Role of Transcranial Doppler and Stump Pressure During Carotid Endarterectomy
Background and Purpose The aim of our study was to clarify the pathophysiology of perioperative cerebral complications during carotid endarterectomy in our series.
Methods By means of transcranial Doppler ultrasonography and stump pressure measurement, we monitored 112 patients who underwent carotid endarterectomy under general anesthesia for symptomatic or asymptomatic severe carotid stenosis.
Results Of 18 patients who underwent carotid endarterectomy with intra-arterial shunt, 2 (11.1%) developed an ischemic stroke. Of the other 94 patients, one suffered a nucleocapsular hemorrhage and 5 had cerebral ischemic complications. In these 5 patients, the duration of clamping was significantly longer (mean±SD, 16.4±1.1 versus 12.7±2.6 minutes; P=.0019), and the decrease of middle cerebral artery mean velocity on clamping was significantly greater (mean±SD, 56.4±4.9% versus 28.8±20.2%; P=.0031), while stump pressure was not significantly different. Microembolic signals were recorded in 70 patients (62.5%) and were not associated with cerebral ischemic complications. The 7 patients who developed cerebral ischemic complications had a significantly higher percentage of stenosis in the contralateral internal carotid artery (mean±SD, 82.0±17.8% versus 29.3±36.4%; P=.0018).
Conclusions The results of our study suggest that the major complications of carotid endarterectomy may be due to hemodynamic factors. Stump pressure alone is not a reliable indicator of hemodynamic changes that predict cerebral ischemia. Particulate microembolism may cause more subtle changes in cerebral parenchyma, but further studies are needed to clarify this point.
Carotid endarterectomy has been shown to be superior to the best medical treatment in symptomatic severe carotid stenosis1,2 and, to a lesser extent, in asymptomatic severe carotid stenosis.3 This benefit could increase if perioperative complication rates were reduced. Our insight into the pathophysiology of cerebral ischemia during CEA is not complete, and both hemodynamic and thromboembolic events can occur. Some studies4–6 emphasize the relevance of microembolic events, particularly if they occur during the dissection phase.5 Others7 support the importance of hemodynamic changes in the genesis of intraoperative ischemic stroke.
TCD has the power to detect both microemboli8,9 and blood flow velocity changes in the basal cerebral arteries; in recent years it has frequently been chosen as the principal intraoperative monitoring system.4–7
Stump pressure is another intraoperative measure that is frequently used as an indicator for selective cross-clamp shunting and has been related to MCA blood flow velocity changes during CEA.9 However, its usefulness in predicting intraoperative ischemic stroke is not yet established.
The aims of our study are as follows: (1) to clarify the pathophysiology of perioperative cerebral complications during CEA in our series and (2) to assess the usefulness of stump pressure as an indicator of hemodynamic changes predicting intraoperative cerebral ischemia.
Subjects and Methods
From January 1994 to December 1996, 123 patients underwent CEA for symptomatic or asymptomatic carotid stenosis (70% to 99%) in our institution; 112 of these (91%) were enrolled in this study. TCD recording was not obtained in the other 11 patients (9%) because of an insufficient bone window. Mean age was 64.7±11.2 years, and 71% were male. Forty-nine patients (44%) had recent symptoms related to carotid stenosis, 56 (50%) had an asymptomatic carotid stenosis, and 7 (6%) had cerebral focal symptoms unrelated to the vascular territory of the stenotic carotid artery. Cerebral CT was normal in 75 patients (67%), showed a symptomatic lesion in 26 (23%), and showed silent infarctions in 17 (15%).
Preoperative assessment included neurological examination, cerebral CT, echo color Doppler scanning and subtraction angiography of both carotid and vertebral arteries, TCD of basal cerebral arteries, and cardiac evaluation. All CEAs were done under general anesthesia (nitrous oxide and halothane 0.25% to 0.5%). All patients took preoperative oral antiplatelet agents. During all operations, intravenous heparin (5000 IU) was given before clamping.
