- Email: [email protected]
Detection of high intensity transient signals (HITS): How and why?
European Journal of Ultrasound 7 (1998) 23 – 29
Review
Detection of high intensity transient signals (HITS): How and why? Dirk W. Droste *, E. Bernd Ringelstein Klinik und Poliklinik fu¨r Neurologie der WWU Mu¨nster, Albert-Schweitzer-Str. 33, D-48129 Mu¨nster, Germany
Abstract HITS (high intensity transient signals) in transcranial Doppler recordings reflect either microemboli, both gaseous and solid, or artefacts. Various frequencies in number of microembolic signals (MES) have been reported in the same clinical condition. A possible explanation for these discrepancies may be different device settings and algorithms for embolus detection. For reproducibility of data, we suggest that studies on MES report the following parameters: (1) Ultrasound device; (2) transducer type; (3) insonated artery; (4) insonation depth; (5) algorithms for signal intensity measurement; (6) scale settings; (7) detection threshold; (8) axial extension of sample volume; (9) fast Fourier transform (FFT) size (number of points used); (10) FFT length (time); (11) FFT overlap; (12) transmitted ultrasound frequency; (13) high pass filter settings; and (14) recording time. No current system of automatic embolus detection has the full sensitivity and specificity required for clinical use. Therefore, each of the signals detected by these devices needs to be checked and verified by an experienced investigator. MES will help to identify the site and activity of the embolizing lesion. Microembolus detection might reduce the observation time and the number of patients needed to perform interventional trials. First, however, MES needs to be validated as a meaningful prognostic parameter. Microemboli originating from prosthetic cardiac valves are mainly gaseous. Therefore, they cannot serve as an indicator of the valves thromboembolic activity or the patient’s stroke risk. © 1998 Elsevier Science Ireland Ltd. All rights reserved. Keywords: High intensity transient signals; Transcranial Doppler; Prosthetic cardiac values; Stroke
1. Introduction
* Corresponding author. Tel.: + 49 251 8348176; fax: + 49 251 8348181.
Emboli, such as thrombus and atheroma, originating from carotid and aortic plaques or cardiac sources are generally accepted as a main cause for cerebral ischemic events in humans. In 1990,
0929-8266/98/$19.00 © 1998 Elsevier Science Ireland Ltd. All rights reserved. PII S 0 9 2 9 - 8 2 6 6 ( 9 8 ) 0 0 0 1 0 - X
24
D.W. Droste, E.B. Ringelstein / European Journal of Ultrasound 7 (1998) 23–29
Fig. 1. Example of a MES in a patient with a carotid artery stenosis.
Spencer et al. (1990) described that clinically asymptomatic circulating cerebral microemboli in patients with carotid artery stenosis could be detected by transcranial Doppler ultrasound (TCD). Fig. 1 gives an example of such a microembolic signal (MES). Although these microemboli are clinically silent, they might be clinically important by indicating an increased risk of stroke. High intensity transient signals (HITS) in transcranial Doppler recordings reflect either gaseous or solid microemboli, or artefacts.
2. Methodological considerations The detection of microembolic particles within the streaming blood is based on the measurement of the backscatter from the emboli. The diffuse backscatter of the ultrasound from the normal flowing blood (including transient erythrocyte aggregates) is usually much lower than the backscatter from solid emboli. The latter, however, is usually much lower than the backscatter from gaseous ones. At present, no conclusion as to the composition and the size of an embolus can be drawn from the embolus’ echo (Russell, 1992). Our group has shown in two studies, that mi-
croemboli originating from prosthetic cardiac valves are mainly gaseous (Droste et al., 1997b; Kaps et al., 1997). Therefore, they cannot serve as an indicator of the valves thromboembolic activity or the patient’s stroke risk. Strongly variable frequencies in number of MES have been reported in the same clinical condition. One explanation for this discrepancy may be the different device settings and variable algorithms used for embolus detection. Technical parameters affecting the detectability of MES are; (1) The relationship between the backscattered power from emboli and that from the blood, i.e. the relative intensity increase; (2) the detection threshold; (3) the size of the sample volume; (4) the fast Fourier transform (FFT) frequency resolution; (5) the FFT temporal resolution; (6) the FFT temporal overlap; (7) the instrumentation’s dynamic range; (8) the transmitted ultrasound frequency; (9) the filters’ settings; and (10) the recording time. The settings of the ultrasound instrumentation strongly influences the detectability of MES (Droste et al., 1994; International Consensus Group on Microembolus Detection, 1998). The investigator should be aware of these effects. Some of these aspects will be addressed later in more detail. It is essential to maintain parameters constant throughout and between
D.W. Droste, E.B. Ringelstein / European Journal of Ultrasound 7 (1998) 23–29
recordings, and to parallel settings in multicentre studies.
