Departments of Anesthesiology (PAM, RT) and Neurological Surgery (RT, KMR, RLG), Washington University School of Medicine, St. Louis, Missouri Neurosurgery 30; 842-846, 1992 ABSTRACT: During carotid endarterectomy (CEA), phenylephrine infusions are commonly used to induce hypertension during carotid clamping in an attempt to increase collateral cerebral blood flow and prevent cerebral ischemia. Although this practice appears to increase the incidence of intraoperative myocardial ischemia during CEA when general anesthesia is employed, whether the limited use of phenylephrine infusions in specific instances of cerebral ischemia, as shown on an electroencephalogram, results in low perioperative rates of both myocardial infarction (MI) and cerebral infarction remains unclear. We studied 171 CEAs done under general anesthesia performed with selective shunting based on the identification of cerebral ischemia by a two-channel computerized electroencephalographic monitor. The use of a phenylephrine infusion was restricted to the following instances of cerebral ischemia: 1) ischemia associated with hypotension that did not resolve within 2 minutes of decreases in anesthetic administration and treatment with fluid and/or colloid; 2) ischemia poorly or slowly responsive to shunt placement, accompanied by either hypo- or normotension; and 3) ischemia poorly or slowly responsive to removal of the carotid clamp, accompanied by either hypo- or normotension. Two non-Q wave MIs (1.2%) occurred, both nonfatal. There were two cerebral infarctions (1.2%) and three deaths not related to MI (1.8%). Based on these findings, in order to decrease the incidence of both MI and cerebral infarction after general anesthesia for CEA, we recommend the restrictive use of phenylephrine-induced hypertension for specific instances of slowly or poorly reversible cerebral ischemia, as shown on the electroencephalogram. KEY WORDS: Alpha-adrenergic agonists; Carotid endarterectomy; Computed electroencephalography; Myocardial infarction; Phenylephrine INTRODUCTION Myocardial infarction (MI) continues to be a major cause of morbidity and mortality after carotid
PATIENTS AND METHODS Patient characteristics One hundred fifty-nine patients undergoing 171 CEAs were studied prospectively. The study protocol was approved by the Institutional Review Board of the Washington University School of Medicine, and informed consent was obtained. The patient population consisted of 117 men and 54 women, with an average age of 64 ± 9 years (range, 36-82 years).
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/30/6/842/2751766 by East Carolina University user on 11 January 2019
AUTHOR(S): Modica, Paul A., M.D.; Tempelhoff, Rene, M.D.; Rich, Keith M., M.D.; Grubb, Robert L., Jr., M.D.
endarterectomy (CEA) (3,18,28). The extent to which anesthetic management may influence the incidence of MI after CEA remains unclear. Neurological assessment of the awake patient using local anesthesia provided by a cervical plexus block allows for sensitive detection of cerebral ischemia. In many studies, the use of local anesthesia for CEA has resulted in relatively low rates of postoperative stroke (0.66-2.5%) (9,11,17,22,30) and MI (02.3%) (11,17,18,23,30). Since CEA performed under local anesthesia requires light levels of sedation for optimal neurological monitoring, patient discomfort and anxiety can sometimes occur. Under these conditions, CEA may be poorly tolerated by both the patient and the surgeon, especially during long procedures or in those complicated by more superior carotid bifurcations requiring upper neck exposure. Furthermore, should the airway become compromised by either excessive sedation or neurological insult, proper anesthetic management becomes exceedingly more difficult. On the other hand, general anesthesia, although more acceptable to many patients and surgeons, increases the difficulty of monitoring cerebral function, and has been associated with more variable, and sometimes higher, rates of postoperative stroke (1-6.7%) (15,17,28,29) and MI (1,3,5,16,17). In one series of CEAs performed under general anesthesia, a 12.9% incidence of MI was reported for a high-risk subgroup of patients with symptomatic coronary artery disease (5). Independent of the choice of anesthesia, it is common practice in many institutions to use infusions of alpha-adrenergic agonists (i.e., phenylephrine) to induce hypertension pharmacologically during carotid cross-clamping, in an attempt to increase collateral cerebral blood flow and prevent cerebral ischemia. This practice may, however, increase the incidence of intraoperative myocardial ischemia during CEA when general anesthesia is employed (26). Based on this association between the administration of alpha-adrenergic agonists and myocardial dysfunction, we employed computerized electroencephalographic (CEEG) monitoring and selective shunting during general anesthesia for CEA to restrict the intraoperative use of phenylephrine infusions to specific instances of slowly or poorly reversible cerebral ischemia on the electroencephalogram (EEG). This study attempted to produce a low rate of perioperative MI while limiting the incidence of postoperative stroke by avoiding pharmacologically induced hypertension during CEA.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Neurosurgery 1992-98 June 1992, Volume 30, Number 6 842 Computerized Electroencephalographic Monitoring and Selective Shunting: Influence on Intraoperative Administration of Phenylephrine and Myocardial Infarction after General Anesthesia for Carotid Endarterectomy Experimental and Clinical Study