Intraoperative TCD monitoring of the ipsilateral MCA was performed by an EME TC 2–64 (Eden Medical Electronicks) instrument with a 2-MHz transducer, at a power of 1000 W/cm2 and a depth of 45 to 55 mm. The probe was placed over the temporal bone and fixed with a headband. All TCD scans were personally monitored by the same operator (C.F.) and were recorded on videotape for further analysis. TCD monitoring was started after the induction of general anesthesia and was prolonged until the final closure of the skin. Special attention was paid to the mean velocity reduction after clamping and to the occurrence of embolic signals (which were identified according to the criteria of Spencer8) in the different phases of the operation (dissection of the carotid artery; cross-clamping; shunting, if performed; release).
Immediately before and 1 minute after clamping, stump pressure was measured by a strain gauge manometer connected through saline-filled tubing to a needle puncturing the common carotid artery. The gauge zero reference was established at the head level of the patient. A cross-clamp shunting was used when one of the following criteria was present: (1) decrease of ipsilateral MCA mean velocity >70% or (2) stump pressure <40 mm Hg, 1 minute after clamping. However, in two patients with a borderline reduction of MCA mean velocity (72% and 76%), the shunt was not applied.
An intraoperative assessment of carotid bifurcation after CEA was routinely performed by means of B-mode echography (linear array, 7.5 MHz). In two patients a residual intimal flap was revealed and was removed by reopening of the arteriotomy.
All patients underwent complete neurological examination on awakening and then periodically during hospitalization. Echo color Doppler scanning of both carotid and vertebral arteries and TCD were performed on day 3 after CEA and cerebral CT or MRI in the case of neurological complications. Patients who had a neurological deficit with complete recovery within 24 hours and normal CT scan were classified as having sustained a TIA, while those with neurological deficits persisting for more than 24 hours were classified as having sustained a stroke. At day 30, of the patients who had a stroke, those with an Oxford Disability Scale10 score of 0 to 2 were classified as having sustained a minor stroke, and those with an Oxford Disability Scale score of 3 to 5 were classified as having sustained a major stroke.
Statistical comparisons were done by means of t test for unpaired data.
No patient died within 30 days of surgery. Three patients suffered a major stroke (2.6%), 3 patients a minor stroke (2.6%), and 2 patients a TIA (1.8%). An intra-arterial shunt was required during 18 CEAs; 2 of these 18 patients (11.1%) developed a cerebral ischemic complication, both during surgery (Table 1⇓). Patient 46 suffered a left minor stroke and showed a critical drop in systemic arterial blood pressure during the shunting phase, which caused a critical reduction in blood flow velocity in the ipsilateral MCA (MCA mean velocity=0 for 3 minutes). Frequent microemboli (25) were recorded during both the dissection and shunting phase and after clamping release. Patient 76 suffered a bilateral ischemic stroke. He had a contralateral occlusion of the internal carotid artery; frequent microemboli (47) were recorded during the shunting phase and after release.
The other 94 CEAs were performed without intra-arterial shunt: 6 patients of 94 (6.4%) showed focal cerebral signs on awakening (Table 2⇓). No other complications were observed within 30 days of surgery. Patient 56 suffered a left nucleocapsular hemorrhage. She showed a large increase in ipsilateral MCA mean velocity after release (139 cm/s, 74% more than the basal value), which was significant for a hyperperfusion syndrome. Patients 4, 6, 23, and 38 suffered ipsilateral TIA or ischemic stroke. Patient 77 suffered a contralateral ischemic stroke; he had an occlusion in the contralateral internal carotid artery.
In the nonshunted group, a statistical comparison was made between the 5 patients who developed cerebral ischemic complications and the uncomplicated cases (n=88). In the 5 complicated cases the duration of clamping was significantly longer (mean±SD, 16.4±1.1 versus 12.7±2.6 minutes; P=.0019), and the decrease of MCA mean velocity on clamping was significantly greater (mean±SD, 56.4±4.9% versus 28.8±20.2%; P=.0031). Combined evaluation of the two parameters allowed us to separate patients with and without cerebral ischemic complications (Figure⇓), suggesting the existence of a biological threshold for clinical complications. No differences were found between the two groups in the values of stump pressure, ipsilateral MCA pulsatility index (baseline and during clamping), or ipsilateral MCA systolic, mean, and diastolic velocity during clamping.
