High-Intensity Transient Signals in Patients With Cerebral Ischemia
Background and Purpose This study investigated the diagnostic relevance of transcranial Doppler monitoring for the detection of high-intensity transient signals (HITS) in patients with recent cerebral ischemic events of various origins.
Methods We prospectively performed bilateral transcranial Doppler monitoring (mean, 48±8 minutes) from both middle cerebral arteries in 280 patients with an acute (<4 weeks) cerebral ischemic event in the middle cerebral artery territory and in 118 asymptomatic control subjects. HITS were analyzed according to a standardized protocol.
Results Symptomatic patients had a significantly (P<.001) higher rate (9.3%) of HITS than asymptomatic control subjects (1.7%). Patients with reversible ischemia (4.2%) and patients with suspected small-vessel syndromes (4.5%) had lower rates of HITS (P<.05) than those with large-vessel territory strokes (14.2%). Brain imaging (CT/MRI) results corresponded with this observation: the occurrence of HITS was significantly higher (P<.001) in patients with a pattern of large-vessel territorial brain infarction (19.0%) than in those with lacunar lesions (0%) or unidentified ischemic lesions (3.4%). Patients with identified sources of potential embolism (12.9%) had HITS (P<.001) more frequently than those without (0%). Patients with cardiac sources of embolism (excluding artificial heart valves) showed fewer HITS (6.2%) than patients with vascular sources (17.1%).
Conclusions The results indicate that HITS occur predominantly in patients with large-vessel territory stroke patterns and persisting deficits that are most likely due to artery-to-artery or cardiogenic embolism. In contrast, patients with small-vessel disease and rapid recovery only occasionally present with HITS. Thus, the detection of HITS may substantially support the classification of the individual pathogenesis of cerebral ischemia, particularly when multiple risk constellations for stroke coexist.
Since the first occasional Doppler observations of HITS in the 1960s during open heart surgery,1 in patients with decompression sickness,2 and in patients undergoing angiography,3 experimental research to elucidate the significance of HITS4 5 has shown a variety of structural components, including gaseous particles within the arterial blood stream passing the Doppler beam, that might cause HITS.4 5 Although methods to differentiate the constituents producing HITS are not yet available,5 it is likely that they represent a combination of platelet aggregates and thrombotic and atherosclerotic material arising from atherosclerotic arterial6 7 8 9 10 and cardiac11 12 13 14 sources. Clinical interest in HITS detection has increased rapidly since HITS have been interpreted to reflect clinically silent microembolism of potential prognostic relevance. Previous data on the incidence and frequency of HITS have been conflicting (Table 1⇓), and many studies have focused on either highly selected patients with defined sources of embolism or groups of patients presenting with a broad spectrum of stroke origins.6 8 11 13 15 16 17 18 Most of these studies used a cross-sectional design10 16 (excluding a recent report that used follow-up data19 ) to estimate the predictive value of HITS for ischemic events. The utility of HITS detection in the evaluation of stroke pathophysiology in patients with variable sources of embolism, structural lesions, and clinical syndromes of ischemic stroke is currently unclear. We therefore undertook a prospective study to clarify this issue.
Subjects and Methods
The occurrence of HITS was investigated in patients with acute onset of signs and symptoms indicating either large-vessel infarction or small-vessel disease in the MCA territory. Control subjects were defined as patients without signs and symptoms indicating large-vessel infarction or small-vessel disease but with a similar distribution of potential sources of cerebral embolism (Table 2⇓). All patients and control subjects underwent a full clinical and diagnostic workup, which included neurological and general physical examination; acute CT/MRI and follow-up CT/MRI (3 to 7 days); noninvasive vascular investigations, including extracranial and transcranial Doppler sonography as well as color-coded duplex scanning; and cardiac investigations consisting of clinical examination, electrocardiogram, Holter monitoring, and transthoracic/transesophageal echocardiography. Additional diagnostic information was taken into account if available (eg, MR angiography or conventional angiography).
The occurrence and frequency of HITS in patients with a recent (<4 weeks) acute cerebral ischemic event of the MCA territory were analyzed in relation to (1) clinical symptomatology, (2) structural ischemic brain lesions, and (3) cardiac and vascular sources of embolism. Patients and control subjects were assigned to these various groups, as follows.
