Prevalence and Time Course of Microembolic Signals in Patients With Acute Stroke
A Prospective Study
Background and Purpose Cerebral emboli can be identified by the presence of typical microembolic signals (MES) in transcranial Doppler (TCD) spectral curves. The usefulness of this technique was studied by evaluating the prevalence and time course of MES in patients with acute stroke. In addition, we examined the influence of anticoagulation therapy on the occurrence of MES. Another study objective was to identify the value of MES in elucidation of the underlying pathology of cerebral ischemia in patients with acute stroke.
Methods We used bilateral TCD monitoring of the middle cerebral artery to search for microemboli in 100 patients with acute nonhemorrhagic stroke in the anterior circulation. Monitoring time was for 30 minutes at admission (examination I), after 24 hours (examination II), and again after 48 hours (examination III).
Results Twenty-two of the 100 patients had to be excluded from the study after examination I because retrospectively they did not fulfill the inclusion criterion or because they had an insufficient bone window. Forty of the patients (51%) showed MES during at least one of the three TCD examinations. In 9 of the 47 patients without MES during examination I (19%), MES could be recorded subsequently during examinations II and III. A statistically significant decrease in the prevalence of MES occurred between examinations I and III (P=.01). The frequency of MES in a single patient decreased between examinations I and II but increased again in examination III, although it did not reach the initial level. Prevalence of MES was the highest during the period up to 6 hours after the onset of symptoms. However, even at >72 hours after the onset of symptoms, a substantial number of MES could be recorded. In 18 of the 21 patients with carotid artery stenosis or occlusion who showed MES (86%), these signals occurred ipsilateral to the affected carotid artery. In 5 of the 13 patients with MES and a potential cardiac source of embolism (38%), MES were observed bilaterally. Forty-one patients were without anticoagulation treatment at the time of examination; 19 of these patients (46%) presented with MES. In contrast, of the 37 patients receiving anticoagulation treatment at the time of the first examination, MES could be recorded in only 12 (32%).
Conclusions Microemboli are a frequent phenomenon in patients with acute stroke arising from a variety of causes, both in the very early stages and several days after the onset of symptoms. The prevalence of MES decreases significantly over time. MES occur more frequently in patients with carotid artery disease than in patients with a potential cardiac source of embolism. Ipsilateral MES are frequent in patients with carotid artery disease, whereas bilateral MES are suggestive of a cardioembolic origin. Anticoagulation treatment appears to decrease the prevalence of MES, but microemboli still occur in patients receiving intravenous therapy with heparin. Because MES occur intermittently, TCD examinations should be repeated several times, even in patients without MES in the first examination, and long-term monitoring equipment is necessary.
The diagnosis of embolic stroke is currently based on the detection of a potential embolic source, usually after the neurological event. The lack of a suitable tool for emboli detection means that it is difficult to distinguish between an embolic and a thrombotic cause of cerebral ischemia. Data regarding the duration, rate, and relationship between embolism and clinical symptoms are limited.1
In the present study, we evaluated the usefulness of this new technique. A study goal was to reveal the prevalence and time course of MES in patients with acute stroke. In addition, we examined the influence of anticoagulation therapy on the occurrence of MES. Another objective was to identify the value of MES in uncovering the underlying pathology of cerebral ischemia in patients with acute stroke.
Subjects and Methods
One hundred patients presenting with acute nonhemorrhagic cerebral ischemia in the anterior circulation were studied prospectively. The onset of symptoms >1 week before the initial examination served as an exclusion criterion. Cerebral CT was performed routinely. On admission, MCAs were monitored over a 30-minute period by the use of TCD with 2-MHz probes. The probes were fixed on the head with an elastic headband. TCD monitoring was performed with the multirange technique. This technique allows the simultaneous examination of different sections within one vessel with a single probe. Probable emboli pass through the different sample volumes at different times, thus showing a delay in time in the off-line evaluation of the raw signal. In contrast, an artifact, which does not travel with the blood flow, will lead to a simultaneous signal in each sample volume. This allows a distinct discrimination between emboli and artifacts. We used a DWL Multidop×4 system. The sample volume length was 6 mm; the distance between the two monitoring depths was 5 mm; and the time resolution was 56 Hz. MES frequency was expressed as the median value, and 95% CI was expressed as the number of signals that occurred within 30 minutes. After the initial TCD examination on admission, two additional TCD examinations were performed at ≈24-hour intervals. MCA Doppler emboli signals were identified according to published criteria.4 5 In addition, they had to be recognized by the multirange system. All monitoring sessions were recorded on digital audiotape for later reevaluation. Three observers with long-term experience in microemboli detection were blinded to the patient data and were asked to note both the total MES count and individual signal positions on the tape. By definition, only signals detected by the three observers and the multirange system were counted. In 2 subjects, there was a disagreement between the classic criteria and the multirange system; these signals were not counted as MES.
