Do Chronic Middle Cerebral Artery Stenoses Represent an Embolic Focus?
A Multirange Transcranial Doppler Study
Background and Purpose It remains uncertain whether the annual stroke risk of 7% to 8% in middle cerebral artery (MCA) stenosis is of embolic or hemodynamic origin. Preliminary reports provide evidence of emboli exiting from acute MCA stenoses, detected by transcranial Doppler (TCD) sonography. With multirange monitoring before and after the stenosis, TCD monitoring may help for the first time to differentiate microemboli exiting from the MCA stenosis from those with a source proximal to the MCA stenosis. We searched for microembolic signals (MES) using multigated monitoring in patients with chronic MCA stenoses.
Methods Fifty-eight patients with 78 chronic stenoses of the MCA were enrolled in the study. Additional sources of embolism were ruled out by extensive clinical workup. Twenty-four patients were treated with coumarin, whereas 28 patients received aspirin. The remaining 6 patients discontinued their medication after a few weeks. The sample volume of the multirange probe was placed on either side of the stenotic area of the MCA.
Results Twenty-three (29.5%) of the stenoses were low grade, 18 (23%) were moderate, and 37 (47.5%) were severe. Thirty-seven (47%) of the stenoses were symptomatic and 41 (53%) were asymptomatic before study entry. During follow-up, 2 strokes and 7 transient ischemic attacks occurred. Computer tomography revealed two watershed-type infarcts. Sufficient insonation of the prestenotic and poststenotic segments of the MCA was possible in 70 stenoses (90%). No MES could be detected during a total of 1740 minutes’ monitoring time distal to the MCA stenoses, regardless of the patients’ medication. MES were also absent in the contralateral MCA.
Conclusions MES are not detectable in patients with chronic MCA stenoses of different degrees. No MES were found in either symptomatic or asymptomatic stenoses, regardless of the patients’ medication. These results indicate that chronic MCA stenoses do not represent a significant embolic source. The absence of MES in the prestenotic Doppler sample volume, the watershed-type infarcts during follow-up, and the absence of small-vessel disease on computed tomography suggests that hemodynamic mechanisms are responsible for recurrent cerebral ischemia.
Middle cerebral artery stenosis is a less common occlusive disease in patients in western countries than in Asian patients.1 The introduction of screening for MCA stenosis in stroke patients by use of TCD sonography has led to improved detection. The annual stroke rate for this disease is ≈7% to 8%.2 3 Hemodynamic and embolic mechanisms are thought to be responsible for recurrent TIA or stroke in this population. Many patients with MCA stenosis suffer from additional cardiovascular or carotid artery diseases that could be the source of embolism. The coincidence of potential embolic diseases could lead to additional difficulties in evaluating the pathogenetic role of MCA stenosis in symptomatic cases. With the TCD multirange system, which allows one to search for microemboli in at least two different regions of the same vessel, it is now possible to search for emboli exiting from MCA stenoses.4 5 Others5 have been able to differentiate emboli exiting from an MCA stenosis from emboli of a more proximal origin. Apart from that preliminary report with only 14 heterogeneous patients, no data concerning the frequency of MES in chronic MCA stenoses have been published. We have therefore investigated the prevalence of MES in patients with isolated and chronic MCA stenosis, in whom other sources of embolism could be ruled out, both before and after MCA stenosis using a multirange TCD system.
Subjects and Methods
From the >6000 examinations per year in the ultrasound laboratories of two centers, potential candidates were identified by reviewing the reports of consecutive TCD studies performed between 1986 and 1995. The two centers cover a population of ≈2.5 million inhabitants. One hundred fifty-three patients with 197 stenoses were selected. Ninety-five patients had to be excluded because they fulfilled at least one of the following exclusion criteria: rapid recanalization after MCA occlusion or marked regression of the stenosis within the first 6 months after the initial diagnosis. Thus, only chronic MCA stenoses lasting ≥12 months were included in the study. An insufficient bitemporal bone window was an additional exclusion criterion. An intensive clinical workup was performed before the classification of a stenosis as symptomatic or asymptomatic, involving CT/MR imaging, echocardiography, Holter ECG, and color Doppler flow imaging to exclude other sources of embolism. A potential cardiac source of embolism detected by echocardiography, lone or nonvalvular atrial fibrillation on ECG, plaques suspicious for ulceration in the CCA or ICA, and vessel lumen obstruction of >40% in the CCA or ICA detected by color-coded duplex sonography served as exclusion criteria.
All patients underwent transcranial color-coded duplex sonography in addition to conventional TCD. For patients who did not undergo DSA, the stenosis had to be confirmed by this ultrasound method before their inclusion in the study.