2.1. Relati6e intensity increase At present, the relative intensity increase of the embolic signal is measured in different ways. Different types of signal analysis are used in the different devices, and can additionally be modified by the user. The relative intensity increase of the embolic signal is usually measured in decibels (dB). In the frequency domain-based analysis, for instance, the peak intensity of a microembolic event, or its mean within a defined time frame and frequency range, can be used. Similarly, the intensity of the background signal may be expressed as a mean value or a median value over variable time periods and frequency ranges, (e.g. at a comparable location as the embolic echo in the preceding cardiac cycle, or comprising arbitrarily defined time frames preceding the embolic event, or considering the whole sweep, even including the signal-free areas of the screen). Thus, for a given embolic signal, different decibel values of relative intensity increase can be calculated using different background and embolic signal intensity measurements. The user should be aware of which technique is used in the particular automated embolus detection systems he is working with; this should be specified. In the same way, manual techniques of calculating signal intensity should always be specified (Markus and Molloy, 1997). Some intensity calculations of the embolic signal and of the background depend on the frequency scale setting (pulse repetition frequency, PRF ), as more or less spectrum-free area is included. Thus, the scale setting should also be indicated and kept constant.
2.2. Detection threshold At present, the various manufacturers and investigators use greatly different parameters and criteria for identifying a short-lasting ultrasound event as microembolic in nature. Particularly, greatly different decibel (dB)-thresholds ranging from 3 to 9 dB have been recommended for discriminating MES from the general background
25
noise and from spontaneous, speckle-like intensity fluctuations of the physiological Doppler flow signals (Siebler et al., 1993; Consensus Committee of the 9th International Cerebral Hemodynamic Symposium, 1995; Markus et al., 1995). The situation is even more complicated because different algorithms are used to calculate the background noise and the intensity of the presumed microembolic signal as such. These technical aspects may have contributed to the striking discrepancies in the prevalence of MES described in the literature in various types of stroke or stroke-prone patients (Babikian et al., 1994; Grosset et al., 1994a,b; Braekken et al., 1995; Markus et al., 1995). There are two possible ways of determining the detection threshold of microemboli in decibels for a given device: either defining the range of spontaneous intensity fluctuations within the Doppler signals of normal controls, or defining fluctuations on a case-by-case basis during emboli-free periods (Markus et al., 1995; Droste et al., 1997a). It is not yet clear whether thresholds defined in the middle cerebral artery can be used for other intracranial arteries, or for poststenotic middle cerebral artery flow spectra as well. To the best of our present knowledge, calibration of individual machines by either normal controls or by intrapatient analysis of the background signal is equally valid. Each device should be individually calibrated, and the approach used must be clearly indicated. Fig. 2 gives an example of such a calibration. A higher detection threshold results in lower sensitivity, but higher specificity and higher intercentre agreement (Markus et al., 1996).
2.3. Artefact rejection Discrimination of true MES from artefacts, e.g. produced by probe displacement, is of crucial importance. Bidirectional signals, i.e. signals above and below the baseline, frequently represent artefacts. However, MES of high echo intensity may occasionally produce bidirectional signals, particularly if gaseous in nature or with inadequate settings of the instrumentation. Investigators new in this field are encouraged to produce artefact signals on purpose, to become
26
D.W. Droste, E.B. Ringelstein / European Journal of Ultrasound 7 (1998) 23–29
Fig. 2. The relative intensity increases of the Doppler speckle background in the absence of MES and of MES in patients with prosthetic cardiac valves and arterial embolic sources given as a percent. A detection threshold of 5 dB and more was chosen for the device (TC 4040, Nicolet-EME) and the setting used. Note that MES from prosthetic heart valves due to their predominantly gaseous composition have a higher relative intensity increase.
familiar with their characteristics. The multigatetechnique (see below) uses sampling of echoes from different depths of the same artery. This allows to trackle the migration of echoes within the flowing blood, i.e. their sequential appearance from the more proximal to the more distal sample volumes. By contrast, an artefact affects all channels simultaneously. Figs. 3 and 4 give examples of an embolic signal and an artefact recorded with the multigate technique. The TCD devices currently available are not yet able to automatically discriminate artefacts from microemboli with satisfying reliability.