Two-dimensional transesophageal echocardiography was performed in all patients. The degree of carotid artery stenosis was also estimated with the use of continuous-wave Doppler sonography and color-coded duplex sonography using standard criteria. Parametric data were expressed as median and 95% CI. The significance of frequency distribution within fourfold tables was calculated with Fisher's exact two-tailed test for independent variables and with the McNemar test for related dichotomous variables. Continuous variables were compared using the Mann-Whitney U test. Larger sets of continuous data were evaluated using Kruskal-Wallis testing. All statistical analysis was performed with SPSS statistical software.
Criteria for the 25 control subjects (mean age, 49 years; age range, 23 to 71 years) included a normal ECG, Doppler ultrasound of the extracranial and intracranial circulations, normal transthoracic echocardiography, and no history of cerebrovascular, peripheral artery, or cardiac disease. In addition, all the control subjects were nonsmokers and lacked a history of hypertension or diabetes mellitus. None of the control subjects were undergoing anticoagulation or antiplatelet therapy at the time of ultrasound examination. Five subjects were patients of the neurological clinic, and 20 subjects were healthy volunteers from the hospital staff. No MES were detected during the 30 minutes of bilateral TCD monitoring with multirange technique in the control subjects.
Fourteen of the 100 patients presented with an insufficient transtemporal bone window. Eight additional patients had to be excluded after the first examination: 2 patients retrospectively had intracerebral bleeding, and 6 had cerebral ischemia in the vertebrobasilar territory. None of these subjects demonstrated MES in the first TCD examination.
MES Prevalence in Study Population
Seventy-eight patients could be evaluated (27 women and 51 men; age range, 29 to 90 years; mean age, 62.5 years). Forty of the patients with acute stroke (51%) presented with MES at least once during the three TCD examinations (median, 3 times; CI, 3.4 times). In 9 of the 47 patients without MES (19%) during the initial examination, MES could be recorded during examinations II and III.
MES Dynamics in Patients With Carotid Artery Disease and Patients With a Potential Cardiac Source of Embolism
Although patients with an affected carotid artery demonstrated an almost constant frequency of MES during the different examinations, patients with a potential cardiac source of embolism showed progressively fewer MES during the three examinations. This difference was not, however, statistically significant (P1=.66, P2=.26, P3=.11).
MES Prevalence Over Time
The prevalence of MES in examinations I to III and their frequency in the single patients are demonstrated in Fig 1⇓, which shows a statistically significant decrease (P=.01) in MES prevalence over time from 31 to 16 patients positive for MES. The number of MES during the single examination decreased from examination I to II, from an median MES frequency of 4 (CI, 5.1) to 3 (CI, 5.1), but increased again during examination III (median, 4; CI, 7.7) without reaching the initial level. The differences in MES frequency in the single patient were not statistically significant (P=.38).
Time Interval Between Onset of Symptoms and TCD Examination
The influence of the time interval between the onset of symptoms and the first TCD examination on the MES prevalence is demonstrated in Fig 2⇓. The prevalence for MES was the highest, with an interval of <6 hours between the onset of symptoms and the TCD examination. The time interval between TCD examination and the onset of symptoms ranged from 1.5 to 168 hours. When considering fixed time intervals, the prevalence decreased initially and increased again 72 hours after onset of symptoms but without reaching the initial level. A statistically significant difference in MES prevalence between the early and late examinations could not be demonstrated (P=.26).
Carotid Artery Disease and MES
Carotid artery stenosis or occlusion was present in 34 cases (44% of all patients). Twenty-one of these 34 patients with carotid artery disease presented with MES. Fig 3⇓ demonstrates the influence of the carotid artery pathology on the frequency of MES occurrence. In 18 of the 21 patients with carotid artery stenosis or occlusion and MES (89%), these embolic signals occurred ipsilateral to the diseased carotid artery.
