Transcranial Doppler Monitoring During Carotid Endarterectomy Helps to Identify Patients at Risk of Postoperative Hyperperfusion

Eur J Vasc Endovasc Surg 18, 222–227 (1999) Article No. ejvs.1999.0846

Transcranial Doppler Monitoring During Carotid Endarterectomy Helps to Identify Patients at Risk of Postoperative Hyperperfusion J. E. Dalman1, I. C. M. Beenakkers2, F. L. Moll3, J. A. Leusink2 and R. G. A. Ackerstaff∗1 Departments of 1Clinical Neurophysiology, 2Anaesthesia and 3Vascular Surgery, St. Antonius Hospital, Postbus 2500, 3430 EM Nieuwegein, The Netherlands Objectives: to investigate whether transcranial Doppler (TCD) monitoring can identify patients at risk of hyperperfusion, and whether active postoperative treatment of selected patients decreases the risk of intracerebral haemorrhage (ICH). Design: a case cohort study of 688 patients undergoing carotid endarterectomy (CEA) with intraoperative TCD monitoring. Methods: sixty-two patients (9%) fulfilled the TCD criteria for hyperperfusion, i.e. >100% increase of peak blood flow velocity or pulsatility index of the middle cerebral artery, compared to preclamp baseline values. In these patients, blood pressure was closely monitored and controlled postoperatively. Results: postoperatively, seven of these patients (11%) exhibited clinical signs or symptoms of hyperperfusion but no cerebral haemorrhage (ICH). This is a significantly better outcome (p<0.005) compared to a 2% incidence of ICH after CEA in previous years in our hospital. Conclusions: patients at risk of hyperperfusion syndrome after CEA can be identified intraoperatively by TCD monitoring. In these selected patients, immediate and adequate postoperative treatment of hypertension results in a decreased risk of intracerebral haemorrhage. Key Words: Carotid endarterectomy; Perioperative monitoring; Transcranial Doppler; Hyperperfusion syndrome; Intracerebral haemorrhage.

Introduction Intracerebral haemorrhage (ICH) is an infrequent but serious complication of carotid endarterectomy (CEA). In the literature, the incidence for this complication varies between 0.5–1.0%.1–5 The mortality from intracerebral haemorrhage after CEA is high (40%). Suggested risk factors include recent and old cerebral infarction, postoperative hypertension, the use of antithrombotic or anticoagulant agents, advanced age, contralateral high-grade carotid artery stenosis, and poor collateral flow. One hypothesis concerning the pathogenesis of intracerebral haemorrhage is postoperative cerebral hyperperfusion. This was initially described by Spetzler et al. as the “normal-perfusionpressure-breakthrough” theory as an explanation for the occurrence of oedema and haemorrhage after excision of a cerebral arteriovenous malformation.6 Sundt et al. were the first to describe the hyperperfusion syndrome after CEA.7 According to them cerebral ∗ Please address all correspondence to: R. G. A. Ackerstaff, Department of Clinical Neurophysiology, St. Antonius Hospital, Postbus 2500, 3430 EM Nieuwegein, The Netherlands. 1078–5884/99/090222+06 $12.00/0  1999 Harcourt Publishers Ltd.

hyperperfusion can result in unilateral hypertensive encephalopathy with a triad of symptoms: seizures, migraine variants and intracranial haemorrhage. Since the publication of Sundt et al., several authors have reported on patients with a post-endarterectomy hyperperfusion syndrome.8–16 Breen and co-workers showed that patients with a hyperperfusion syndrome after CEA had, on computer tomography (CT) scans of the brain, severe white matter oedema ipsilateral to the side of surgery.17 They conclude that brain oedema with neurological signs should be included as a serious but potentially reversible component of the postoperative hyperperfusion syndrome. Most authors recommend strict control of blood pressure to prevent these complications. Transcranial Doppler (TCD) ultrasonography can be used to measure blood flow velocities in the middle cerebral artery (MCA) during CEA and there are some reports on the use of TCD monitoring in patients with a postoperative hyperperfusion syndrome.16,18–22 However, prospective studies on the role of transcranial Doppler monitoring in preventing the hyperperfusion syndrome after CEA have not been published. In a previous study of 233 endarterectomies