2.4. Documentation and quality control At present, the most widely used documentation system is the recording of the pre-FFT audio signal ( = raw data) on digital audio tapes (DAT)
(Bush and Evans, 1993). This allows the data to be subjected to quality control and the re-evaluation of regions of interest. It also allows for off-line analysis. For scientific purposes, observer bias can be avoided by a blinded analysis of the audio tapes by different observers. It is important to ensure reproducibility both between and within centres in the identification of MES. For inter-observer studies, it is important to guarantee that observers select the same MES. A statistical method which encounters this is required (e.g. probability of specific agreement, rather than counting the total number of emboli recorded by each observer) (Markus et al., 1996). Exchange and analysis of data among centres is encouraged. For multicentre studies, the use of identical devices and a standardised, identical setting of the equipment is strongly recommended.
D.W. Droste, E.B. Ringelstein / European Journal of Ultrasound 7 (1998) 23–29
Fig. 3. The left template shows an embolic signal within the Doppler spectrum. On the right side, the pre-FFT (fast Fourier form) data are given for all the four channels. The x-axis is stretched to better demonstrate the time delay between the four channels.
2.5. Automatic embolus detection Currently, most centres record the Doppler signal on digitalised audio tapes and evaluate these tapes acoustically and visually off-line and blinded to the clinical diagnosis. A recording time of 1 h has been shown to be a good compromise between the time needed to detect rare MES and the tolerance of both the patient and the investigator (Droste et al., 1996). Besides the 1 h recording time, more than another hour is needed for the evaluation of the tape and its proper analysis. This makes embolus detection very time consuming and laborious.
27
There have been several approaches to automated embolus detection. Some older software only rely on the relative intensity increase of the embolic signal as compared to the background spectrum to define an embolic event (Droste et al., 1996). The problem of this method is that artefacts can produce similar relative intensity increases. By considering only unidirectional signals within the flow spectrum and excluding bidirectional signals, the latter method became more refined (Markus et al., 1993; Casty, 1994). However, bidirectional MES do also exist. The use of a trained neural network is another attractive development for semi-automated embolus detection (Siebler et al., 1992, 1993). The present, most promising development is the multi-gated transcranial Doppler approach, taking into account the fact that an embolus moves inside the insonated artery, whereas an artefact does not. The software traces the embolus at two different depths, and uses the time delay of the high intensity transients a moving embolus produces within two locally separated sample volumes in the same artery. This is contrary to an artefact which affects the signal at both insonation depths simultaneously (Figs. 3 and 4). None of the current systems, however, has reached a sensitivity and specificity required for clinical use. Automated embolus detection has already become helpful to define regions of interest. So far, each of the signals detected by these devices needs to be verified by an experienced investigator.
3. Clinical impact
Fig. 4. The left template shows an artefact within the Doppler spectrum. The artefact was produced by tapping against the probe. On the right side the pre-FFT data are given for all the four channels. There is no time delay between the four channels.
Several studies in patients with symptomatic high-grade carotid artery stenosis have shown that MES can be found more frequently in the first days following a stroke (‘smoking guns’, Fig. 5). The prognostic value of the presence and number of MES has not yet been clearly demonstrated, The following potential advances in the treatment of cerebrovascular patients have been suggested by pioneers in this field, but have not
28
D.W. Droste, E.B. Ringelstein / European Journal of Ultrasound 7 (1998) 23–29
References
Fig. 5. Number of MES per hour in 26 patients with symptomatic high-grade carotid artery stenosis. There is a net decline of MES following the event.