Subgroups Determined by Clinical Symptomatology
Symptomatic patients were categorized as those with signs or symptoms indicating large- or small-vessel infarction in the MCA territory.20 21 22 The control group consisted of clinically asymptomatic subjects.
Subgroups Determined by Structural Ischemic Brain Lesions
Brain imaging patterns were categorized as acute territorial stroke patterns indicating large-vessel disease23 ; acute hemodynamic stroke patterns (cortical or subcortical) indicating large-vessel disease23 ; lacunes (<1.5 cm in diameter) and subcortical white matter lesions indicating small-vessel disease23 ; and no ischemic lesion.
Subgroups Determined by Cardiac and/or Vascular Source of Embolism
Vascular source of embolism24 ipsilateral to the side of the ischemic event included carotid stenosis with ≥70% lumen obstruction including occlusions (mainly of atherosclerotic origin, occasionally due to dissection); carotid stenosis with <70% lumen obstruction; and other vascular source of embolism (eg, stenosis of the intracranial arteries and atherosclerotic lesions of the aortic arch).
Cardiac source of embolism included valvular dysfunction excluding artificial heart valves (since HITS originating from artificial heart valves25 have been reported to reflect other phenomena, eg, microcavitations different from structural embolic particles); intracardiac thrombi; relevant arrhythmia (eg, atrial fibrillation and atrial flutter with or without intermittent sinus rhythms); and other sources of cardiac embolism (eg, cardiac aneurysm, relevant wall hypokinesia, dilatative cardiomyopathy).
An additional subgroup included those with combinations of vascular and cardiac diseases potentially serving as sources of embolism.
Flow velocity monitoring was performed with the use of (1) a single-gate transcranial pulsed Doppler system in 164 patients and 74 control subjects (TCD7, DWL) and (2) a new multigate transcranial pulsed Doppler system26 in 136 patients and 44 control subjects (Multidop X, DWL). Both MCAs were insonated simultaneously with a 2-MHz probe through the transtemporal window at an insonation depth of 50 to 56 mm.24 In multigate Doppler examinations, an intergate distance of 5 mm was used. Registrations were performed with the patient in a supine position with monitoring of blood pressure and heart rate. Doppler signals were registered for 30 to 60 minutes (mean registration time, 48.2±8.1 minutes). The Doppler signals were analyzed acoustically by an experienced investigator on-line and by another independent investigator off-line.
Differentiation of HITS from artifacts was based on criteria established during recent consensus conferences.27 28 All signals with a typical crisp sound and a signal intensity of >4 dB spectral broadening intensity were evaluated. In these cases, the contralateral Doppler spectra were taken into account (Figure⇓). Artifacts were assumed if signals were registered simultaneously on both sides or were bidirectional. With the multigate technology, only short unilateral signals showing intergate latencies were classified as HITS.
We used the Student's t test to compare group means and the χ2 test to compare group proportions. However, in case of unacceptably low cell counts, Fisher's exact test was used. The quantitative risk measure (OR) was determined by the Mantel-Haenszel procedure. To analyze the relationship between a dependent variable and multiple independent categorical variables, we used a logistic regression analysis. The level for significance was set at P<.05. The calculations were completed with the use of statistical software packages (SAS and BMDP).
HITS in Asymptomatic Control Subjects
We found HITS in 2 of 118 clinically asymptomatic control patients (1.7%). Five control subjects (4.2%) had lacunar lesions with or without white matter lesions, whereas 113 (95.8%) showed no structural ischemic deficits. Forty-six control subjects (39.0%) had a potential source of embolism: 31 had an asymptomatic vascular source of embolism (10 with carotid stenosis with <70% lumen reduction, 12 with carotid stenosis with ≥70% lumen reduction, 3 with carotid occlusion, and 6 with intracranial internal carotid artery stenosis or dissection); 11 control subjects had an asymptomatic cardiac source of embolism; and 4 control subjects had vascular and cardiac embolic sources. In contrast, 72 control subjects (61.0%) had no identifiable source of embolism. Two control subjects (1.7%) showed HITS: 1 with a vascular source of embolism (significant carotid stenosis) and 1 with a cardiac source of embolism (atrial fibrillation). Brain imaging in control subjects showing HITS was normal. The incidence of HITS was significantly (OR, 5.4; 95% CI, 2.3 to 12.8; P<.001) lower in control subjects (1.7%) than in patients (9.3%).
