Detection of Cerebral Microemboli by Means of Transcranial Doppler Monitoring Before and After Carotid Endarterectomy
Background and Purpose The main purpose of carotid endarterectomy (CEA) for neurologically symptomatic high-grade extracranial carotid artery stenosis is to remove the suspected source of cerebral microemboli. Transcranial Doppler (TCD) ultrasonography has the potential for detecting solid microemboli in the basal cerebral arteries. Therefore, TCD monitoring provides the opportunity to assess the rate of microemboli to the brain in patients with symptomatic high-grade carotid artery stenosis and to verify whether these phenomena have ceased after CEA.
Methods TCD monitoring was performed in 41 patients to detect high-intensity transient signals indicating microemboli in the middle cerebral artery before and after CEA. In the event that, within 1 week after CEA, TCD monitoring revealed ongoing cerebral microemboli on the side of surgery, the examination was repeated 3 months later.
Results High-intensity transient signals were detected preoperatively on the side of the affected carotid artery in 22 patients (54%; mean, 10.2 per hour; range, 1 to 88). Linear regression analysis demonstrated a trend toward an inverse relationship between the number of high-intensity transient signals per hour and the time interval since the last episode of neurological symptoms (P<.1). CEA resulted in a significant reduction in the number of high-intensity transient signals per hour 7 days after surgery (mean, 6.0 versus 0.4 per hour; median, 0 versus 0; n=37; P<.005) and 3 months later (mean, 6.3 versus 0 per hour; median, 1.3 versus 0; n=41; P<.0001).
Conclusions Clearly, TCD monitoring can be helpful in assessing the effect of CEA in removing the suspected source of cerebral microemboli. Ongoing microemboli to the brain should prompt reassessment of the operated carotid artery or a search for other potential sources of cerebral embolism. Carotid artery plaques seem to produce cerebral microemboli for a limited period, which implies that the prophylactic effect of CEA might decrease if the operation is delayed.
Recently, two large studies reported a marked effect on stroke prevention resulting from carotid endarterectomy (CEA) in the event of high-grade carotid artery stenosis in patients with recent ipsilateral transient ischemic attacks of the brain or eye or a nondisabling ischemic stroke.1 2 However, the association between high-grade carotid artery stenosis and neurological symptoms is poorly understood, and our insight into the pathophysiology of cerebral ischemic deficits is far from complete. Although thromboembolism in the brain is considered of major importance, the hemodynamic and thromboembolic theories for explaining the occurrence of cerebral ischemia are not mutually exclusive. Transcranial Doppler (TCD) ultrasonography has the potential for detecting solid microemboli in the basal cerebral arteries.3 Therefore, TCD monitoring provides the opportunity to evaluate the thromboembolic theory by assessment of the rate of microemboli to the brain in patients with neurologically symptomatic high-grade carotid artery stenosis and to thereby verify whether these phenomena have ceased after CEA. To address these issues, we studied TCD monitoring of the middle cerebral artery (MCA) before and after CEA.
Subjects and Methods
From February 1993 through November 1993, a total of 112 patients underwent CEA in our institution, of whom 41 (37%) were enrolled in this study. The sampling was determined by practical reasons; neither clinical condition nor neurological history were applied as selection criteria. Thirty-four patients were men and 7 were women. Mean age was 63 years (range, 42 to 80 years). All patients underwent preoperative neurological examination. Brain computed tomography (CT) was done in 38 patients and showed normal results (n=25) or cerebral infarction (n=13). Fourteen of the 38 patients (37%) had suffered one or more episodes of transient monocular blindness, 13 (34%) transient ischemic attacks of the brain, 9 (24%) nondisabling ischemic stroke, and 2 (5%) major stroke. The 3 patients without previous CT had no neurological symptoms. The number of episodes of ischemic symptoms was registered. One patient had chronic atrial fibrillation. In the other patients, no cardiac embolic sources were identified. During the study, 36 patients were taking antiplatelet treatment (always including aspirin). The other 5 patients were taking oral anticoagulation therapy, which was continued postoperatively.
The degree of carotid artery stenosis was assessed by digital subtraction angiography or duplex scanning. Thirty-nine patients had a high-grade stenosis (70% to 99%) of the operated carotid artery. One of the remaining 2 patients underwent CEA because of neurologically symptomatic restenosis of a previously operated internal carotid artery. The other patient had repetitive transient ischemic attacks with a nonstenotic but ulcerated lesion in the ipsilateral internal carotid artery. During surgery, 14 patients were judged to need a shunt according to our institutional criteria based on intraoperative electroencephalographic, TCD, and stump pressure measurements.4 In 15 patients, the arteriotomy was closed with a vein patch. To prevent early thrombotic occlusion of the operated carotid artery, all patients received intravenous heparin before the vessels were clamped. One patient had developed occlusion of the symptomatic common carotid artery before surgery, and in this case a vascular bypass operation was performed.