yet been proven unequivocally. In asymptomatic patients, this technique may identify those with an active embolic source, i.e. microembolus detection would allow for preclinical identification of a subgroup of patients at high risk for stroke. In symptomatic patients, after an index event, microembolus detection might be able to pinpoint those individuals at high risk for recurrent stroke (Siebler et al., 1995). Furthermore, this technique could help to identify the site of the embolizing lesion, particularly in patients with competing sources of embolism. The ultrasound-based detection of microembolism might also serve as a surrogate marker in interventional trials. In patients with a first-ever ischemic event and a high grade carotid artery stenosis, the prevalence of a recurrent stroke is low ( :7% per annum) (European Carotid Surgery Trialists Collaborative Group, 1991). However, in symptomatic ICA stenosis, the prevalence of clinically silent ES in recordings of 20 min to 4 h is much higher ( : 21–100%) (Siebler et al., 1993; Grosset et al., 1994b; Markus et al., 1995). Microembolus detection might reduce the observation time and the number of patients needed to perform interventional trials, but still requires to be validated as a meaningful prognostic parameter.
Babikian VL, et al. Clinical correlates of high-intensity transient signals detected on transcranial Doppler sonography in patients with cerebrovascular disease. Stroke 1994;25:1570 – 3. Braekken SK, et al. Incidence and frequency of cerebral embolic signals in patients with a similar bileaflet mechanical heart valve. Stroke 1995;26:1225 – 30. Bush G, Evans DH. Digital audio tape as a method of storing Doppler ultrasound signals. Physiol Meas 1993;14:381–6. Casty M. Automated detection of microemboli in transcranial Doppler signals. Ultraschall Med 1994;15:253 – 8. Consensus Committee of the 9th International Cerebral Hemodynamic Symposium. Basic identification criteria of Doppler microembolic signals. Stroke 1995;26:1123. Droste DW, et al. The effect of different settings of ultrasound pulse amplitude, gain and sample volume on the appearance of emboli studied in a transcranial Doppler model. Cerebrovasc Dis 1994;4:152 – 6. Droste DW, et al. Variability in occurrence of embolic signals in long term transcranial Doppler recordings. Neurol Res 1996;18:25 – 30. Droste DW, et al. Bigated transcranial Doppler for the detection of clinically silent circulating emboli in normal persons and patients with prosthetic cardiac valves. Stroke 1997a;28:588 – 92. Droste DW, et al. Oxygen inhalation can differentiate gaseous from non-gaseous microemboli detected by transcranial Doppler ultrasound. Stroke 1997b;28:2453– 6. European Carotid Surgery Trialists Collaborative Group. MRC European carotid surgery trial: Interim results for symptomatic patients with severe (70 – 99%) or with mild (0 – 29%) carotid stenosis. Lancet 1991;337:1235 – 43. Grosset DG, et al. Ultrasound detection of cerebral emboli in patients with prosthetic heart valves. J Heart Valve Dis 1994a;3:128 – 32. Grosset DG, et al. Doppler emboli signals vary according to stroke subtype. Stroke 1994b;25:382 – 4. Kaps M, et al. Clinically silent microemboli in patients with artificial prosthetic aortic valves are predominantly gaseous and not solid. Stroke 1997;28:322 – 5. Markus H, et al. Computerized detection of cerebral emboli and discrimination from artifact using Doppler ultrasound. Stroke 1993;24:1667 – 72. Markus HS, et al. How good is intercenter agreement in the identification of embolic signals in carotid artery disease? Stroke 1996;27:1249 – 52. Markus HS, Molloy J. Use of a decibel threshold in detecting Doppler embolic signals. Stroke 1997;28:692 – 5. Markus HS, et al. Asymptomatic cerebral embolic signals in symptomatic and asymptomatic carotid artery disease. Brain 1995;118:1005– 11. Russell D. The detection of cerebral emboli using Doppler ultrasound. In: Newell DW, Aaslik R, editors. Transcranial Doppler. New York: Raven Press, 1992:52 – 8.
D.W. Droste, E.B. Ringelstein / European Journal of Ultrasound 7 (1998) 23–29 Siebler M, et al. Silent cerebral embolism caused by neurologically symptomatic high-grade carotid stenosis. Brain 1993;116:1005–15. Siebler M, et al. Cerebral microembolism and the risk of ischemia in asymptomatic high-grade internal carotid artery stenosis. Stroke 1995;26:2184–6.
29
Siebler M, et al. Detection of intracranial emboli in patients with symptomatic extracranial carotid artery disease. Stroke 1992;23:1652 – 4. Spencer MP, et al. Detection of middle cerebral artery emboli during carotid endarterectomy using transcranial Doppler ultrasonography. Stroke 1990;21:415 – 23.