Intraoperative management: Technique of general anesthesia and criteria for phenylephrine infusion and placement of carotid bypass shunt With each patient receiving 100% oxygen (O2) by mask, general anesthesia was slowly induced with either thiopental (3-4 mg/kg, intravenously) or etomidate (Amidate, Abbott, North Chicago, IL) (0.31.35 mg/kg, intravenously). After facilitation of endotracheal intubation with vecuronium (Norcuron, Organon, West Orange, NJ) (0.1-0.2 mg/kg, intravenously), anesthesia was maintained in all patients with a mixture of 70% nitrous oxide (N2O) in O2, low-concentration (0.1-0.3% end-tidal) isoflurane (Forane, Anaquest, Madison, WI) and a variable-rate (1-2 µg/kg/h) infusion of fentanyl, each titrated so as to achieve sensitive EEG/CSA tracings while maintaining systolic blood pressure (SBP) within 20% of each patient's baseline ward value. Baseline ward SBP was defined as the mean of the values obtained by ward nurses during the first 24 hours after hospital admission. Hypertension (i.e., increase in SBP > 20% of baseline) was treated by increasing isoflurane (to 0.3% end-tidal) and/or fentanyl (to 2 µg/kgh), and administering vasodilators (e.g., nitroglycerin, intravenously) if needed. Conversely, hypotension (i.e., decrease in SBP > 20% of baseline) without signs of a concomitant ischemic EEG event was treated by decreasing isoflurane (to 0.1% endtidal) and/or fentanyl (to 1 µg/kg/h), and administering infusions of isotonic fluid (e.g., normal saline) and/or colloid (e.g., 5% albumin) if needed. Additional monitoring during anesthesia included an ECG, pulse oximetry, use of an esophageal stethoscope with temperature probe, mass spectrometry, and radial artery cannulation for continuous measurement of arterial blood pressure. Ventilation was mechanically controlled to maintain arterial pCO2 at 35-45 mm, as measured by intermittent blood gas analysis. In all 171 cases, heparin (5,000 units, intravenously) was administered 4 minutes before cross-clamping of the carotid arteries. An infusion of phenylephrine was administered only when the following criteria were met: 1. An ischemic EEG event associated with hypotension occurred and did not resolve within 2
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Electroencephalography Before the induction of anesthesia, a commercially available CEEG monitor (Interspec-Neurotrac, Conshohocken, PA) with compressed spectral array (CSA) capability was employed. This monitor displays two channels (left and right cerebral hemispheres) of raw EEG and CSA data recorded from five subdermal, fine-gauge, platinum needle electrodes (Grass, Quincy, MA). These electrodes were placed in a symmetrical frontal-mastoid montage that closely corresponds to F7T5 and F8-T6, in the international 10-20 System. Details of our method of CEEG monitoring, including electrode placement, have been described previously (14,29) . Throughout each procedure, a continuous on-line, two-channel, raw EEG tracing was obtained with a simultaneous printout of the CSA plus spectral edge frequency (SEF) using a frequency range of 1 to 30 Hz displayed in 2-second epochs. The SEF marker represents the highest significant frequency present in the sampled EEG spectrum. To achieve a sensitive EEG/CSA tracing, the depth of anesthesia was carefully controlled to produce a delta, theta, low alpha pattern with an SEF of 12 to 15 Hz recorded at a gain ranging between 20 to 80 µV. An ischemic EEG event (29) was defined as 1. a more than one-third amplitude attenuation of the raw EEG (by visual inspection of the printed tracing compared with both the other hemispheric
signal and earlier recordings) sustained for more than 30 seconds, and/or 2. a unilateral decrease in the SEF to less than 50% of its average value (over the preceding minute) sustained for more than 30 seconds. All detected ischemic EEG events were subclassified into the following three types (29): 1. Early ischemic EEG event: Any ischemic EEG event detected during the period of carotid crossclamping, when it was still technically possible to place a bypass shunt. 2. Late ischemic EEG event: Any ischemic EEG event detected late in the procedure, when it was no longer technically feasible to place a shunt (e.g., arteriotomy closure, shunt removal). 3. Ischemic EEG event associated with hypotension.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/30/6/842/2751766 by East Carolina University user on 11 January 2019
