statistical software.
- Received October 11, 1996.
- Revision received November 25, 1996.
- Accepted November 25, 1996.
- Copyright © 1997 by American Heart Association
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.
Grosset DG, Georgiadis D, Abdullah I, Bone I, Lees KR. Doppler emboli signals vary according to stroke subtype. Stroke. 1994;25:382-384.
Tong DC, Albers GW. Transcranial Doppler–detected microemboli in patients with acute stroke. Stroke. 1995;26:1588-1592.
Koennecke H-C, Mast H, Trocio SH Jr, Sacco RL, Ma W, Mohr JP, Kim P, Thompson JLP. Prevalence of microemboli in patients with acute ischemic stroke. Neurology. 1996(suppl 2);45:A387. Abstract.
Daffertshofer M, Ries S, Schminke U, Hennerici M. High-intensity transient signals in patients with cerebral ischemia. Stroke. 1996;27:1844-1849.
Siebler M, Kleinschmidt A, Sitzer M, Steinmetz H, Freund H-J. Cerebral microembolism in symptomatic and asymptomatic high-grade internal carotid artery stenosis. Neurology. 1994;44:615-618.
Sliwka U, Job F-P, Wissuwa D, Diehl RR, Flachskampf F-A, Hanrath P, Noth J. Occurrence of transcranial Doppler high-intensity transient signals in patients with potential cardiac sources of embolism: a prospective study. Stroke. 1995;26:2067-2070.
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.
Bogousslavsky J, Van Melle G, Regli F. The Lausanne Stroke Registry: analysis of 1000 consecutive patients with first stroke. Stroke. 1988;19:1083-1092.
Diehl RR, Sliwka U, Rautenberg W, Schwartz A. Evidence for embolization from a posterior cerebral artery thrombus by transcranial Doppler monitoring. Stroke. 1993;24:606-608.
Fujioka KA, Douville CM. Anatomy and freehand examination techniques. In: Newell DW, Aaslid R, eds. Transcranial Doppler. New York, NY: Raven Press Publishers; 1992:9-31.
Consensus Committee of the Ninth International Cerebral Hemodynamic Symposium. Basic identification criteria of Doppler microembolic signals. Stroke. 1995;26:1123.
Markus H. Importance of time-window overlap in the detection and analysis of embolic signals. Stroke. 1995;26:2044-2047.
Adams HP, Bendixen BH, Kappelle J, Biller J, Love BB, Gordon DL, Marsh EE, and the TOAST Investigators. Classification of subtype of ischemic stroke: definitions for use in a multicenter clinical trial. Stroke. 1993;24:35-41.
Bogousslavsky J, Cachin C, Regli F, Despland P-A, Van Melle G, Kappenberger L, for the Lausanne Stroke Registry Group. Cardiac sources of embolism and cerebral infarction: clinical consequences and vascular concomitants: the Lausanne Stroke Registry. Neurology. 1991;41:855-859.
Amarenco P, Hauw J-J. Cerebellar infarction in the territory of the superior cerebellar artery: a clinicopathologic study of 33 cases. Neurology. 1990;40:1383-1390.
Mast H, Thompson JLP, Völler H, Mohr JP, Marx P. Cardiac sources of embolism in patients with pial artery infarcts and lacunar lesions. Stroke. 1994;25:776-781.
Venketasubramanian N, Sacco RL, Di Tullio M, Sherman D, Homma S, Mohr JP. Vascular distribution of paradoxical emboli by transcranial Doppler. Neurology. 1993;43:1533-1535.
Klingelhöfer J, Schwarze J, Sander D, Matzander G. Dynamics and spatial distribution of emboli in the anterior and posterior cerebral circulation in patients with right-to-left shunt. J Neuroimaging. 1995;5(suppl):S66. Abstract.
Caplan LR. Vertebrobasilar occlusive disease. In: Barnett HJM, Mohr JP, Stein BM, Yatsu FM, eds. Stroke: Pathophysiology, Diagnosis, and Management. 2nd ed. New York, NY: Churchill Livingstone, Inc: 1992:443-515.
Von Büdingen HJ, Staudacher T. Evaluation of vertebrobasilar disease. In: Newell DW, Aaslid R, eds. Transcranial Doppler. New York, NY: Raven Press Publishers; 1992:167-195.
Bogousslavsky J, Regli F, Maeder P, Meuli R, Nader J. The etiology of posterior circulation infarcts: a prospective study using magnetic resonance imaging and magnetic resonance angiography. Neurology. 1993;43:1528-1533.
Schwartz C, Mitchell J. Atheroma of the carotid and vertebral arterial systems. Br Med J. 1961;2:1057-1063.
Moossy J. Morphology, sites, and epidemiology of cerebral atherosclerosis. Proc Assoc Res Nerv Ment Dis. 1966;51:1-22.
Valton L, Larrue V, Arrue P, Geraud G, Bes A. Asymptomatic cerebral embolic signals in patients with carotid stenosis: correlation with appearance of plaque ulceration on angiography. Stroke. 1995;26:813-815.
Sitzer M, Müller W, Siebler M, Hort W, Kniemeyer H-W, Jäncke 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.