TCD monitoring was carried out with a computer-controlled Doppler ultrasound device (TC2000, Eden Medical Electronics Inc) with a 2-MHz pulsed-wave transducer firmly affixed to the lateral temporal region to minimize movement artifacts. Continuous surveillance of the monitoring sessions ensured optimal quality of the Doppler signal and identification of patient movements. Doppler signals were recorded on stereo videotape and analyzed off-line with dedicated software (eme), which allowed the determination of relative power spectra and signal duration. In an unblinded fashion, high-intensity transient signals (HITS) indicating microemboli were identified according to the criteria given by Spencer et al and Russell et al.3 5 6 To differentiate HITS from other artifacts, the former were defined as having a duration of less than 100 milliseconds and an energy of at least 3 dB above the background Doppler blood velocity spectrum. Such signals are unidirectional and occur at random within the cardiac cycle. Furthermore, most HITS produce a specific sound known as “chirps” or “blips.” HITS were identified without the use of automatic emboli-detection software. The observed HITS were counted and adjusted to the number per hour.
TCD monitoring was planned before and at 1 to 7 days after surgery. Data on monitoring times are presented in the Table⇓. In most patients, the preoperative TCD monitoring was performed within 1 week before surgery. The mean interval in days between the onset of the last episode of neurological symptoms and the time of preoperative monitoring was 87 days (median, 70; range, 0 to 360 days; n=36). In 2 of the 38 neurologically symptomatic patients, the time interval could not be accurately retrieved. In 37 patients, the postoperative monitoring was done 1 to 7 days (mean, 3.2 days; median, 2 days) after CEA. Three patients showed ongoing HITS. Only in these cases was the examination repeated after 3 months. In another 4 patients, the postoperative TCD monitoring had to be postponed for several weeks for practical reasons. In the final analysis, only results at 3 months were included for the latter patients.
Selecting the most suitable time for postoperative TCD monitoring was another objective of the study. Therefore, 15 patients (37%) also underwent 15 to 30 minutes of TCD monitoring of the MCA on the affected side 1 to 2 hours after surgery. Subsequently, TCD monitoring of the opposite MCA for the same period of time was performed as a control procedure (Table⇑).
The results were analyzed for statistical significance by nonparametric tests (Mann-Whitney and Wilcoxon) and by linear regression analysis. Probability values lower than .05 were considered to indicate significance.
In 22 patients (54%), HITS were detected preoperatively on the side of the affected carotid artery (mean, 10.2 per hour; range, 1 to 88 per hour). TCD monitoring of the contralateral MCA showed HITS in 2 of the 10 patients. Both these patients had HITS on the symptomatic side as well. In one of these two patients, the contralateral carotid artery also showed a high-grade stenosis. In the second patient, the contralateral carotid artery was occluded with evidence of transhemispheric blood flow through the anterior communicating artery on angiography. In the patient with chronic atrial fibrillation, no HITS were found before and after surgery. During TCD monitoring, none of the patients developed new neurological symptoms.
Linear regression analysis demonstrated a trend toward an inverse relationship between the number of preoperative HITS per hour in the MCA on the side of surgery and the time since the last ischemic symptoms (P<.1, Figure⇓). Transformation of the number of HITS to a normal distribution did not influence the result of the analysis. We failed to identify a relationship between the number of episodes of neurological symptoms in the relevant carotid territory and the number of preoperative HITS in the ipsilateral MCA. Moreover, no correlation was found between the number of HITS per hour and the type of ischemic symptoms or preoperative medication.
CEA resulted in a statistically significant reduction in the number of HITS within 7 days (mean, 6.0 versus 0.4 per hour; median, 0 versus 0 per hour; n=37; P<.005) and 3 months (mean, 6.3 versus 0 per hour; median, 1.3 versus 0 per hour; n=41; P<.0001) after the operation. The persistence of HITS in 3 patients was unrelated to the time interval since surgery. Three months later, these 3 patients were free of HITS.