Potential Cardiac Sources of Embolism and MES
A potential cardiac source of embolism defined by transesophageal echocardiography or ECG could be diagnosed in 29 patients (36%). We found atrial fibrillation (10 patients); PFO (9 patients); coronary artery disease plus ejection fraction of <30%, including at least three wall segments of hypokinesia/akinesia (2 patients); intra-atrial aneurysm of the septum (3 patients); sclerosis of the aortic valve (3 patients); thrombus in the ventricle (1 patient); and infectious endocarditis (1 patient). MES could be recognized in 13 cases (45%). In 4 of the 9 patients with PFO, MES were detected. The exclusion of the subgroup of patients with PFO did not affect the results. In 5 of the 13 (38%), MES occurred bilaterally. In patients with a potential cardiac source of embolism and unilateral MES, these embolic signals were detected in the left MCA in 7 and in the right MCA in only 1. Unilateral MES occurred more frequently in patients with a carotid artery stenosis than in patients with a cardioembolic source of stroke. On the other hand, patients with a cardioembolic source of the cerebral ischemia had more bilateral MES than did patients in whom carotid artery disease was responsible for the stroke. Although our data indicate an obvious difference, it did not reach a statistically significant level (P=.086).
Coexisting Cardiac and Carotid Disease
Coexisting cardiac and carotid disease was diagnosed in nine patients. MES occurred in four of these patients (44%). The frequency was 2/3/6/65 per 30 minutes. This subpopulation did not show a higher MES count compared with patients with a single embolic source.
MES in Patients With Normal Results in Diagnostic Workup
In 20 of the 78 stroke patients, diagnostic work-up showed normal results. However, 8 of these patients (40%) presented with MES during at least one of the three TCD examinations, in comparison to the 32 of 58 patients with MES and a proven embolic source for their ischemia (55%).
Effect of Anticoagulation Therapy
The effect of anticoagulation with intravenous heparin on the MES prevalence was investigated only in the initial TCD examination on admission. Only at this time were the numbers of patients with (38 [49%]) and without (40 51%]) anticoagulation comparable. The occurrence of MES in patients with and without anticoagulation is demonstrated in Fig 4⇓. Patients receiving anticoagulation treatment tended to show less MES, although this difference was not statistically significant (P=.15).
Nineteen patients were not receiving intravenous heparin treatment during the initial examination and demonstrated MES. During follow-up, 11 of them had MES although they were receiving heparin treatment, whereas 5 did not show MES after the initiation of intravenous heparin treatment. Two patients did not receive heparin and continued to show MES, whereas 1 patient had no further MES without heparin.
Embolic infarcts were visible on the CT scan in 64 patients. Five patients were diagnosed with watershed-type infarcts, and 9 patients had small-vessel disease. The 40 patients with MES demonstrated 34 embolic infarcts, four watershed-type infarcts, and two CT scans showing small-vessel disease. The distribution of infarct types in the patients with and without MES was comparable. Patients positive for MES did not show more embolic infarcts than those without MES during the Doppler examination.
There is strong evidence from both experimental and clinical data that the MES observed with TCD represent microemboli of different origins.6 7 8 9 In the present study, we found that MES occur frequently in a large population of patients presenting with acute stroke in the anterior cerebral circulation of different origins. We were able to demonstrate that MES occur over a long period after the initial clinical event. In addition, it could be shown that the dynamics of MES prevalence changes over time in these patients. The need for recurrent TCD examination was demonstrated by the finding that the occurrence of MES is inconstant and that they may be absent during a single examination. MES detection may be helpful in discriminating between different potential sources of embolism (ie, artery to artery or cardioembolic strokes).10 TCD monitoring may help to monitor the dynamic process of embolism in embolic strokes.