HITS Related to Clinical Syndromes
Of a total of 280 patients, 141 (50.4%) patients presented with symptomatology suggestive of large-vessel disease, 67 (23.9%) patients showed syndromes indicating small-vessel disease, and 72 (25.7%) patients reported reversible ischemic symptoms and were therefore difficult to classify. The occurrence of HITS was significantly (P<.017) higher in patients with large-vessel syndromes (14.2%) than in those with lacunar syndromes (4.5%) or reversible ischemia (4.2%) (Table 3⇓).
HITS Related to Sources of Embolism
Differentiating patients with cerebral ischemia on the basis of the potential origin of embolism (Table 4⇓), we found no HITS in patients without an identified source of embolism (n=78), whereas patients with a potential source of embolism (n=202) showed HITS in 12.9% (OR, 21.4; 95% CI, 4 to 114.5; P<.001). In patients with vascular sources of embolism (n=105), HITS occurred in 17.1%. Patients with cardiac sources of embolism (n=65) showed HITS in 6.2% (P=.026), and patients with both vascular and cardiac sources of embolism (n=32) generated HITS in 12.5% (Table 4⇓). Comparison between the different types of embolic sources within the vascular and cardiac subgroups showed no statistical differences (Table 4⇓). The frequency of HITS was 4.0±3.9/30 min in patients with vascular sources of embolism and 19.1±20.4/30 min in patients with cardiac sources of embolism. Thirty-one patients with a vascular source of embolism were found to have a contralateral carotid atherosclerotic lesion; however, HITS related to these asymptomatic lesions were not detected.
HITS Related to Brain Imaging
When patients were analyzed on the basis of brain imaging studies (CT, MRI, or both), 125 patients presented with an acute ischemic lesion suggesting large-vessel disease (territorial infarction, n=116; hemodynamic stroke pattern, n=9). Ninety-three patients showed single or multiple lacunes indicating some form of microangiopathic lesions (78 with additional white matter lesions on CT/MRI scans). Four patients had an acute territorial infarction combined with lacunar lesions. Patients with reversible ischemic events (n=72) showed CT/MRI scans without identifiable ischemic lesions in 58 cases, lacunar lesion(s) in 11 cases, and a clinically silent territorial infarction in 3 cases. Comparing the initial clinical classification to brain imaging patterns, we found the expected association (84.1%) between clinical classification of large-vessel disease (OR, 31.4; 95% CI, 15.5 to 63.7; P=.001) and brain imaging patterns of territorial or hemodynamic infarction.
Patients with territorial infarctions had a significantly (OR, 44.5; 95% CI, 2.7 to 744.7; P=.001) higher rate of HITS (19.0%) than patients with lacunar lesions (0%) and patients without identifiable structural ischemic lesions (3.4%) (Table 5⇓). The highest incidence of HITS was found in patients with large-vessel syndrome, vascular source of embolism, and territorial infarction in follow-up brain imaging (22.7%). We analyzed clinical classification, identification of source of embolism, and occurrence of HITS for their accuracy in predicting the development of an embolic territorial or hemodynamic stroke pattern versus a microangiopathic or lacunar lesion in brain imaging using a logistic regression analysis with maximum likelihood estimation. Clinical classification showed a high accuracy of 84.1%. The differentiation between an embolic versus a lacunar stroke mechanism could be increased to an accuracy of 88.2% when the identification of source of embolism and particularly the occurrence of HITS were taken into account.
HITS Related to Therapy
The therapeutic regimen at the time of the investigation was analyzed to establish its effect on incidence of HITS. Patients without a specific treatment, ie, prophylactic subcutaneous heparin (5000 IU TID) only or no treatment, showed a nonsignificant trend for a higher incidence of HITS than patients who were treated with sufficient anticoagulation (patients with vascular source of embolism, 20.9% versus 10.8%; patients with cardiac source of embolism, 8.6% versus 3.6%) (Table 6⇓). Only a small number of patients (insufficient for statistical analysis) were receiving platelet aggregation inhibitor at the time of the Doppler investigation.