Analysis of microembolic signals was performed by combining data of patients with and without shunt (n=112). Microemboli were recorded in 70 patients (62.5%), with a mean (±SD) of 16.6±17.1 signals per patient and a mean (±SD) duration of TCD monitoring of 108±26 minutes. The prevalence and frequency of microembolic events did not differ between patients with (n=7) and without (n=105) cerebral ischemic complications.
Sixteen patients showed microemboli during the dissection phase (mean±SD, 9.7±9.3 signals per patient); 6 of these had ≥10 emboli. This finding was not significantly associated with cerebral ischemic complications (Table 3⇓).
The last analysis was performed on the whole group and took into account the degree of stenosis in the contralateral carotid artery. The patients who developed cerebral ischemic complications (n=7) had a significantly higher percentage of stenosis in the contralateral internal carotid artery (mean±SD, 82.0±17.8% versus 29.3±36.4%; P=.0018).
The cerebral complications of CEA may be intraoperative (neurological symptoms revealed on awakening) or postoperative (within 30 days of CEA). Such complications are believed to be due to various mechanisms: hemodynamic failure during cross-clamping, thromboembolism during or after CEA, cerebral hemorrhage, and early thrombosis of the internal carotid artery that has undergone CEA.11
The most obvious mechanism of intraoperative cerebral ischemia is hemodynamic compromise of a cerebral hemisphere during cross-clamping. Previous studies12,7 have demonstrated the importance of this mechanism in patients with poor collateral circulation and critical reduction (>70%) in MCA blood flow velocity during clamping.
In our study all patients with critical reduction in MCA mean velocity (>70%) on clamping were shunted to avoid cerebral ischemia during CEA. In the nonshunted group, the 5 patients who developed cerebral ischemia showed both a significantly greater percent reduction in ipsilateral MCA mean velocity and a significantly longer duration of clamping in comparison with the uncomplicated patients.
It appears that a reduction in MCA mean blood flow velocity in the range of 50% to 70% may cause cerebral ischemia if the duration of clamping is sufficiently long. This may prompt surgeons to use a shunt not only when the MCA mean velocity is critically reduced (>70%) but even in the case of a 50% to 70% reduction if they foresee long-duration clamping. Meticulous removal of the plaque and accurate washing of the internal surface of the artery are, of course, other factors that influence safety. We want to emphasize the usefulness of intraoperative echographic assessment of carotid bifurcation, which allowed us to identify and remove two residual intimal flaps.
Most patients with a severe reduction in MCA mean velocity had a low stump pressure. However, in the nonshunted group stump pressure was not significantly different between patients with and without ischemic complications. We conclude, in agreement with previous studies,7 that stump pressure alone is not a reliable indicator of hemodynamic changes that predict cerebral ischemia.
The relevance of degree of stenosis in the contralateral carotid artery as a risk factor for intraoperative complications during CEA is not clear because the evidence in the literature is conflicting.13–16 In our study the patients who suffered from intraoperative ischemic complications showed a significantly higher degree of stenosis in the contralateral internal carotid artery. Two patients (patient 76 in the shunted group and patient 77 in the nonshunted group) with an occlusion of the left internal carotid artery had a left ischemic stroke during CEA of the stenotic right carotid artery. In these patients, preoperative TCD had shown that both left and right MCAs were supplied by the right carotid artery.