Patients were defined as symptomatic when they had an infarct that was ipsilateral to the MCA stenosis on CT/MR imaging or a TIA with clear symptoms of the MCA territory (brachio-facial pronounced hemiparesis or aphasia). Symptoms contralateral to the affected MCA or ischemia in the anterior or posterior cerebral artery were neglected.
Finally, 58 white patients with low-grade (peak flow velocity of 140 to 180 cm/s), middle (181 to 220 cm/s), or high-grade (>220 cm/s) MCA stenosis (38 men, 20 women; mean age, 57±12 years) were enrolled in the study. Of the total of 78 stenoses, 37 (47%) were symptomatic (2 low grade, 8 moderate, and 27 severe) and 41 (53%) were asymptomatic (21 low grade, 10 moderate, and 10 severe) at the time of the ultrasound monitoring. Of the 78 stenoses, 32 were treated with coumarin and 38 with aspirin, and no specific medication was used in 8 stenoses (Table 1⇓). Vascular risk factors were found, including arterial hypertension (76%), smoking (67%), hyperlipidemia (59%), and diabetes mellitus (34%).
TCD was performed with a pulsed Doppler device (Multidop X4, DWL) with multirange equipment. The multirange embolus detection software TCD-8 for MDX, version 8.00 K, was used on-line. One channel harvested flow signals from the proximal MCA segment or distal carotid siphon and the second from the distal MCA segment. The aim was to place the sample volumes on either side of the stenotic area of the MCA. This was not possible in all patients. Thus, monitoring was performed over prestenotic and poststenotic areas, prestenotic and intrastenotic areas, and intrastenotic and poststenotic areas. The sample volumes were chosen to be as low as possible to avoid overlap. Data were stored for later off-line analysis. Probable emboli originating proximal to the MCA stenosis passed through the different sample volumes at different times, thus showing a delay in time in the off-line evaluation of the raw signal. Emboli exiting from the MCA stenosis would only pass the second channel, located distal to the stenosis, and would produce a typical signal in the pre-FFT signal but no signal in the prestenotic sample volume. Therefore, an embolic signal would not be recorded in the proximal sample volume but only in the distal one. Artifacts produce a typical pattern in both segments in the pre-FFT signal at the same time. Signals were only accepted as MES when they fulfilled the above-mentioned multirange criteria and the recently published criteria for microemboli detection.6 The bilateral monitoring time of both MCAs was 30 minutes. To ensure that only chronic MCA stenoses were included in the study, the latency of emboli detection after the initial detection of MCA stenosis was between 12 and 108 months (mean, 53±27 months). In patients with DSA, the first TCD examination was performed 1 week before or after the DSA. Doppler examination for emboli detection was performed between 13 and 110 months (mean, 55±30 months) after the initial DSA.
Quantification of the MCA stenosis was performed using the anteroposterior and lateral projections. All angiograms were performed and reevaluated in the departments of neuroradiology of the two centers by experienced neuroradiologists blinded to the TCD results. The stenosis grading by DSA in this study was as follows: normal (no lumen reduction), low grade (pallor of contrast medium), moderate (lumen reduction >50% and <70%), and severe (subtotal stenosis with delayed distal filling and/or border-zone shift).
Statistical analysis was performed using the Mann-Whitney test for nonnormally distributed data. Spearman’s rank test was applied to examine potential correlations of nonparametric data. Significance was declared at the P<.05 level.
The 58 patients presented with a total of 78 MCA stenoses. Twenty-three stenoses (29.5%) were low grade (peak flow velocity of 140 to 180 cm/s), 18 (23%) were moderate (181 to 220 cm/s), and 37 (47.5%) were severe (>220 cm/s).
Thirty-seven (47.5%) of the stenoses were symptomatic (2 low grade, 8 moderate, and 27 severe) and 41 (53%) were asymptomatic (21 low grade, 10 moderate, and 10 severe). In 41 cases (53%), the stenoses were additionally investigated by DSA. Estimating the different degrees of MCA stenoses, the correlation between the gold standard DSA and TCD was .914 (P<.001; Spearman’s rank correlation coefficient). Although an excellent correlation was found for severe stenoses, 7 low-grade and 3 moderate stenoses were not revealed by angiography (Table 2⇓).
Sufficient insonation of the prestenotic and poststenotic segments of the MCA was possible in 70 stenoses (90%). The mean distance of the sample volume centers was 13.7±2 mm. The sample volume ranged from 5 to 7 mm (mean, 6.1±0.8 mm). In 2 stenoses, the distal segment could not be insonated sufficiently throughout the entire monitoring period because the distal signal was not large enough to ensure reliable monitoring quality. In 5 stenoses, an unequivocal prestenotic signal could not be found because of the extent of the stenoses.