All patients received detailed medical and neurological evaluation before surgery. The preoperative medical findings are presented in Table 1. Previous history of MI was based upon the clinical history, an electrocardiogram (ECG), or the levels of the cardiac isoenzymes creatinine kinase (CK)-MB or lactic dehydrogenase (LDH). The patients' neurological history before surgery is presented in Table 2. Preoperative history of either a recent (≤6 weeks) or distant (>6 weeks) cerebral infarction was based on the neurological history, clinical examination, and computed tomographic findings for each patient. Cerebral angiography was performed in all patients before surgery. Of the 14 patients studied who had asymptomatic stenosis of the internal carotid artery, 12 had previously treated symptomatic lesions of the contralateral ICA. Grading of the degree of surgical risk was carried out according to the Mayo Clinic classification of Sundt et al. (28) as follows: Grade 1: neurologically stable patients with no major medical or angiographically defined risks and with unilateral or bilateral ulcerative-stenotic carotid disease; Grade 2: neurologically stable patients with no major medical risks but with significant angiographically determined risks; Grade 3: neurologically stable patients with major medical risks with or without significant angiographically determined risks; and Grade 4: major neurological risks, with or without associated major medical or angiographically determined risks.
RESULTS Table 3 summarizes the morbidity and mortality in this series of 171 CEAs according to the Mayo Clinic grading of risk (28). No fatal MIs (0%) occurred. There were two nonfatal postoperative MIs (1.2%), both in Grade 3 patients, each associated with 1 to 2 mm of ST segment depression on ECG and confirmed by significant increases in CK-MB, as Q waves were never evident on their serial ECGs. The first MI was accompanied by confusion, tachypnea, and mild pulmonary edema on the third postoperative day, whereas the second MI was associated with chest tightness and dyspnea on the first postoperative day. Neither of these 2 patients who suffered a postoperative MI received an intraoperative infusion of phenylephrine, although both suffered an early ischemic EEG event that resolved shortly after shunt placement.
DISCUSSION In this report of 171 CEAs performed with general anesthesia and CEEG-guided shunt placement, selective use of phenylephrine infusion for specific instances of slowly or poorly reversible cerebral ischemia, as shown on CEEG, was associated with only two perioperative MIs, both nonfatal, and a 1.2% incidence of postoperative cerebral infarction. Our 0% rate of fatal MI is identical to that found in previous reports of CEA performed using local anesthesia, including those that also commonly used induced hypertension with phenylephrine during carotid clamping (11,18,30). In contrast to our findings and those for local anesthesia described above, the combined use of general anesthesia and phenylephrine infusion during CEA leads to a greater incidence of intraoperative myocardial ischemia (26). This may explain the often higher rates of MI, especially fatal MI, associated with general anesthetic techniques for CEA (3,13,16,17). Previous studies have reported a decrease in ventricular wall stress when a decrease in blood pressure was associated with an increasing concentration of inhalant anesthetics (24). In a report of 60 CEAs analyzed with transesophageal echocardiography, the maintenance of systemic blood pressure with deep--1.4 minimum alveolar concentration (MAC) (4)--inhalation anesthesia (50% N20 in O2 accompanied by either halothane or isoflurane) and phenylephrine was compared with lighter inhalation anesthesia without phenylephrine with regard to new segmental wall motion and wall thickening abnormalities (26). Increased wall stress was associated with a threefold greater incidence of myocardial ischemia, as evidenced by new segmental
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Diagnosis of new postoperative myocardial infarctions and neurological deficits and statistical analysis Beginning in the recovery room upon emergence from anesthesia, all study patients were questioned about symptoms of chest pain and shortness of breath; observed for hypertension, hypotension, arrhythmias (including tachycardia), and altered mental status. In addition, they were examined for the presence of new postoperative motor, sensory, or speech disorders. Thereafter, these postoperative medical and neurological evaluations were performed two or more times daily until discharge for at least 4 and up to 7 days postoperatively. Serial 12-lead ECGs and CK-MB isoenzymes were obtained daily for 3 postoperative days for any of the above signs or symptoms suggestive of myocardial ischemia (7). The diagnosis of perioperative MI was made by an independent cardiologist and was based on a rise in blood levels of CK-MB isoenzyme greater than 12 IU/L. New neurological deficits persisting for 24 hours or more were classified as permanent deficits, that is, cerebral infarction.