Many studies have used TCD for microemboli detection, and criteria for MES detection have been published. Nevertheless, there remain difficulties in discriminating real embolic signals from artifacts. To increase the power of our results, all signals fulfilled the published criteria for MES but, in addition, were identified with the use of the multirange ultrasound technique mentioned above.11
The repetition of the TCD examinations revealed that a high number of embolic events occurred even several days after the initial occurrence of the clinical deficit. The results reported here suggest that continuous subclinical microemboli may be substantially underestimated through clinical or radiological evaluation alone. These data presume that the underlying pathology in stroke patients is a dynamic process that is incomplete after the clinical event.12 13 For this reason, there may be arguments for the use of immediate anticoagulation therapy with intravenous heparin in patients with acute stroke on admission (after having ruled out an intracerebral bleeding) to prevent further neurological deficits. According to our data, the thromboembolic activity appears to be highest in the first hours after the onset of symptoms. This could be due to the sudden activation of a possible embolic source.12 13 The significant decrease in MES prevalence over time indicates an incomplete healing process and/or the possible effect of anticoagulation therapy. However, MES occur several days after the onset of symptoms in a substantial number of patients. These data show that patients with an acute ischemic stroke have subclinical microemboli for a long period, which is in accordance with other findings restricted to patients with carotid artery stenosis.14 Clinical data from patients with recurrent embolization in acute cardioembolic stroke demonstrate recurrent brain or systemic embolization in 20% of these patients, which fits well with our data.15 In the present study, 16 of 78 patients (21%) still demonstrated MES at examination III as a sign of an active embolization process. Previously, information has been absent on whether this high number of embolic events leads to additional minimal brain damage in patients with acute ischemic strokes.16
It is possible that a high number of MES may cause subclinical brain damage. For example, a postmortem brain study of patients without overt neurological deficit after recent cardiopulmonary bypass showed multiple small capillary and arteriolar dilatations, suggesting fat and air embolism.17
Controversy still exists concerning the duration of TCD monitoring necessary for emboli detection. We chose a period of 30 minutes, whereas other researchers have preferred a monitoring period of 15 to 60 minutes. In an extreme case, monitoring was performed for several hours.18 In our opinion, 30 minutes represents a well-chosen time frame, especially in patients with acute stroke, in whom the tolerability of the procedures for the patient is an important factor. Other reasons for a 30-minute monitoring period are the increase in artifact signals as the monitoring time is prolonged and the time consumption of this investigation (the observers must listen to the signal with a constant level of attention). These reasons have led many other groups to use a 30-minute monitoring period.9 10 19 20 In this discussion of timing, it must be kept in mind that each monitoring period represents only a snapshot of the cerebral circulation, regardless of the length of the chosen monitoring interval. Our data for repetitive ultrasound examinations support the need for easily tolerated and methodologically simple long-term monitoring equipment; in 19% of our patients with MES, these signals first occurred not in the initial but instead in the second or third TCD examination. For this reason, we suggest that MES in our patients did not occur at a constant frequency.
The potential embolic activity of carotid artery stenosis increases with the degree of the stenosis. This phenomenon has been demonstrated by others.21 22 High-grade carotid artery stenoses not only represent a hemodynamic problem but also, according to our data, involve a high risk of artery-to-artery embolism.23
It is well known that in the distal part of the carotid artery, occlusions may be a potential embolic source.24 25 With MES detection, we were able to demonstrate a high incidence of embolic events distal from an ipsilateral carotid artery occlusion. For this reason, microemboli detection may help to decide how long anticoagulation treatment in patients with carotid artery occlusion should be continued. In our opinion, patients without MES in the three examinations do not need further heparin treatment, whereas in patients with MES, heparin treatment should be prolonged. Our data also support the supposition that stroke mechanisms in acute carotid artery occlusions are not merely hemodynamic but also embolic.
Cardioembolic ischemic strokes are one of the major complications in patients with potential sources of embolism. From 15% to 30% of ischemic strokes are of cardiac origin. MES occurrence has been reported in this subgroup.26 Tong et al19 found an association between the presence of MES and known cardioembolic risk factors. They also suggested a relationship between microemboli and a history of prior stroke. Bilateral MES occurrence appears to provide strong evidence for a cardioembolic source in patients with acute stroke. The method of bilateral TCD monitoring may help to discriminate between cardioembolic and other sources of the ischemic stroke. Nevertheless, a high percentage of patients with a cardioembolic disease present with unilateral MES. Based on the present study, the most obvious reason is the relatively short monitoring time. Reasons for this study limitation were discussed above. Collection of dynamic data such as MES is very vulnerable in terms of fluctuation of MES frequency. Only well-tolerable monitoring procedures applied over hours may improve this situation in the future.
Bilateral or contralateral MES to an unilateral carotid artery stenosis could be caused by aortic plaques not diagnosed with echocardiography. Plaques in the aorta ascendance were not systematically assessed in our study. In one patient with a carotid artery stenosis and contralateral microemboli, a stenosis of the MCA ipsilateral to the MES was probably responsible for the microemboli signals. Contralateral MES in patients with a carotid artery stenosis are a reason for invasive and extensive workup in a search for embolic sources.