TCD and Hyperperfusion in Carotid Endarterectomy

with intraoperative TCD monitoring in our institution, the incidence of intracerebral haemorrhage after surgery was analysed.23 This complication was correlated with clinical features and transcranial Doppler parameters. Five patients (2%) developed an ipsilateral intracerebral haemorrhage. During the procedure, at carotid artery clamp release, four of these five patients had either a significant increase of peak blood flow velocity (PFV) in the ipsilateral middle cerebral artery, or an increase of pulsatility index (PI) of the Doppler signal. Statistical analysis led to the conclusion that patients at risk for ICH could reliably be identified based on factors occurring before the event. An increase of peak blood flow velocity >175% or pulsatility index >100% after declamping predicted intracerebral haemorrhage more accurately than the occurrence of headache and hypertension. Based on the results of this study, we changed our perioperative treatment of patients to prevent the occurrence of ICH. The purpose of the present study was twofold: Firstly, to determine if intraoperative TCD monitoring can be used as a reliable modality to identify and select patients at risk for hyperperfusion after CEA. Secondly, to establish if immediate and adequate treatment of hypertension in selected patients results in a decreased risk of intracerebral haemorrhage.

Materials and Methods From November 1992 until February 1998, 789 elective carotid endarterectomies were performed in our institution (ipsilateral stenosis >70%). Patients with a carotid endarterectomy in combination with coronary bypass surgery were excluded. Intraoperative transcranial Doppler monitoring was possible in 475 men (69%) and 213 women, i.e. in 87% of all patients. In the remaining 101 patients TCD monitoring was not possible because of technical or logistical problems. The methods of intraoperative EEG and TCD monitoring have been described elsewhere.24,25 In this study, we were especially interested in the changes of peak blood flow velocities (PFV) and of the Gosling pulsatility indices (PI) of the MCA mainstem on the side of surgery at declamping. Therefore, we measured PFV and PI 1 min before test clamping and for the first 3 min after clamp release, when the desobstruction had been completed. The changes of PFV and PI were calculated as the percentual increase compared to intraoperative preclamp values. We classified patients with a >100% increase of PFV, or >100% increase of PI, or both, as patients at risk for a hyperperfusion syndrome. In the selected patients, blood pressure

223

was closely monitored and controlled with medication intraoperatively after declamping and postoperatively on the medium care unit or the surgical ward. All patients were operated on under general anaesthesia. One hour before start of surgery, 10 mg diazepam was administered as preanaesthetic medication. Anaesthesia was induced with thiopental 4–6 mg, fentanyl 3–5 lg/kg and atracurium 0.5 mg/ kg. The trachea was intubated and all patients were mechanically ventilated. Maintenance of anaesthesia was achieved with 66% nitrous oxide (N2O) in oxygen and repeat doses of fentanyl and atracurium. If needed, low-dose isoflurane was given (0.5–1%). All patients were operated under slightly hypocapnic (end-tidal CO2 concentrations 4.0–5.0%) and hypertensive conditions (blood pressure 10–15% above the level accepted as normal for the patient). Monitoring included heart rate, peripheral oxygen saturation, invasive arterial blood pressure, end-tidal concentrations of isoflurane, nitrous oxide and carbon dioxide. Patients with a history of angina pectoris or myocardial infarction received nitroglycerine intravenously during the operation and in the postoperative period. Three minutes before cross-clamping of the carotid artery, patients received 5000 IU heparin intravenously; protamine reversal was not used. A Javid shunt was used in cases of severe EEG asymmetry and/or a reduction of PFV >70% during test clamping. All patients remained on the medium-care unit for at least 24 h postoperatively. Blood pressure was monitored intra-arterially during this time. In patients at risk of hyperperfusion, a postoperative TCD investigation was performed to determine if cerebral blood flow velocities were within the patient’s preoperative limits and adjustments were made to prevent hyperperfusion. In the event of undesirable high blood pressure (i.e. systolic pressure above 140 mmHg, or above the threshold systolic pressure determined by TCD), blood pressure was actively lowered with intravenous or oral antihypertensive medication. During and directly after the operation, clonidine was used. If needed, this was administered as an intravenous bolus. When necessary, it was continued as boluses intramuscularly on the first postoperative day. If the pulse rate allowed it, then sometimes labetolol was started as a continuous intravenous infusion. After 24 h, oral medication was given. Captopril was the preferred antihypertensive drug. The majority of patients had antithrombotics, aspirin 100 mg daily, which was continued. Only specific cardiac pathology sometimes required anticoagulants. Epileptic seizures were controlled with anticonvulsants. In the event of focal cerebral deficit or seizures, a CT scan of the brain Eur J Vasc Endovasc Surg Vol 18, September 1999