Interest in the phenomenon of HITS has increased significantly in recent years because of the assumption that detection of HITS enables detection of ongoing embolic events in humans. Despite great enthusiasm, the exact nature of HITS in the individual patient remains unclear. For example, an increased risk of cerebral ischemia in patients with artificial heart valves who characteristically show a high frequency of HITS (>40 000/d29 ) has been expected.30 However, this contention could not be verified,18 thus negating specific therapeutic implications.31 Signal pattern analysis suggests that these HITS represent a homogeneous phenomenon suggestive of gaseous particles generated by a cavitation mechanism.11 26 However, the situation is more complex in the majority of patients with other potential sources of (cardiac and vascular) embolism.
HITS are only exceptionally associated with clinical signs or symptoms of cerebral ischemia. In accordance with others,29 we observed HITS only in single cases to be followed by immediate focal cerebral ischemia during routine transcranial Doppler monitoring in patients with cardiac or vascular sources of embolism. This can easily be explained by the fact that single HITS most likely reflect very small particles (10 μm in diameter for gaseous bubbles and 20 to 100 μm for solid material)32 but rarely represent emboli large enough to cause critical occlusion of even small-caliber distal arterial segments.
The data presented here support the potential clinical significance of HITS in the pathogenesis of recent cerebral ischemic events in patients with clinical and brain imaging signs indicating large-vessel disease. This, however, differs for patients with small-vessel disease or recent transient ischemic attacks, despite transient ischemic episodes in patients with a vascular source of embolism, who showed HITS in 2 of 21 cases (9.5%). This hypothesis is supported by the observation that HITS are more frequent in symptomatic patients, particularly those with clinical nonlacunar stroke, large-vessel stroke pattern in brain imaging, and a potential source of embolism. Further evidence arises from the fact that in the subgroup of patients with bilateral vascular sources of embolism, HITS could not be detected on the contralateral asymptomatic side.33 We could demonstrate that the occurrence of HITS in patients with cerebral ischemia is closely linked to the presence of sources of embolism. Particularly in patients with acute cerebral ischemia in the presence of a source of embolism, HITS occurred most frequently in patients with “major” stroke and structural (large-vessel) deficit in brain imaging.
Regarding the frequency of HITS, our results confirm recent observations by Markus et al,17 32 who, in a similar series excluding patients with artificial heart valves, found 2.2 HITS per 20 minutes in patients with carotid disease. Different criteria for detection of HITS, problems of artifact suppression, and overestimation of signals in semiautomated systems may be responsible for somewhat higher frequencies in other studies.16 We used two different technologies: single-gate and multigate. In a series of 10 patients investigated with both single-gate and multigate Doppler, we found a high agreement (97.6%) between both methods concerning occurrence and frequency of HITS. However, an advantage of the new multigate technology is a direct estimate of the dynamic properties of embolic material passing through the arterial lumen,26 which is a further guarantee for artifact minimization and enables a simple and safe diagnosis of HITS. Thus, we decided to continue the prospective clinical study with the multigate Doppler. We found no significant differences between these methods when we analyzed the data set.
Most previous studies discussed the occurrence of HITS as a possible predictor of further ischemic events. This can only be determined by follow-up studies.19 In contrast, the primary goal of our study was to analyze the value of HITS detection in the diagnostic workup to clarify the pathophysiological mechanism of a focal ischemic event, particularly in patients presenting with multiple potential causes of cerebral ischemia. This distinction, although often difficult in the acute setting, particularly when results of brain imaging are noncontributory, may also have therapeutic implications, since patients with embolic stroke may benefit from neuroprotective and fibrinolytic treatment more than patients with small-vessel disease. The significant association between HITS and embolic stroke in this study underscores the importance of transcranial Doppler monitoring of acute stroke patients in clinical decision making.
Selected Abbreviations and Acronyms
|HITS||=||high-intensity transient signals|
|MCA||=||middle cerebral artery|
This study was supported in part by the Deutsche Forschungsgemeinschaft, Forschergruppe (KU/258-1), and by the Forschungsfond Fakultät Mannheim, University Heidelberg 68/95. We are grateful to P. Kuprion for assistance with the ultrasound investigations. We are also indebted to Dr H.-P. Altenburg for advice regarding statistical methodology and particularly to Dr S. Meairs and Ms Maria Garcia for their editorial comments.
Reprint requests to Michael Daffertshofer, MD, Department of Neurology, University Heidelberg, Klinikum Mannheim, Theodor-Kutzer-Ufer, 68135 Mannheim, Germany. E-mail firstname.lastname@example.org.
- Received February 26, 1996.
- Revision received May 23, 1996.