In the 171 CEAs analyzed, there were 35 patients in whom ischemic EEG events occurred. Of these 35 patients, 12 were treated with an infusion of phenylephrine at some time intraoperatively, and none suffered a permanent deficit. In the 23 remaining patients in whom ischemic EEG events were detected, treatment with phenylephrine was not required at any time intraoperatively. In 1 of these 23 cases, a late EEG event accompanied by significant hypertension (i.e., SBP > 20% of baseline) occurred shortly after release of carotid cross-clamping. The patient who suffered this ischemic EEG event awoke with a deficit that was permanent and did not recover. He was classified as a Grade 4 patient, and died 3 days later from an autopsy-proven pulmonary embolism. Overall, two patients (1.2%) were left with permanent deficits postoperatively and three (1.8%) died, including the Grade 4 patient described above. The second permanent deficit and death occurred in a Grade 1 patient who initially was neurologically intact. After approximately 12 hours, however, he suffered a cerebral infarction due to acute carotid occlusion and died 4 days later. Finally, the third death occurred 6 days postoperatively as the result of an acute subdural hematoma in a Grade 4 patient treated with heparin for 48 hours for unstable angina. Serial ECGs and cardiac (CK-MB) isoenzymes were not significant for MI in this patient.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/30/6/842/2751766 by East Carolina University user on 11 January 2019
minutes after isoflurane and fentanyl were decreased and isotonic fluid and/or colloid treatment was instituted. 2. An early ischemic EEG event that did not resolve within 2 minutes after shunt placement occurred, accompanied by either hypotension (i.e., a decrease in SBP > 20% of the baseline that responded poorly to decreasing isoflurane and fentanyl and instituting treatment with fluid and/or colloid) or SBP no higher than 20% above the baseline. The decision to place a bypass shunt during the period of carotid cross-clamping was based on the presence or absence of an early ischemic EEG event on the two-channel EEG/CSA monitor. 3. A late ischemic EEG event that did not resolve within 2 minutes of cross-clamp removal occurred, accompanied by either hypotension (i.e., a decrease in SBP > 20% of baseline that responded poorly to decreasing isoflurane and fentanyl and instituting treatment with fluid and/or colloid) or SBP no higher than 20% above the baseline.
Received for publication, August 14, 1991; accepted, September 18, 1991. Presented in part at the 65th Annual Meeting of the International Anesthesia Research Society, San Antonio, Texas, February 9, 1991. Reprint requests: R. Tempelhoff, M.D., Department of Anesthesiology, Division of Neuroanesthesia, Washington University School of Medicine, Box 8054, 660 S. Euclid Avenue, St. Louis, MO 63110. REFERENCES: (1-30)
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
symptomatic (17-55%) and asymptomatic (49-66%) coronary artery disease in patients scheduled for CEA (5,8,16) . Some recent investigations implicate isoflurane and/or N2O, the inhalation anesthetics used in our study, as causes of myocardial ischemia in patients with coronary artery disease (20,21), leading us to question whether their use during CEA is appropriate. Yet, various investigations, including this study, do not document an increase in adverse myocardial effects during CEA employing N2O and isoflurane (3, 26) . In a review of 2,223 CEAs, a fatal MI rate of 0.25% was found in the 784 CEA patients receiving a 1.0-MAC mixture of isoflurane in 50% N2O and O2, a significantly lower incidence as compared with the 1,439 other CEAs performed with different anesthetic agents (3). Using transesophageal echocardiography during CEA, Smith et al. (26) found no difference in the incidence of myocardial ischemia between patients anesthetized with isoflurane and those receiving halothane, each at similar MAC levels with 50% N2O and O2. Also, the addition of N2O to isoflurane did not increase the incidence of myocardial ischemia, as measured by transesophageal echocardiography and 12-lead ECG, in patients undergoing carotid artery surgery (10). In view of these reports, as well as our finding of only two MIs in 171 CEAs, both of which were nonfatal, it appears that whatever effect N2O and isoflurane might have in patients with coronary artery disease, it is not clinically evident in patients undergoing