224

J. E. Dalman et al.

Table 1. Characteristics of patients fulfilling the TCD criteria for hyperperfusion syndrome as compared to the control group matched for sex, age and date of operation. Hyperperfusion n=62 Age Male gender Carotid artery Right Left Hypertension Diabetes mellitus Symptomatology Asymptomatic Symptomatic Contralateral carotid artery Occlusion 65–99% stenosis Normal or <65% Shunt

Controls n=61

Mean PFV↑ Mean PI↑

68 68 (range 46–91 years) (range 45–80 years) 40 (65)∗ 37 (61) n.s. 29 33 30 9

(47) (53) (48) (15)

28 33 28 7

(46) (54) (46) (11)

Table 2. Intraoperative increase of mean (95% confidence interval) PFV and PI.

n.s.

Hyperperfusion

Controls

146% (122–170) 68% (47–89)

16% (8–24) 16% (9–23)

Table 3. Postoperative signs and symptoms of the 62 patients fulfilling the TCD criteria for the hyperperfusion syndrome (atrisk group) compared to the control group.

n.s. n.s.

Hyperperfusion n=62

Controls n=61

7 5 2 19 – –

2 (3) – – 10 (16) – –

10 (16) 52 (84)

20 (33) 41 (67)

p<0.005 p<0.005

18 (29) 10 (16) 34 (55)

11 (18) 9 (15) 41 (67)

p<0.05 n.s. p<0.05

Headache Focal cerebral deficit Seizures Postoperative hypertension Oedema Intracerebral haemorrhage

25 (40)

17 (28)

p<0.05

∗ Numbers between parentheses are percentages.

∗ Numbers between parentheses are percentages.

was made. To prevent cerebral oedema, corticosteroids were given intraoperatively to patients at risk of hyperperfusion. Usually, this was done at the discretion of one of our vascular surgeons. Patients identified as being at risk of hyperperfusion were compared to a control group matched for age, sex, and date of operation. The control group was created from the group of 688 CEA patients with TCD monitoring. Patients’ characteristics were analysed to assess significant differences in preoperative symptomatology, vascular risk factors, side of surgery, severity of contralateral internal carotid artery disease, intraoperative shunt use and postoperative morbidity. Statistical evaluation was performed with a v2-test. A p-value of less than 0.05 was considered significant.

Results In the total group of 789 carotid endarterectomies, the combined perioperative stroke and death rate was 3.3%. Of the 688 patients, 62 patients (9%) with a mean age of 68 years (range 46–91) fulfilled the TCD criteria for hyperperfusion. None of these patients developed an intracerebral haemorrhage. The relevant patient characteristics of the at-risk and control groups are summarised in Table 1. There were no significant differences between the two groups concerning the incidence of preoperative hypertension and diabetes mellitus. In the control group, there were significantly more asymptomatic patients. The incidence of an occlusion of the contralateral carotid artery and the use Eur J Vasc Endovasc Surg Vol 18, September 1999

(11)∗ (8) (3) (31)

p<0.005 n.s. n.s. p<0.005 – –

of a shunt intraoperatively was significantly higher in the hyperperfusion group. In the hyperperfusion group the mean (95% confidence interval) increase of the PFV and PI in the ipsilateral MCA were respectively 146% (122–170) and 68% (47–89) (Table 2). The mean increase of PFV and PI in the control group were respectively 16% (8–24) and 16% (9–23). Postoperatively, seven patients (11%) of the at-risk group had clinical signs or symptoms of hyperperfusion (Table 3). Two of the patients had seizures, occurring on the 5th and 7th postoperative day, and five patients (8%) had focal cerebral symptoms. Only one of these patients experienced headache ipsilateral to the side of surgery. Besides, six other patients also had headache, but no focal cerebral deficit. Nineteen patients (31%) had postoperative hypertension. None of the seven patients with focal cerebral deficit and seizures had a striking number of intraoperative cerebral microemboli. In the control group there were no patients with intracerebral oedema or haemorrhage, two patients (3%) had headaches and 10 patients (16%) postoperative hypertension. There were no patients with seizures, focal neurologic deficit, intracerebral oedema or haemorrhage. The occurrence of postoperative hypertension and headache was significantly different in the two groups. One patient in the hyperperfusion group developed a hemiparesis contralateral to the side of surgery 1 h after completion of the operation. At re-exploration, the internal carotid artery was occluded by a fresh thrombus. After removal of the thrombus, both PFV and PI increased >100% at clamp release. Postoperatively, the patient still had a hemiparesis, and a CT scan of the brain showed a large infarction in the

TCD and Hyperperfusion in Carotid Endarterectomy

MCA territory, with oedema. The clinical symptoms of this patient were probably due to a thromboembolic stroke after the first operation and were not the consequence of a hyperperfusion syndrome after the second intervention. Nonetheless, we decided to include this patient, because he was certainly at risk for hyperperfusion, with a recently infarcted hemisphere.