Because several patients presented with multiple embolic sources (ie, with embolic sources in the heart as well as in the carotid artery system), bilateral TCD monitoring of the MCAs may be only partially successful in localizing the embolic source. A useful further step would be dual-probe recording on the common carotid arteries and MCAs; central (cardiac or aortic) embolic sources produce emboli at both common carotid arteries, but internal carotid artery stenoses produce only intracranial signals.
Astonishingly, a high number of patients without evident embolic sources presented with MES. MES in this subgroup could be due to undiagnosed plaques in the aorta ascendance or in the intracranial part of the carotid artery or to intercurrent arrhythmia not detected by Holter monitoring. MES in these patients should be a reason for expanding the search for embolic sources to include invasive diagnostic procedures.
Heparin treatment in patients with progressive ischemic stroke appears to have a beneficial influence on the prognosis.27 Anticoagulation therapy with intravenous heparin may therefore decrease the prevalence of MES. Ries et al28 reported a decrease in MES frequency in patients with symptomatic carotid artery stenosis who were receiving intravenous heparin treatment. In contrast, Forteza et al13 detected no significant difference between the MES-positive and MES-negative groups with regard to heparin and aspirin intake. Sliwka et al26 did not find a correlation between the level of anticoagulation with heparin or phenprocoumon in patients with potential cardiac sources of embolism and MES. The effect of anticoagulation treatment on the occurrence of MES was investigated extensively in the population of patients with mechanical heart valve prostheses. None of the researchers could find a correlation between MES and the anticoagulation level.8 29 30 31 In our population of patients with stroke of different origins, there was evidence that intravenous heparin may have an possible inhibitory effect on the prevalence of MES. This trend was clear but did not reach statistical significance. We suspect that the relation between these medications and MES occurrence has yet to be fully elucidated. Further studies assessing individual patients longitudinally with the use of longer study periods, TCD monitoring times over hours, and repetitive analysis of anticoagulation parameters may solve the problem of whether intravenous heparin has an effect on MES prevalence in patients with acute stroke.
The CT data in our study are in accordance with those of Grosset et al,20 who could also demonstrate that MES are common in patients with acute stroke, with the notable exception of lacunar stroke. The few MES in our population of patients with small vessel lesion on CT arises probably from coincidental atherosclerotic disease, which is known to occur in this patient group.
In conclusion, we believe that TCD monitoring is a helpful noninvasive method for investigating the underlying pathology of an acute cerebral ischemic event. We were able to demonstrate that MES and acute ischemic events are, to a large extent, associated. We therefore suggest that repetitive MES occurrence may be one marker for increased risk for recurring stroke. MES detection may help to discriminate among different embolic mechanisms. Furthermore, ultrasound monitoring may help to reveal patients who are at higher risk for recurring ischemic events. Repetitive TCD examinations may improve our understanding of the dynamic process of cerebral ischemia. In the future, this method may be useful for monitoring the effect of preventive drugs in acute stroke trials by measuring the number of embolic signals to the brain. Further progress in ultrasound technology should provide easy-to-use and well-tolerated monitoring equipment. Therefore, it is hoped that long-term ultrasound monitoring studies over a period of hours will be possible, yielding more information on stroke patients for better treatment and prophylaxis.
Selected Abbreviations and Acronyms
|MCA||=||middle cerebral artery|
|PFO||=||patent foramen ovale|
For helpful suggestions, we thank Dr Stuart Fellows, Department of Neurology, RWTH Aachen. For statistical support, we thank Prof Klaus Willmes v. Hinckeldey. This work forms part of the doctoral theses of Ariane Lingnau and Wolf-Dirk Stohlmann.
Reprint requests to Dr Ulrich Sliwka, MD, Klinikum der Friedrich-Schiller-Universität Jena, Klinik für Neurologie, Philosophenweg 3, D-07740 Jena, Germany.
- Received July 17, 1996.
- Revision received November 12, 1996.
- Accepted November 12, 1996.
- Copyright © 1997 by American Heart Association
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.
Russell D, Madden P, Clark WM, Sandset PM, Zivin JA. Detection of arterial emboli using Doppler ultrasound in rabbits. Stroke. 1991;22:253-258.