. .
Review
Detection of high intensity transient signals (HITS): How and why? Dirk W. Droste *, E. Bernd Ringelstein Klinik und Poliklinik fu¨r Neurologie der WWU Mu¨nster, Albert-Schweitzer-Str. 33, D-48129 Mu¨nster, Germany
Abstract HITS (high intensity transient signals) in transcranial Doppler recordings reflect either microemboli, both gaseous and solid, or artefacts. Various frequencies in number of microembolic signals (MES) have been reported in the same clinical condition. A possible explanation for these discrepancies may be different device settings and algorithms for embolus detection. For reproducibility of data, we suggest that studies on MES report the following parameters: (1) Ultrasound device; (2) transducer type; (3) insonated artery; (4) insonation depth; (5) algorithms for signal intensity measurement; (6) scale settings; (7) detection threshold; (8) axial extension of sample volume; (9) fast Fourier transform (FFT) size (number of points used); (10) FFT length (time); (11) FFT overlap; (12) transmitted ultrasound frequency; (13) high pass filter settings; and (14) recording time. No current system of automatic embolus detection has the full sensitivity and specificity required for clinical use. Therefore, each of the signals detected by these devices needs to be checked and verified by an experienced investigator. MES will help to identify the site and activity of the embolizing lesion. Microembolus detection might reduce the observation time and the number of patients needed to perform interventional trials. First, however, MES needs to be validated as a meaningful prognostic parameter. Microemboli originating from prosthetic cardiac valves are mainly gaseous. Therefore, they cannot serve as an indicator of the valves thromboembolic activity or the patient’s stroke risk. © 1998 Elsevier Science Ireland Ltd. All rights reserved. Keywords: High intensity transient signals; Transcranial Doppler; Prosthetic cardiac values; Stroke
1. Introduction
* Corresponding author. Tel.: + 49 251 8348176; fax: + 49 251 8348181.
Emboli, such as thrombus and atheroma, originating from carotid and aortic plaques or cardiac sources are generally accepted as a main cause for cerebral ischemic events in humans. In 1990,
0929-8266/98/$19.00 © 1998 Elsevier Science Ireland Ltd. All rights reserved. PII S 0 9 2 9 - 8 2 6 6 ( 9 8 ) 0 0 0 1 0 - X
24
D.W. Droste, E.B. Ringelstein / European Journal of Ultrasound 7 (1998) 23–29
Fig. 1. Example of a MES in a patient with a carotid artery stenosis.
Spencer et al. (1990) described that clinically asymptomatic circulating cerebral microemboli in patients with carotid artery stenosis could be detected by transcranial Doppler ultrasound (TCD). Fig. 1 gives an example of such a microembolic signal (MES). Although these microemboli are clinically silent, they might be clinically important by indicating an increased risk of stroke. High intensity transient signals (HITS) in transcranial Doppler recordings reflect either gaseous or solid microemboli, or artefacts.
2. Methodological considerations The detection of microembolic particles within the streaming blood is based on the measurement of the backscatter from the emboli. The diffuse backscatter of the ultrasound from the normal flowing blood (including transient erythrocyte aggregates) is usually much lower than the backscatter from solid emboli. The latter, however, is usually much lower than the backscatter from gaseous ones. At present, no conclusion as to the composition and the size of an embolus can be drawn from the embolus’ echo (Russell, 1992). Our group has shown in two studies, that mi-
croemboli originating from prosthetic cardiac valves are mainly gaseous (Droste et al., 1997b; Kaps et al., 1997). Therefore, they cannot serve as an indicator of the valves thromboembolic activity or the patient’s stroke risk. Strongly variable frequencies in number of MES have been reported in the same clinical condition. One explanation for this discrepancy may be the different device settings and variable algorithms used for embolus detection. Technical parameters affecting the detectability of MES are; (1) The relationship between the backscattered power from emboli and that from the blood, i.e. the relative intensity increase; (2) the detection threshold; (3) the size of the sample volume; (4) the fast Fourier transform (FFT) frequency resolution; (5) the FFT temporal resolution; (6) the FFT temporal overlap; (7) the instrumentation’s dynamic range; (8) the transmitted ultrasound frequency; (9) the filters’ settings; and (10) the recording time. The settings of the ultrasound instrumentation strongly influences the detectability of MES (Droste et al., 1994; International Consensus Group on Microembolus Detection, 1998). The investigator should be aware of these effects. Some of these aspects will be addressed later in more detail. It is essential to maintain parameters constant throughout and between
D.W. Droste, E.B. Ringelstein / European Journal of Ultrasound 7 (1998) 23–29
recordings, and to parallel settings in multicentre studies.