No MES could be detected during a total of 1740 minutes of monitoring time either before, after, or within the MCA stenoses. MES were also absent in the contralateral MCA. For this reason, there were no differences regarding the frequency of MES between symptomatic or asymptomatic patients or for patients receiving anticoagulation or aspirin therapy.
Eight patients in the coumarin group (25%) and three in the aspirin group (8%) suffered an ipsilateral TIA (n=9) or stroke (n=2) (Table 1⇑). Only patients with high-grade and symptomatic stenoses suffered from recurring ischemic events. A statistically significant correlation between the prevalence of symptoms and the type of medication could not be found, however (P=NS by Wilcoxon test). Both patients with strokes had hemodynamic infarcts between the neighboring territories in the frontoparasagittal watershed area on CT.
There are no commonly accepted TCD criteria for the classification of angiographically confirmed MCA stenoses,7 8 9 10 but using our own criterion of an elevation of the systolic peak flow velocity, we found a good correlation with DSA, especially in moderate and severe stenoses. The discrepancy between DSA and TCD findings in low-grade MCA stenoses has been described previously.10 Although anteroposterior and lateral views are commonly obtained with the DSA technique, the latter is of little value in the detection of low-grade MCA stenosis in the commonly involved M1 segment. Although DSA is considered the “gold standard” for the evaluation of extracranial and intracranial artery disease, angiographic routine diagnosis is inexact owing to the lack of multiple projections. In other studies comparing TCD, DSA, and MR angiography, TCD revealed elevated systolic blood velocities ranging from 140 to 209 cm/s, whereas DSA was normal. Considering the strict criteria for TCD diagnosis of an MCA stenosis, these findings again raise the question of the diagnostic reliability of routinely performed DSA, especially since MR angiography supported the TCD results in these studies.
The noninvasive diagnosis of MCA stenosis has been obtained by TCD in both symptomatic and asymptomatic patients. A concept for the treatment of symptomatic and asymptomatic patients with different degrees of MCA stenosis has yet to be developed, but in the last 2 years, data concerning long-term follow-up of these patients have been published.11 12 13 Nevertheless, it remains to be discussed whether recurring ischemic events are of embolic origin from the MCA or other embolic sources or are of hemodynamic origin. Nabavi and colleagues5 were the first to use multirange Doppler equipment to search for microembolic events exiting from MCA stenoses in such patients. They found MES exiting from MCA stenoses in 2 patients with symptomatic MCA stenoses. No MES could be found in a single patient with a long history of MCA stenoses, which is in accordance with our results. However, their data were weakened by the small number of patients (14) with heterogenous cerebrovascular disease. Emboli detected by TCD in the posterior cerebral artery with an initial high-grade stenosis during slow recanalization over the course of 1 week were described for the first time by Diehl et al.14 This observation gave an impression of the progress of spontaneous thrombolysis. With multirange monitoring before and after the stenosis, TCD monitoring may help for the first time to differentiate microemboli exiting from the MCA stenosis (signal exclusively after the stenosis) from those with a source proximal to the MCA stenosis (signal can be recorded both before and after the MCA stenosis). In our population, acute MCA stenoses or rapid regression of the stenosis in the first 6 months as a sign of recanalization of an MCA occlusion or stenosis was an exclusion criterion. Therefore, our data deal with the embolic potency of chronic MCA stenoses.
Ultrasound and angiography data support the concept that acute MCA stenoses in stroke patients are highly suggestive of partial recanalization of MCA occlusions and that acute MCA stenosis represents an embolic source.5 14 15 16 In contrast, chronic MCA stenoses, which were the subject of our study, do not seem to represent an embolic source. These results seem to be independent of the degree of stenosis, the medication used, and the presence or absence of clinical symptoms attached to the stenoses.
Nevertheless, in our population, 11 patients suffered from TIA or stroke during follow-up. What might be the reasons for these clinical events?
Proximal emboli exiting from potential cardiac or artery-to-artery embolisms were excluded in our study. In addition, we could not find any microemboli with the multirange monitoring method before or after the stenosis as a sign of emboli exiting from a proximal embolic source. Furthermore, the lack of territorial infarctions in CT is an additional argument against a thromboembolic theory in our symptomatic patients.
Local thrombosis of the small vessels seen in patients with Binswanger’s disease could be another pathophysiological mechanism. However, none of our patients with TIA or stroke had evidence of small-vessel disease on CT.