Discussion Sundt et al. were the first to describe the hyperperfusion syndrome after CEA.7 They analysed 1145 carotid endarterectomies. Regional cerebral blood flow (rCBF) measurements were performed in all patients. Postoperatively, 20 patients exhibited signs or symptoms of hyperperfusion, and five of these patients developed intracerebral haemorrhage. All patients had an increase of rCBF >100% in comparison with baseline values. According to the authors, this suggested paralysis of cerebral autoregulatory mechanisms. The brain was incapable of controlling its blood flow and became passively dependent on the perfusion pressure. Therefore, in a previously hypoperfused area with impaired cerebral autoregulation, restoration of normal perfusion pressure after CEA can lead to hyperperfusion. Breen et al. described five patients (2.7%) with severe ipsilateral white matter oedema after carotid surgery.17 All patients presented 5–8 days after surgery with hypertension, headaches, seizures and focal neurologic symptoms. One patient died from herniation secondary to massive oedema. In the other patients, the CT abnormalities and neurologic signs were resolved within 3 weeks. Breen et al. concluded that brain oedema with focal neurologic signs should be included as a serious but potentially reversible component of the postoperative hyperperfusion syndrome. The hyperperfusion syndrome is a clinical diagnosis based on a number of non-specific signs and symptoms. It is important to diagnose patients as soon as possible, preferably before the occurrence of intracerebral oedema or haemorrhage. Headache is a very common, but subjective, phenomenon and occurs in a lot of patients after carotid surgery. Seizures and focal neurological deficit can have other underlying causes. Reigel et al. identified periodic lateralised epileptiform discharges (PLEDs) on EEG as “a sensitive but nonspecific objective indicator” for the hyperperfusion syndrome.11 In fact, PLEDs are indicative of epileptic activity and not particularly helpful in identifying patients at risk for a hyperperfusion syndrome. High-grade carotid artery stenosis, poor collateral flow and contralateral occlusion have been identified

225

as risk factors for the hyperperfusion syndrome. McCarthy et al. concluded that patients with an occlusion of the contralateral internal carotid may have a greater risk of developing an ICH after CEA.26 In our patient group, which fulfilled the TCD criteria of hyperperfusion, there was a higher incidence of contralateral occlusion. Nevertheless, some patients who have no known risk factors still develop a hyperperfusion syndrome. Therefore, selecting patients at risk for hyperperfusion and ICH on the basis of preoperative risk factors is not reliable. Several reports underline the role of TCD in selecting patients at risk of a postoperative hyperperfusion syndrome.16,18–22 Powers et al. made a diagnosis of hyperperfusion syndrome with serial TCD in two patients.16 Sbarigia et al. and Naylor et al. showed that patients with a reduced preoperative cerebrovascular reserve capacity had a significant increase in MCA blood flow velocities on the side of surgery compared to basal preoperative values and, therefore, were at risk for a hyperperfusion syndrome.18,20 Chambers et al. could not predict hyperperfusion from TCD criteria.19 They did not, however, perform intraoperative TCD monitoring. Jørgensen et al. showed that, in patients with a postendarterectomy hyperperfusion syndrome, ipsilateral MCA mean flow velocities were pressure-dependent. Reduction of arterial pressure leads to normalisation of flow velocities and resolved episodes of headache and seizures in symptomatic patients.22 We based our cut-off values on the results of the studies of Sundt, Piepgras, and Jansen.3,7,23 Piepgras et al. did a retrospective analysis of 2362 patients undergoing CEA.3 To identify the role of hyperperfusion, regional cerebral blood flow was assessed during and after surgery. Hyperperfusion was defined as “at least 100% increase of baseline cerebral blood flow”, and 11.6% of the patients showed such an increase. Fourteen patients developed an ICH, and nine of these patients had a more than 100% increase of rCBF. In our study, using a combination of cut-off values of TCD parameters, 9% of the patients fulfilled the criteria of a postoperative hyperperfusion syndrome. With strict postoperative control and treatment of hypertension in this selected group of patients, no patient developed an ICH. Moreover, in the matched control group no patient exhibited clinical signs or symptoms of hyperperfusion syndrome. In other words, all patients clinically at risk were correctly identified. On the other hand, of the 62 patients who fulfilled our TCD criteria and who were strictly monitored and treated after surgery, only seven finally developed clinical signs and symptoms of a hyperperfusion syndrome and no patients had an ICH. Eur J Vasc Endovasc Surg Vol 18, September 1999