2.1. Relati6e intensity increase At present, the relative intensity increase of the embolic signal is measured in different ways. Different types of signal analysis are used in the different devices, and can additionally be modified by the user. The relative intensity increase of the embolic signal is usually measured in decibels (dB). In the frequency domain-based analysis, for instance, the peak intensity of a microembolic event, or its mean within a defined time frame and frequency range, can be used. Similarly, the intensity of the background signal may be expressed as a mean value or a median value over variable time periods and frequency ranges, (e.g. at a comparable location as the embolic echo in the preceding cardiac cycle, or comprising arbitrarily defined time frames preceding the embolic event, or considering the whole sweep, even including the signal-free areas of the screen). Thus, for a given embolic signal, different decibel values of relative intensity increase can be calculated using different background and embolic signal intensity measurements. The user should be aware of which technique is used in the particular automated embolus detection systems he is working with; this should be specified. In the same way, manual techniques of calculating signal intensity should always be specified (Markus and Molloy, 1997). Some intensity calculations of the embolic signal and of the background depend on the frequency scale setting (pulse repetition frequency, PRF ), as more or less spectrum-free area is included. Thus, the scale setting should also be indicated and kept constant.
2.2. Detection threshold At present, the various manufacturers and investigators use greatly different parameters and criteria for identifying a short-lasting ultrasound event as microembolic in nature. Particularly, greatly different decibel (dB)-thresholds ranging from 3 to 9 dB have been recommended for discriminating MES from the general background
25
noise and from spontaneous, speckle-like intensity fluctuations of the physiological Doppler flow signals (Siebler et al., 1993; Consensus Committee of the 9th International Cerebral Hemodynamic Symposium, 1995; Markus et al., 1995). The situation is even more complicated because different algorithms are used to calculate the background noise and the intensity of the presumed microembolic signal as such. These technical aspects may have contributed to the striking discrepancies in the prevalence of MES described in the literature in various types of stroke or stroke-prone patients (Babikian et al., 1994; Grosset et al., 1994a,b; Braekken et al., 1995; Markus et al., 1995). There are two possible ways of determining the detection threshold of microemboli in decibels for a given device: either defining the range of spontaneous intensity fluctuations within the Doppler signals of normal controls, or defining fluctuations on a case-by-case basis during emboli-free periods (Markus et al., 1995; Droste et al., 1997a). It is not yet clear whether thresholds defined in the middle cerebral artery can be used for other intracranial arteries, or for poststenotic middle cerebral artery flow spectra as well. To the best of our present knowledge, calibration of individual machines by either normal controls or by intrapatient analysis of the background signal is equally valid. Each device should be individually calibrated, and the approach used must be clearly indicated. Fig. 2 gives an example of such a calibration. A higher detection threshold results in lower sensitivity, but higher specificity and higher intercentre agreement (Markus et al., 1996).
2.3. Artefact rejection Discrimination of true MES from artefacts, e.g. produced by probe displacement, is of crucial importance. Bidirectional signals, i.e. signals above and below the baseline, frequently represent artefacts. However, MES of high echo intensity may occasionally produce bidirectional signals, particularly if gaseous in nature or with inadequate settings of the instrumentation. Investigators new in this field are encouraged to produce artefact signals on purpose, to become
26
D.W. Droste, E.B. Ringelstein / European Journal of Ultrasound 7 (1998) 23–29
Fig. 2. The relative intensity increases of the Doppler speckle background in the absence of MES and of MES in patients with prosthetic cardiac valves and arterial embolic sources given as a percent. A detection threshold of 5 dB and more was chosen for the device (TC 4040, Nicolet-EME) and the setting used. Note that MES from prosthetic heart valves due to their predominantly gaseous composition have a higher relative intensity increase.
familiar with their characteristics. The multigatetechnique (see below) uses sampling of echoes from different depths of the same artery. This allows to trackle the migration of echoes within the flowing blood, i.e. their sequential appearance from the more proximal to the more distal sample volumes. By contrast, an artefact affects all channels simultaneously. Figs. 3 and 4 give examples of an embolic signal and an artefact recorded with the multigate technique. The TCD devices currently available are not yet able to automatically discriminate artefacts from microemboli with satisfying reliability.