Our hypothesis suggests that recurrent ischemic events in our patients are of hemodynamic origin. Insufficient collateral supply in combination with exhausted vasomotor reactivity distal to the MCA stenoses is likely to be the reason for watershed-type infarcts, which were observed on CT in the two patients with stroke during follow-up.17 18 These watershed-type infarcts are also observed in persons with moyamoya disease, in which the collateralizing capacity of the circle of Willis is severely reduced owing to occlusions of the basal arteries. Therefore, the motivation for anticoagulation therapy and prescription of platelet aggregation inhibitors is not to prevent the patients from experiencing recurrent embolic events. The therapeutic approach should prevent the progression of MCA stenoses and support the development of collateral pathways.
We conclude that microemboli detection with the multirange technique is helpful in differentiating the embolic source not only in the extracranial vessels but also in the intracranial arteries. Our data indicate that patients with chronic MCA stenoses seem to have a low risk for embolic events exiting from the MCA stenoses. The annual stroke rate of 7% to 8% in the literature and the recurring events in our population may be predominately explained by hemodynamic mechanisms. Local atherosclerotic mechanisms or intermittent embolic activity is possible but very unlikely in our population. Therefore, anticoagulation therapy or medication with platelet aggregation inhibitors should prevent the progression of MCA stenoses and support the development of collateral pathways.
The pathological mechanisms in acute MCA occlusions or partial MCA recanalizations in cerebral ischemia are different, and MCA stenoses may represent a potential embolic source in these patients. This must be evaluated in more extensive studies of peracute stroke patients.
Selected Abbreviations and Acronyms
|CCA||=||common carotid artery|
|DSA||=||digital subtraction angiography|
|FFT||=||fast Fourier transformation|
|ICA||=||internal carotid artery|
|MCA||=||middle cerebral artery|
|TIA||=||transient ischemic attack|
Ulrich Sliwka, Christof Klötzsch, Octavian Popescu, Katrin Brandt, and Peter Schmidt were responsible for the conception and design of the study and for analysis and interpretation of the data. Peter Berlit and Johannes Noth revised the article critically for important intellectual content. For helpful suggestions, we thank Dr Stuart Fellows, Department of Neurology, RWTH Aachen, Germany, and Prof Weiller, Friedrich-Schiller-Universität Jena, Department of Neurology, Germany.
- Received April 4, 1997.
- Revision received May 2, 1997.
- Accepted May 6, 1997.
- Copyright © 1997 by American Heart Association
Sacco RL, Kargman DE, Gu Q, Zamanillo MC. Race-ethnicity and determinants of intracranial atherosclerotic cerebral infarction: the Northern Manhattan Stroke Study. Stroke. 1995;26:14-20.
Bogousslavsky J, Barnett HJM, Fox AJ, Hachinski VC, Taylor W. Atherosclerotic diseases of the middle cerebral artery. Stroke. 1986;17:1112-1120.
Georgiadis D, Goeke J, König M, Hill M, Nabavi DG, Stögbauer F, Zunker P, Ringelstein EB. A novel technique for identification of Doppler microembolic signals based on the coincidence method: in vitro and in vivo evaluation. Stroke. 1996;27:683-686.
Nabavi DG, Georgiadis D, Mumme T, Zunker P, Ringelstein EB. Detection of microembolic signals in patients with middle cerebral artery stenosis by means of a bigate probe: a pilot study. Stroke. 1996;27:1347-1349.
Consensus Committee of the Ninth International Cerebral Hemodynamic Symposium. Basic identification criteria of Doppler microembolic signals. Stroke. 1995;26:1123.
Rorick MB, Nichols FT, Adams RJ. Transcranial Doppler correlations with angiography in detection of intracranial stenosis. Stroke. 1994;25:1931-1934.
The Warfarin-Aspirin Symptomatic Intracranial Artery Disease Study Group. The Warfarin-Aspirin Symptomatic Intracranial Artery Disease Study. Neurology. 1995;45:1488-1493.
Kühnen J, Steinke W, Kern R, Artemis N, Hennerici M. Long-term course of symptomatic and asymptomatic middle cerebral artery stenosis. Neurology. 1996;46:281. Abstract.
Sliwka U, Klötzsch C, Popescu O, Brandt K, Berlit P, Noth J. Anticoagulation versus antiaggregation in patients with middle cerebral artery stenosis. Cerebrovasc Dis. 1996;6(suppl 3):8. Abstract.
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.
Ringelstein EB, Biniek R, Weiller C, Ammeling B, Nolte PN, Thron A. Type and extent of hemispheric infarctions and clinical outcome in early and delayed middle cerebral artery recanalization. Neurology. 1992;42:289-298.
Weiller C, Ringelstein EB, Reiche W, Buell U. Clinical and hemodynamic aspects of low-flow infarctions. Stroke. 1991;22:1117-1123.