- Accepted June 13, 1996.
- Copyright © 1996 by American Heart Association
Rautenberg W, Schwartz A, Hennerici M. Transkranielle Dopplersonographie während der zerebralen Angiographie. In: Widder B, ed. Transkranielle Dopplersonographie bei zerebrovaskulären Erkrankungen. Heidelberg, Germany: Springer-Verlag; 1987:144-148.
Russell D, Madden KP, 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.
Lash S, Newell D, Spence A, Douville C, Byrd S, Winn HR. Artery-to-artery cerebral emboli detection with transcranial Doppler: analysis of eight cases. J Stroke Cerebrovasc Dis. 1993;3:15-22.
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.
Ries S, Bäzner H, Rautenberg W, Hennerici M. Emboliesuche mit trankraniellen Dopplersonographie: Entscheidungshilfe bei der Identifikation zur Karotischirugie. Vasa. 1993;42(suppl):84. Abstract.
Ries S, Schminke U, Daffertshofer M, Hennerici M. Emboli detection by TCD in patients with cerebral ischemia. Cerebrovasc Dis. 1994;4:20. Abstract.
Siebler M, Sitzer M, Rose G, Bendfeldt D, Steinmetz H. Silent cerebral embolism caused by neurologically symptomatic high-grade carotid stenosis: event rates before and after carotid endarterectomy. Brain. 1993;116:1005-1015.
Georgiadis D, Grosset DG, Kelman AW, 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.
Braekken SK, Russell D, Brucher R, Svenning J. Cerebral embolus monitoring during open heart surgery. Stroke. 1994;25:745. Abstract.
Georgiadis D, Kaps M, Siebler M, Hill M, König M, Berg J, Kahl M, Zunker P, Diehl B, Ringelstein EB. Variability of Doppler microembolic signal counts in patients with prosthetic cardiac valves. Stroke. 1995;26:439-442.
Sliwka U, Job FP, Wissuwa D, Diehl RR, Flachskamp FA, Hanrath P, Noth J. Occurrence of transcranial Doppler high-intensity transient signals in patients with potential cardiac source of embolism. Stroke. 1996;26:2067-2070.
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.
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.
Grosset DG, Georgiadis D, Abdullah I, Bone I, Lees KR. Doppler emboli signals vary according to stroke subtype. Stroke. 1994;25:382-384.
Siebler M, Nachtmann A, Sitzer M, Rose G, Kleinschmidt A, Rademacher J, Steinmetz H. Cerebral microembolism and the risk of ischemia in asymptomatic high-grade internal carotid artery stenoses. Stroke. 1995;26:2184-2186.
Hennerici M, Daffertshofer M. Patterns of motor dysfunction after stroke. In: Fisher M, Bogousslavsky J, eds. Current Review of Cerebrovascular Disease. 2nd ed. Philadelphia, Pa: Current Medicine; 1995:93-106.
Neuerburg-Heusler D, Hennerici M. Gefässdiagnostik mit Ultraschall: Doppler- und farbkodierte Duplexsonographie der grossen Körperarterien und -venen. 2nd ed. Stuttgart, Germany: Thieme; 1995.
Georgiadis D, Mackay TG, Kelman AW, Grosset DG, Wheatley DJ, Lees KR. Differentiation between gaseous and formed embolic materials in vivo: application in prosthetic heart valve patients. Stroke. 1994;25:1559-1563.
Hennerici M. Cerebral embolism. Cerebrovasc Dis. 1995;5:69.
Consensus Committee of the Ninth Cerebral Hemodynamic Symposium. Basic identification criteria of Doppler microembolic signals. Stroke. 1995;26:1123.
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, Mallinson A, Grosset DG, Lees KR. Coagulation activity and emboli counts in patients with prosthetic cardiac valves. Stroke. 1994;25:1211-1214.
Russell D, Braekken SK, Brucher R, Svennevig J. Automatic cerebral embolus detection. Cerebrovasc Dis. 1994;4:235. Abstract.
Markus HS, Droste DW, Brown MM. Ultrasonic detection of cerebral emboli in carotid stenosis. Lancet. 1993;341:1606.
Ries S, Valikovics A, Daffertshofer M, Hennerici M. The usefulness of a multigate detection of high-intensity transient signals for the analysis of cerebral microembolism. Neurology. 1996;46:388-389.