TCD monitoring was also performed 1 to 2 hours after CEA in 15 patients. In this early postoperative phase, there was a significant increase in the number of HITS per hour in the ipsilateral MCA (mean, 76.7 versus 1.5 per hour; median, 32.0 versus 0 per hour; P<.005). The number of HITS was not related to the perioperative medication or the use of a shunt or vein patch. Other possible relevant variables, such as vascular risk factors and surgical technique, were not evaluated. In 2 of the 15 patients, early postoperative TCD monitoring also revealed multiple HITS (42 and 120 per hour) in the opposite MCA. Both patients had a high-grade stenosis of the contralateral internal carotid artery as well. On preoperative TCD monitoring, one of the two patients had also shown HITS in the contralateral MCA (4 per hour), and in the other patient no HITS were detected. Three months after surgery, no HITS were found in either patient.
The main purpose of CEA for neurologically symptomatic high-grade extracranial carotid artery stenosis is to remove the suspected source of cerebral embolism. Clearly, the results of this study demonstrate the effect of CEA of preventing embolization to the ipsilateral MCA. HITS have been found to represent microemboli in in vitro and in vivo studies.5 6 7 8 9 10 The impact of cerebral microemboli on the functional integrity of the brain remains to be elucidated. None of the detected HITS in our patients resulted in neurological symptoms. Apart from a recent case report of embolic stroke during TCD examination after carotid compression,11 earlier studies on TCD monitoring also have failed to detect symptomatic embolism.
Recently, a study of TCD monitoring in 14 patients before and after CEA reported similar results.12 In that study, all patients had shown HITS in the MCA on the affected side before the operation. It is not clear whether this should be attributed to the prolonged period of monitoring or to the selection of patients. In the present study, only 54% of the patients preoperatively showed HITS in the MCA on the side of surgery. In the event of a low frequency of cerebral microemboli, the period of TCD monitoring might be too short to record HITS. In our study, 6 of the 41 patients demonstrated only one or two HITS per hour on TCD monitoring, and in 19 we found none. Siebler et al8 found a mean of less than two microemboli per hour in the ipsilateral MCA in 3 patients with neurologically symptomatic carotid artery stenosis. It is reasonable to assume that some of our patients who did not demonstrate HITS in up to 60 minutes of TCD monitoring actually experienced sporadically occurring microemboli beyond the period of observation. Moreover, microemboli may occur unnoticed in other cerebral arteries, outside the Doppler sample volume. Obviously, the choice to limit insonation to the MCA, which is the most accessible and suitable cerebral artery for TCD monitoring, is a compromise between feasibility and thoroughness.
It might be argued that the criteria to identify HITS are subject to controversy. Studies from different centers have reported widely differing numbers of HITS in symptomatic carotid artery stenosis. The reasons for these differences are still unclear but may, at least in part, be due to different criteria used for identification of HITS. Using the criteria defined by Spencer et al and Russell et al,3 5 6 we found a relatively low percentage of positive TCD monitorings (54%) in the present study compared with previous studies. Validation of the criteria may be expected from international multicenter interobserver-agreement studies.
Before operation, one of our patients with a unilateral occlusion of the internal carotid artery and a high-grade stenosis on the other side showed HITS on both sides in the MCA. Postoperatively, this patient was free of HITS in either MCA. Recently, transhemispheric passage of microemboli has been demonstrated in patients with unilateral internal carotid artery occlusion.13 We believe that in our patient a similar crossing through the anterior circle of Willis was responsible for cerebral microemboli on the opposite side.
The inverse trend between the number of preoperative HITS per hour in the MCA on the affected side and the interval since the last ischemic episode might have important implications. Given that cerebral embolism is the major cause of recurrent cerebral ischemia, the prophylactic effect of CEA in patients with symptomatic high-grade carotid artery stenosis might decrease with time. This hypothesis is supported by the results of the European Carotid Surgery Trial,1 which show that in patients not allocated to the CEA group the main excess of disabling or fatal ipsilateral ischemic stroke occurred in the first year. Moreover, Bamford et al14 reported a marked clustering in the first 6 months of partial anterior circulation infarcts, suggesting only a limited period during which sources actively release showers of cerebral microemboli. Plaque morphology may be a relevant factor in the pathogenesis of cerebrovascular events. Using Doppler color flow imaging, Steinke et al15 found more ulcerated plaques in symptomatic than in asymptomatic carotid artery stenosis. With time, atheromatous plaques in the carotid artery may show a tendency to restore endothelialization, thereby reducing the risk of embolism. Studies using TCD monitoring, combined with ultrasonic imaging and specimens of the carotid artery, can help to improve our understanding of the natural history of symptomatic carotid artery stenosis as a source of cerebral microemboli.