226

J. E. Dalman et al.

When aiming to prevent hyperperfusion syndrome after carotid surgery, anaesthesia and perioperative treatment of hypertension are very important. Anaesthetic agents have different effects on cerebral blood flow and autoregulation. Isoflurane is the volatile anaesthetic of choice in neurosurgical operations because of comparatively less vasodilating effects than other halogenated anaesthetics at equipotent doses. Strebel et al. showed that, when more than 2 MAC isoflurane is given, cerebral autoregulation will be disturbed.27 However, the effect of isoflurane on cerebral autoregulation is dose-dependent and below 1 MAC minimal. In our patients only doses below 1 MAC were given. Nitrous oxide gives a small rise in cerebral blood flow, intracranial pressure, and cerebral blood volume. Less than 70% nitrous oxide probably has no influence on cerebral autoregulation. Nitrous oxide combined with volatile anaesthetics in different concentrations has dose-dependent vasodilational effects on cerebral blood flow. If blood pressure drops, the effects are different and unpredictable. Most authors recommend strict control of blood pressure in the postoperative period. To prevent adverse effects on cerebral blood flow, we changed our postoperative antihypertensive regimen in patients at risk for hyperperfusion. Therefore, we preferred drugs that give cerebral vasoconstriction and not cerebral vasodilation. We used clonidine as an antihypertensive drug during and after the operation.28 Labetolol was sometimes used in patients without bradycardia. Labetolol is a selective postsynaptic alpha1-antagonist and a non-selective beta-antagonist that has no direct effects on cerebral blood flow.29 Blood pressure regulation is recommended until cerebral autoregulation is restored. In our hospital, we treat patients for 6 months after surgery with oral antihypertensive medication, preferably captopril. In conclusion, TCD monitoring is a practical and sensitive method that can identify patients who are at risk for hyperperfusion after CEA. The increase of PFV and PI after clamp release has a low specificity, but a high sensitivity, for predicting patients at risk for hyperperfusion. We have been able to reduce the occurrence of intracerebral haemorrhage by careful blood-pressure monitoring and control in at-risk patients, changing our perioperative treatment regimen concerning blood pressure.

References 1 Pomposelli FB, Lamparello PJ, Riles TS et al. Intracranial hemorrhage after carotid endarterectomy. J Vasc Surg 1988; 7: 248–255.