Markus HS, Brown MM. Differentiation between different pathological cerebral embolic materials using transcranial Doppler in an in vitro model. Stroke. 1993;24:1-5.
Tegeler CH. High-intensity transient signals detected by Doppler ultrasonography: searching for answers. Cerebrovasc Dis. 1994;4:379-382.
Consensus Committee of the Ninth International Symposium on Cerebral Hemodynamics. Basic identification criteria of microembolic signals. Stroke. 1995;26:1123.
Pugsley W. The use of Doppler ultrasound in the assessment of microemboli during cardiac surgery. Perfusion. 1989;4:115-122.
Russel D, Madden KP, Clark WM, Sabdset PM, Zivin JA. Detection of arterial emboli using Doppler ultrasound in rabbits. Stroke. 1991;22:253-258.
Sliwka U, Diehl RR, Meyer B, Schöndube 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.
Markus H. Transcranial Doppler detection of circulating cerebral emboli. Stroke. 1993;24:1667-1672.
Grosset DG, Georgiadis D, Kelman AW, Lees KR. Quantification of ultrasound emboli signals in patients with cardiac and carotid disease. Stroke. 1993;24:1922-1924.
Georgiadis D, Goeke J, Hill M, König M, Navabi DG, Stögbauer F, Zunker P, Ringelstein EB. A novel technique for identification of Doppler microembolic signals based on the coincidence method. Stroke. 1996;27:683-686.
Tegeler CH, Burke GL, Dalley GM, Stump DA. Carotid emboli predict poor outcome in stroke. Stroke. 1993;24:186. Abstract.
Forteza AM, Babikian VL, Hyde C, Winter M, Pochay V. Effect of time and cerebrovascular symptoms on the prevalence of microembolic signals in patients with cervical carotid stenosis. Stroke. 1996;27:687-690.
Siebler M, Sitzer M, Rose G, Bendfeldt D, Steinmetz H. Silent cerebral embolism caused by neurologically symptomatic high-grade stenosis. Brain. 1993;116:1005-1015.
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.
Norris JW, Zhu CZ. Silent stroke and carotid stenosis. Stroke. 1992;23:483-485.
Tong DC, Bolger A, Albers GW. Incidence of transcranial Doppler-detected cerebral microemboli in patients referred for echocardiography. Stroke. 1994;25:2138-2141.
Grosset DG, Georgiadis D, Abdullah Ibone I, Lees KR. Doppler emboli signals vary according to stroke subtype. Stroke. 1994;25:382-384.
Valton L, Larrue V, Arrue P, Geraud G, Bes A. Asymptomatic cerebral embolic signals in patients with carotid stenosis. Stroke. 1995;26:813-815.
Siebler M, Kleinschmidt A, Sitzer M, Steinmetz H, Freund HJ. Cerebral microembolism in symptomatic and asymptomatic high-grade internal carotid artery stenosis. Neurology. 1994;44:615-618.
Ringelstein EB, Zeumer H, Angelou D. The pathogenesis of strokes from internal carotid artery occlusion: diagnostic and therapeutical implications. Stroke. 1983;14:867-875.
Delcker A, Diener HC, Wilhelm H. Source of cerebral embolic ischemia in occlusion of internal carotid artery. Stroke. 1996;27:798. Abstract.
Sliwka U, Job FP, Wissuwa D, Diehl RR, Flachskampf FA, Hanrath P, Noth J. Occurrence of transcranial Doppler high-intensity transient signals in patients with potential cardiac sources of embolism. Stroke. 1995;26:2067-2070.
Dahl T, Sandset PM, Abildgaard U. Heparin treatment in 52 patients with progressive ischemic stroke. Cerebrovasc Dis. 1994;4:101-105.
Ries S, Schminke U, Daffertshofer M, Schindlmayr C, Hennerici M. High intensity transient signals and carotid artery disease. Cerebrovasc Dis. 1995;5:124-127.
Georgiadis D, Grosset DG, Kelman A, Faichney A, Lees KR. Prevalence and characteristics of intracranial microemboli signals in patients with different types of prosthestic cardiac valves. Stroke. 1994;25:587-592.
Sturzenegger M, Beer JH, Rihs F. Monitoring combined antithrombotic treatments in patients with prosthetic heart valves using transcranial Doppler and coagulation markers. Stroke. 1995;26:63-69.