2.4. Documentation and quality control At present, the most widely used documentation system is the recording of the pre-FFT audio signal ( = raw data) on digital audio tapes (DAT)
(Bush and Evans, 1993). This allows the data to be subjected to quality control and the re-evaluation of regions of interest. It also allows for off-line analysis. For scientific purposes, observer bias can be avoided by a blinded analysis of the audio tapes by different observers. It is important to ensure reproducibility both between and within centres in the identification of MES. For inter-observer studies, it is important to guarantee that observers select the same MES. A statistical method which encounters this is required (e.g. probability of specific agreement, rather than counting the total number of emboli recorded by each observer) (Markus et al., 1996). Exchange and analysis of data among centres is encouraged. For multicentre studies, the use of identical devices and a standardised, identical setting of the equipment is strongly recommended.
D.W. Droste, E.B. Ringelstein / European Journal of Ultrasound 7 (1998) 23–29
Fig. 3. The left template shows an embolic signal within the Doppler spectrum. On the right side, the pre-FFT (fast Fourier form) data are given for all the four channels. The x-axis is stretched to better demonstrate the time delay between the four channels.
2.5. Automatic embolus detection Currently, most centres record the Doppler signal on digitalised audio tapes and evaluate these tapes acoustically and visually off-line and blinded to the clinical diagnosis. A recording time of 1 h has been shown to be a good compromise between the time needed to detect rare MES and the tolerance of both the patient and the investigator (Droste et al., 1996). Besides the 1 h recording time, more than another hour is needed for the evaluation of the tape and its proper analysis. This makes embolus detection very time consuming and laborious.
27
There have been several approaches to automated embolus detection. Some older software only rely on the relative intensity increase of the embolic signal as compared to the background spectrum to define an embolic event (Droste et al., 1996). The problem of this method is that artefacts can produce similar relative intensity increases. By considering only unidirectional signals within the flow spectrum and excluding bidirectional signals, the latter method became more refined (Markus et al., 1993; Casty, 1994). However, bidirectional MES do also exist. The use of a trained neural network is another attractive development for semi-automated embolus detection (Siebler et al., 1992, 1993). The present, most promising development is the multi-gated transcranial Doppler approach, taking into account the fact that an embolus moves inside the insonated artery, whereas an artefact does not. The software traces the embolus at two different depths, and uses the time delay of the high intensity transients a moving embolus produces within two locally separated sample volumes in the same artery. This is contrary to an artefact which affects the signal at both insonation depths simultaneously (Figs. 3 and 4). None of the current systems, however, has reached a sensitivity and specificity required for clinical use. Automated embolus detection has already become helpful to define regions of interest. So far, each of the signals detected by these devices needs to be verified by an experienced investigator.
3. Clinical impact
Fig. 4. The left template shows an artefact within the Doppler spectrum. The artefact was produced by tapping against the probe. On the right side the pre-FFT data are given for all the four channels. There is no time delay between the four channels.
Several studies in patients with symptomatic high-grade carotid artery stenosis have shown that MES can be found more frequently in the first days following a stroke (‘smoking guns’, Fig. 5). The prognostic value of the presence and number of MES has not yet been clearly demonstrated, The following potential advances in the treatment of cerebrovascular patients have been suggested by pioneers in this field, but have not
28
D.W. Droste, E.B. Ringelstein / European Journal of Ultrasound 7 (1998) 23–29
References
Fig. 5. Number of MES per hour in 26 patients with symptomatic high-grade carotid artery stenosis. There is a net decline of MES following the event.
yet been proven unequivocally. In asymptomatic patients, this technique may identify those with an active embolic source, i.e. microembolus detection would allow for preclinical identification of a subgroup of patients at high risk for stroke. In symptomatic patients, after an index event, microembolus detection might be able to pinpoint those individuals at high risk for recurrent stroke (Siebler et al., 1995). Furthermore, this technique could help to identify the site of the embolizing lesion, particularly in patients with competing sources of embolism. The ultrasound-based detection of microembolism might also serve as a surrogate marker in interventional trials. In patients with a first-ever ischemic event and a high grade carotid artery stenosis, the prevalence of a recurrent stroke is low ( :7% per annum) (European Carotid Surgery Trialists Collaborative Group, 1991). However, in symptomatic ICA stenosis, the prevalence of clinically silent ES in recordings of 20 min to 4 h is much higher ( : 21–100%) (Siebler et al., 1993; Grosset et al., 1994b; Markus et al., 1995). Microembolus detection might reduce the observation time and the number of patients needed to perform interventional trials, but still requires to be validated as a meaningful prognostic parameter.