CEA. During both normocapnic and hypocapnic states under general anesthesia for CEA, pharmacologically induced hypertension to a level approximately 20% higher than preoperative SBP produced regional cerebral blood flow and carotid artery back (stump) pressure responses during carotid clamping that could not be predicted in individual cases (2). These findings indicate that in patients with cerebrovascular disease, pressure cannot be equated with flow (12). Thus, the routine use of induced hypertension without concomitant EEG signs of cerebral ischemia cannot be advocated, especially in view of its potentially hazardous consequences of myocardial ischemia during general anesthesia in this population. Based on these previous findings and on the results of our study, in order to decrease the incidence of both MI and cerebral infarction after general anesthesia for CEA with EEG monitoring, we recommend the restrictive use of phenylephrine-induced hypertension for specific instances of slowly or poorly reversible cerebral ischemia as detected by EEG.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/30/6/842/2751766 by East Carolina University user on 11 January 2019
wall motion and wall thickening abnormalities, and was found in the patients who received phenylephrine (26) . An additional factor that may also contribute to the development of myocardial ischemia in patients treated with phenylephrine during CEA is the effect of alpha-adrenergic coronary vasoconstriction (6). In view of these findings and our results, the restrictive use of a phenylephrine infusion for specific instances of slowly or poorly reversible cerebral ischemia as detected by EEG appears to decrease the occurrence of perioperative MI during general anesthesia for CEA. It is possible that the low rate of perioperative MI in this study may have been the result of a lower severity of preoperative cardiovascular and neurological characteristics in our patient population; however, the mean age of our patients, 64 years, is close to the range (62-64 years) of other reports (1,5,17, 18) . Also, only 8.2% of the CEAs in this series were performed for asymptomatic carotid stenosis, a finding similar to or less than that of previous reports (1,11,17,18) , and most were performed as a second procedure in patients with previously treated symptomatic lesions of the contralateral internal carotid artery. The incidence of preoperative hypertension (71.9%) in our study is also close to or higher than that of earlier reports (1,5,11,18,23). A previous MI was reported in 20.5% of our patients, a finding similar to the range (7-24%) reported in other, comparable series (1,5,11,18). In addition, 42.7% of our CEAs were associated with a history of angina, a percentage almost identical to that recorded in earlier studies (41-44%) (16,18). Finally, over three fourths (76%) of our CEAs were performed in patients in Mayo Clinic Grades 3 and 4, a much higher incidence of patients with major medical and/or neurological risks as compared with the CEA patients reported on by Sundt et al. (28). Overall, when compared with patients in other reports, the patients in our CEA series appear to have been at the same or higher risk for postoperative complications. In this study, one of the two MIs presented without pain, a finding consistent with those of earlier reports, in which approximately 50% of the MIs following surgery were also painless (19,25,27). There is still the possibility that undiagnosed silent MIs may have occurred in some of the remaining 169 patients we studied. Based on the recommendation of Goldman (7) , we were alert for signs and symptoms other than chest pain suggestive of myocardial ischemia and obtained serial ECGs and CK-MB isoenzymes for 3 postoperative days whenever the possibility of an MI was felt to be present. Also, our study patients were monitored for 4 to 7 days postoperatively, which is consistent with the findings of the most recent prospective studies showing that perioperative MIs always occur within 4 postoperative days after noncardiac surgery (19,25). The high mortality associated with postoperative MI, ranging from 23% (16,25) to 69% (27), also makes it unlikely that many painless MIs could have occurred in our series without there being a considerable increase in the number of postoperative deaths. It is well known that there is a high incidence of
2.
4.
5.
6. 7. 8.
9. 10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22. 23.
24.
25.
26.