In most patients, postoperative TCD monitoring was feasible within 1 week after CEA. The finding of continuing cerebral microemboli at that time should prompt reassessment of the operated carotid artery or a search for other potential sources of microemboli. However, monitoring a few hours after surgery gives paradoxical results and should be interpreted carefully to avoid unnecessary reexploration of the operated carotid artery. On the other hand, early postoperative TCD monitoring may provide insight into the mechanisms of repair of the human vascular wall after CEA. Animal studies have demonstrated a layer of fibrin and platelets on the endarterectomized surface within 30 minutes after reestablishment of blood flow.16 17 Heparinization failed to prevent the development of the layer of fibrin and platelets but dramatically reduced mural thrombus formation. Therefore, early postoperative TCD monitoring can be expected to detect predominantly platelet aggregates that have been dislodged from the surgical site.
Surprisingly, two patients showed a remarkable but temporary increase in the number of HITS distal to an asymptomatic high-grade carotid artery stenosis on early postoperative TCD monitoring opposite to the side of surgery. A possible fibrinolytic activity of heparin is still the subject of controversy.18 It is possible that the use of intravenous heparin during surgery promotes dislodgment of embolic material from plaque in the contralateral carotid artery. Moreover, as many as 30% of patients receiving heparin will develop heparin-induced thrombocytopenia. Especially in the event of reexposure to heparin, patients were found to be at risk for thromboembolic events, including early stroke after CEA.19 20 The presence of anticardiolipin antibodies21 as well as a hypercoagulable state might also increase the risk for thromboembolism. Screening for antibodies or hemostatic and fibrinolytic indexes was not part of our protocol.
Our study demonstrates that detection of cerebral HITS with TCD monitoring is a reliable method of assessing the effect of CEA in removing the source of microemboli in patients with neurologically symptomatic high-grade carotid artery stenosis. Further research, by means of TCD monitoring, may provide pathophysiological evidence of the result of CEA in patients with asymptomatic high-grade carotid artery stenosis. Moreover, detection of cerebral HITS might be helpful in identifying those patients with lesser degrees of carotid artery stenosis who are still at high risk for ischemic stroke. The European Carotid Surgery Trial1 and the North American Symptomatic Carotid Endarterectomy Trial2 continue randomization of patients with moderate (30% to 69%) carotid artery stenosis to determine whether CEA is beneficial in reducing the risk of ischemic stroke. Evidence of microemboli from the carotid artery to the brain on TCD monitoring might prove to be a reason for advocating surgery before high-grade stenosis has developed, especially in patients who do not respond to medical treatment.
- Received April 12, 1994.
- Revision received October 6, 1994.
- Accepted October 7, 1994.
- Copyright © 1995 by American Heart Association
Spencer MP. Detection of cerebral arterial emboli. In: Newell DW and Aaslid R, eds. Transcranial Doppler. New York, NY: Raven Press Publishers; 1992:215-230.
Spencer MP, Thomas GI, Nicholls SC, Sauvage LR. Detection of middle cerebral artery emboli during carotid endarterectomy using transcranial Doppler ultrasonography. Stroke. 1990;21:415-423.
Russell D, Madden KP, Clark WM, Sandset PM, Zivin JA. Detection of arterial emboli using Doppler ultrasound in rabbits. Stroke. 1991;22:253-258.
Albin MS, Bunegin L, Garcia C, McKay W. The transcranial Doppler can image microaggregates of intracranial air and particulate matter. J Neurosurg Anesthesiol. 1989;1:134-135.
Siebler M, Sitzer M, Steinmetz H. Detection of intracranial emboli in patients with symptomatic extracranial carotid artery disease. Stroke. 1992;23:1652-1654.
Markus HS, Brown MM. Differentiation of different pathological cerebral embolic materials using transcranial Doppler in an in vitro model. Stroke. 1993;24:1-5.
Markus HS, Loh A, Brown MM. Computerized detection of cerebral emboli and discrimination from artifact using Doppler ultrasound. Stroke. 1993;24:1667-1672.
Khaffaf N, Karnik R, Winkler W-B, Valentin A, Slany J. Embolic stroke by compression maneuver during transcranial Doppler sonography. Stroke. 1994;25:1056-1057.
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, Lees KR. Transhemispheric passage of microemboli in patients with unilateral internal carotid artery occlusion. Stroke. 1993;24:1664-1666.
Steinke W, Hennerici M, Rautenberg W, Mohr JP. Symptomatic and asymptomatic high-grade carotid stenoses in Doppler color-flow imaging. Neurology. 1992;42:131-138.
Becker PS, Miller VT. Heparin-induced thrombocytopenia. Stroke. 1989;20:1449-1459.
Barbut D, Hinton RB, Supino P, Borer JS. Anticardiolipin antibodies following carotid endarterectomy: association with increased stroke risk. Stroke. 1994;25:258. Abstract.