Babikian VL, et al. Clinical correlates of high-intensity transient signals detected on transcranial Doppler sonography in patients with cerebrovascular disease. Stroke 1994;25:1570 – 3. Braekken SK, et al. Incidence and frequency of cerebral embolic signals in patients with a similar bileaflet mechanical heart valve. Stroke 1995;26:1225 – 30. Bush G, Evans DH. Digital audio tape as a method of storing Doppler ultrasound signals. Physiol Meas 1993;14:381–6. Casty M. Automated detection of microemboli in transcranial Doppler signals. Ultraschall Med 1994;15:253 – 8. Consensus Committee of the 9th International Cerebral Hemodynamic Symposium. Basic identification criteria of Doppler microembolic signals. Stroke 1995;26:1123. Droste DW, et al. The effect of different settings of ultrasound pulse amplitude, gain and sample volume on the appearance of emboli studied in a transcranial Doppler model. Cerebrovasc Dis 1994;4:152 – 6. Droste DW, et al. Variability in occurrence of embolic signals in long term transcranial Doppler recordings. Neurol Res 1996;18:25 – 30. Droste DW, et al. Bigated transcranial Doppler for the detection of clinically silent circulating emboli in normal persons and patients with prosthetic cardiac valves. Stroke 1997a;28:588 – 92. Droste DW, et al. Oxygen inhalation can differentiate gaseous from non-gaseous microemboli detected by transcranial Doppler ultrasound. Stroke 1997b;28:2453– 6. European Carotid Surgery Trialists Collaborative Group. MRC European carotid surgery trial: Interim results for symptomatic patients with severe (70 – 99%) or with mild (0 – 29%) carotid stenosis. Lancet 1991;337:1235 – 43. Grosset DG, et al. Ultrasound detection of cerebral emboli in patients with prosthetic heart valves. J Heart Valve Dis 1994a;3:128 – 32. Grosset DG, et al. Doppler emboli signals vary according to stroke subtype. Stroke 1994b;25:382 – 4. Kaps M, et al. Clinically silent microemboli in patients with artificial prosthetic aortic valves are predominantly gaseous and not solid. Stroke 1997;28:322 – 5. Markus H, et al. Computerized detection of cerebral emboli and discrimination from artifact using Doppler ultrasound. Stroke 1993;24:1667 – 72. Markus HS, et al. How good is intercenter agreement in the identification of embolic signals in carotid artery disease? Stroke 1996;27:1249 – 52. Markus HS, Molloy J. Use of a decibel threshold in detecting Doppler embolic signals. Stroke 1997;28:692 – 5. Markus HS, et al. Asymptomatic cerebral embolic signals in symptomatic and asymptomatic carotid artery disease. Brain 1995;118:1005– 11. Russell D. The detection of cerebral emboli using Doppler ultrasound. In: Newell DW, Aaslik R, editors. Transcranial Doppler. New York: Raven Press, 1992:52 – 8.
D.W. Droste, E.B. Ringelstein / European Journal of Ultrasound 7 (1998) 23–29 Siebler M, et al. Silent cerebral embolism caused by neurologically symptomatic high-grade carotid stenosis. Brain 1993;116:1005–15. Siebler M, et al. Cerebral microembolism and the risk of ischemia in asymptomatic high-grade internal carotid artery stenosis. Stroke 1995;26:2184–6.
29
Siebler M, et al. Detection of intracranial emboli in patients with symptomatic extracranial carotid artery disease. Stroke 1992;23:1652 – 4. Spencer MP, et al. Detection of middle cerebral artery emboli during carotid endarterectomy using transcranial Doppler ultrasonography. Stroke 1990;21:415 – 23.
. .