Eur J Vasc Endovasc Surg Vol 18, September 1999

2 Schroeder TV, Sillesen H, Boesen J, Laursen H, Soelberg Sorensen P. Intracerebral haemorrhage after carotid endarterectomy. Eur J Vasc Surg 1987; 1: 51–60. 3 Piepgras DG, Morgan MK, Sundt TF, Yanagihara T, Mussman LM. Intracerebral hemorrhage after carotid endarterectomy. J Neurosurg 1988; 68: 532–536. 4 Bruetman ME, Fields WS, Crawford ES, DeBakey ME. Cerebral hemorrhage in carotid artery surgery. Arch Neurol 1963; 9: 26–35. 5 Solomon RA, Loftus CM, Quest DO, Correll JW. Incidence and etiology of intracerebral hemorrhage following carotid endarterectomy. J Neurosurg 1986; 64: 29–34. 6 Spetzler RF, Wilson ChB, Weinstein Ph et al. Normal perfusion pressure breakthrough theory. Clin Neurosurg 1978; 25: 651–672. 7 Sundt TM, Sharbrough FW, Piepgras DG et al. Correlation of cerebral blood flow and electroencephalographic changes during carotid endarterectomy. With results of surgery and hemodynamics of cerebral ischemia. Mayo Clinic Proceedings 1981; 56: 533–543. 8 Harrison PB, Wong MJ, Belzberg A, Holden J. Hyperperfusion syndrome after carotid endarterectomy. CT changes. Neuroradiology 1990; 33: 106–110. 9 Bernstein M, Fleming JFR, Deck JHN. Cerebral hyperperfusion after carotid endarterectomy: a cause of cerebral hemorrhage. Neurosurgery 1984; 15: 50–56. 10 Penn AA, Schomer DF, Steinberg GK. Imaging studies of cerebral hyperperfusion after carotid endarterectomy. J Neurosurg 1995; 83: 133–137. 11 Reigel MM, Hollier LH, Sundt TM et al. Cerebral hyperperfusion syndrome: a cause of neurologic dysfunction after carotid endarterectomy. J Vasc Surg 1987; 5: 628–634. 12 Schroeder T, Sillesen H, Sørensen O, Engell HC. Cerebral hyperperfusion following carotid endarterectomy. J Neurosurg 1987; 66: 824–829. 13 Harten van B, Gool van WA, Bienfait HME, Stam J. Brain edema after carotid endarterectomy. Neurology 1997; 48: 544–545. 14 Harten van B, Gool van WA, Legemate DA. Het cerebrale hyperperfusiesyndroom na carotisendarteriectomie. Ned Tijdschr Geneeskd 1997; 141: 2461–2464. 15 Mansoor GA, White WB, Grunnet M, Ruby ST. Intracerebral hemorrhage after carotid endarterectomy associated with ipsilateral fibrinoid necrosis: a consequence of the hyperperfusion syndrome? J Vasc Surg 1996; 23: 147–151. 16 Powers AD, Smith RR. Hyperperfusion syndrome after carotid endarterectomy: a transcranial Doppler evaluation. Neurosurgery 1990; 26: 56–60. 17 Breen JC, Caplan LR, DeWitt LD et al. Brain edema after carotid surgery. Neurology 1996; 46: 175–181. 18 Naylor AR, Whyman MR, Wildsmith JAW et al. Factors influencing the hyperaemic response after carotid endarterectomy. Br J Surg 1993; 80: 1523–1527. 19 Chambers BR, Smidt V, Koh P. Hyperperfusion postendarterectomy. Cerebrovasc Dis 1994; 4: 32–37. 20 Sbarigia E, Speziale F, Giannoni MF, Colonna M, Panico MA, Fiorani P. Post-carotid endarterectomy hyperperfusion syndrome: preliminary observations for identifying at risk patients by transcranial Doppler sonography and the acetazolamide test. Eur J Vasc Surg 1993; 7: 252–256. 21 Magee TR, Davies AH, Horrocks M. Transcranial Doppler evaluation of cerebral hyperperfusion syndrome after carotid endarterectomy. Eur J Vasc Surg 1994; 8: 104–106. 22 Jørgensen LG, Schroeder TV. Defective cerebrovascular autoregulation after carotid endarterectomy. Eur J Vasc Surg 1993; 7: 370–379. 23 Jansen C, Sprengers AM, Moll FL et al. Prediction of intracerebral haemorrhage after carotid endarterectomy by clinical criteria and intraoperative transcranial Doppler monitoring: results of 233 operations. Eur J Vasc Surg 1994; 8: 220–225. 24 Jansen C, Vriens EM, Eikelboom BC et al. Carotid endarterectomy with transcranial Doppler and electroencephalographic monitoring, a prospective study in 130 operations. Stroke 1993; 24: 665–669.

TCD and Hyperperfusion in Carotid Endarterectomy

25 Jansen C, Moll FL, Vermeulen FEE et al. Continuous transcranial Doppler ultrasonography and electroencephalography during carotid endarterectomy: a multimodal monitoring system to detect intraoperative ischemia. Ann Vasc Surg 1993; 7: 95–101. 26 McCarthy WJ, Wang R, Pearce WH, Flinn WR, Yao JST. Carotid endarterectomy with an occluded contralateral carotid artery. Am J Surg 1993; 166: 168–172. 27 Strebel S, Kaufman M, Anselmi L, Schaefer HG. Nitrous oxide is a potent cerebrovasodilator in humans when added to isoflurane. Acta Anesthesiol Scand 1995; 39: 653–658.

227

28 Lee H-W, Caldwell JE, Dodson B, Talke P, Howley J. The effect of clonidine on cerebral blood flow velocity, carbon dioxide cerebral vasoreactivity, and response to increased arterial pressure in human volunteers. Anesthesiology 1997; 87: 553–558. 29 Muzzi D, Black S, Locasso TJ, Cucchiara RF. Labetolol and esmolol in the control of hypertension after intracranial surgery. Anesth Analg 1990; 70: 68–71. Accepted 22 January 1999

Eur J Vasc Endovasc Surg Vol 18, September 1999