Microemboli often occur during CEA, with a prevalence ranging from 69% to 93% in the literature.4,17,18 In our study microembolic signals were recorded in 70 patients (62.5%). The different prevalence reported in the literature can be due to different criteria for the identification of the microembolic signals, different timing of registration, and differences in the use of intravenous heparin during operation and in surgical technique for removal of the plaque. The clinical impact of the detection of embolic signals is not fully known. Many microemboli generated at the time of clamp release may represent bubbles without clinical relevance for the most part, while microemboli generated during the dissection phase are probably particulate. Jansen et al5 showed a significant relationship between the occurrence of >10 embolic signals during dissection of the arteries and the finding of new MR ischemic lesions; no such relationship was found with the occurrence of new CT ischemic lesions. Moreover, according to Gaunt et al,17 the finding of >10 particulate emboli during initial carotid dissection correlates with a significant deterioration in postoperative cognitive functions. In another study, Ackerstaff et al4 showed a significant relationship between microembolism during CEA (>10 emboli during the dissection phase) and the occurrence of clinically evident ischemic complications; these were TIAs. Recently Spencer18 found that a high number of microembolic signals correlated with cerebrovascular complications, with a tendency for embolism to be associated with less severe grades of ischemic complications.
Our patients did not undergo systematic neuroradiological examinations (CT or MRI) after CEA. We did not find a significant correlation between microemboli occurrence and intraoperative ischemic complications. However, a role of embolism in the pathophysiology of cerebral ischemia cannot be excluded in patients 46 and 76. Both presented a high number of microemboli; patient 46 also had a critical drop of arterial blood pressure, with critical reduction of MCA mean velocity during shunting. Patient 76 had a contralateral (left) occlusion of the internal carotid artery and suffered two watershed infarctions in the right hemisphere and an infarction in the territory of the anterior cerebral artery in the left hemisphere. The localization of the ischemic lesions may suggest a hemodynamic pathogenesis; however, a shower of emboli with transhemispheric passage cannot be excluded.19
The results of our study and the review of the literature suggest that the major complications of CEA may be due to hemodynamic factors (critical reduction in MCA blood flow velocity during cross-clamping, or a 50% to 70% reduction in MCA mean velocity if clamping is sufficiently long). Particulate microembolism may cause more subtle changes in cerebral parenchyma and deterioration of cognitive functions. Further studies are needed to clarify this point.
Selected Abbreviations and Acronyms
|MCA||=||middle cerebral artery|
|TCD||=||transcranial Doppler sonography|
|TIA||=||transient ischemic attack|
This study was supported in part by the National Research Council (CNR), Italy.
Reprint requests to Cinzia Finocchi, Department of Neuroscience and Neurological Rehabilitation, University of Genova, Via De Toni 5, 16131 Genova, Italy.
Presented in part at the Joint 3rd World Congress and 5th European Stroke Conference, Munich, Germany, September 1–4, 1996.
- Received March 11, 1997.
- Revision received August 19, 1997.
- Accepted September 15, 1997.
- Copyright © 1997 by American Heart Association
Jansen C, Ramos MLP, Van Heesewijk JPM, Moll FL, Van Gijn J, Ackerstaff RGA. Impact of microembolism and hemodynamic changes in the brain during carotid endarterectomy. Stroke. 1994;25:992–997.
Van Zuilen EV, Moll FL, Vermeulen FEE, Mauser HW, Van Gijn J, Ackerstaff RGA. Detection of cerebral microemboli by means of transcranial Doppler monitoring before and after carotid endarterectomy. Stroke. 1995;26:210–213.
Spencer MP. Detection of cerebral arterial emboli. In: Newell DW, Aaslid R, eds. Transcranial Doppler. New York, NY; Raven Press Publishers; 1992:215–230.
Spencer MP, George IT, Moefbring MA. Relation between MCA blood flow velocity and stump pressure during carotid endarterectomy. Stroke. 1992;23:1439–1445.
Jansen C, Vriens EM, Eikelboom BC, Vermeurlen FEE, Van Gijn J, Ackerstaff RGA. Carotid endarterectomy with transcranial Doppler and electroencephalographic monitoring: a prospective study in 130 operations. Stroke. 1993;24:665–669.
Spencer MP. Transcranial Doppler monitoring and causes of stroke from carotid endarterectomy. Stroke. 1997;28:685–691.
Georgiadis D, Grosset DG, Lees KR. Transhemispheric passage of microemboli in patients with unilateral internal carotid artery occlusion. Stroke. 1993;24:1664–1666.