Modica PA, Tempelhoff R: A comparison of computerized EEG with internal carotid artery stump pressure for detection of ischemia during carotid endarterectomy. J Neurosurg Anesth 1:211-218, 1989. Morawetz RB, Zeiger HE, McDowell HA Jr, McKay RD, Warner PD, Gelman S, Halsey JH: Correlation of cerebral blood flow and EEG during carotid occlusion for endarterectomy (without shunting) and neurologic outcome. Surgery 96:184-189, 1984. O'Donnell TF Jr, Callow AD, Willet C, Payne D, Cleveland RJ: The impact of coronary artery disease on carotid endarterectomy. Ann Surg 198:705-712, 1983. Peitzman AB, Webster MW, Loubeau JM, Grundy BL, Bahnson HT: Carotid endarterectomy under regional (conductive) anesthesia. Ann Surg 196:59-64, 1982. Prough DS, Scuderi PE, Stullken E, Davis CH Jr: Myocardial infarction following regional anaesthesia for carotid endarterectomy. Can Anaesth Soc J 31:192-196, 1984. Rao TLK, Jacobs KH, El-Etr AA: Reinfarction following anesthesia in patients with myocardial infarction. Anesthesiology 59:499505, 1983. Reiz S, Balfors E, Sorensen MB, Ariola S Jr, Friedman A, Truedsson H: Isoflurane-a powerful coronary vasodilator in patients with coronary artery disease. Anesthesiology 59:9197, 1983. Reiz S: Nitrous oxide augments the systemic and coronary haemodynamic effects of isoflurane in patients with ischaemic heart disease. Acta Anaesthesiol Scand 27:464-469, 1983. Rich NM, Hobson RW II: Carotid endarterectomy under regional anesthesia. Am Surg 41:253-259, 1975. Riles TS, Kopelman I, Imparato AM: Myocardial infarction following carotid endarterectomy: A review of 683 operations. Surgery 85:249-252, 1979. Roizen MF, Hamilton WK, Sohn YJ: Treatment of stress-induced increases in pulmonary capillary wedge pressure using volatile anesthetics. Anesthesiology 55:446450, 1981. Shah KB, Kleinman BS, Sami H, Patel J, Rao TLK: Reevaluation of perioperative myocardial infarction in patients with prior myocardial infarction undergoing noncardiac operations. Anesth Analg 71:231-235, 1990. Smith JS, Roizen MF, Cahalan MK, Benefiel DJ, Beaupre PN, Sohn YJ, Byrd BF, Schiller NB, Stoney RJ, Ehrenfeld WK, Ellis JE, Aronson S: Does anesthetic technique make a difference? Augmentation of systolic blood pressure during carotid endarterectomy: Effects of phenylephrine versus light anesthesia and of isoflurane versus halothane on the incidence of myocardial ischemia.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/30/6/842/2751766 by East Carolina University user on 11 January 2019
3.
Asiddao BB, Donegan JH, Whitesell RC, Kalbfleisch JH: Factors associated with perioperative complications during carotid endarterectomy. Anesth Analg 61:631-637, 1982. Boysen G, Engell HC, Henriksen H: The effect of induced hypertension on internal carotid artery pressure and regional cerebral blood flow during temporary carotid clamping for endarterectomy. Neurology 22:1133-1144, 1972. Cucchiara RF, Sundt TM Jr, Michenfelder JD: Myocardial infarction in carotid endarterectomy patients anesthetized with halothane, enflurane, or isoflurane. Anesthesiology 69:783-784, 1988. Eger EI II, Saidman LJ, Brandstater B: Minimum alveolar anesthetic concentration, a standard of anesthetic potency. Anesthesiology 26:756-763, 1965. Ennix CL Jr, Lawrie GM, Morris GC Jr, Crawford ES, Howell JF, Reardon MJ, Weatherford SC: Improved results of carotid endarterectomy in patients with symptomatic coronary artery disease: An analysis of 1,546 consecutive carotid operations. Stroke 10:122125, 1979. Fiegl EO: The paradox of adrenergic vasoconstriction. Circulation 76:737-745, 1987. Goldman L: Cardiac risks and complications of noncardiac surgery. Ann Intern Med 98:504513, 1983. Hertzer NR, Lees CD: Fatal myocardial infarction following carotid endarterectomy: Three hundred thirty-five patients followed 611 years after operations. Ann Surg 194:212218, 1981. Imparato AM, Ramirez A, Riles T, Mintzer R: Cerebral protection in carotid surgery. Arch Surg 117:1073-1078, 1982. Kozmary SV, Lampe GH, Benefiel DJ, Cahalan MK, Strum D, Wauk LZ, Whitendale T, Eger EI II: Does addition of N2O to isoflurane increase myocardial ischemia in carotid artery surgery patients? Anesth Analg 68:S152, 1989 (abstr). Lee KS, Davis CH Jr, McWhorter JM: Low morbidity and mortality of carotid endarterectomy performed with regional anesthesia. J Neurosurg 69:483-487, 1988. McKay RD, Sundt TM, Michenfelder JD, Gronert GA, Messick JM, Sharbrough FW, Piepgras DG: Internal carotid artery stump pressure and cerebral blood flow during carotid endarterectomy: Modification by halothane, enflurane, and innovar. Anesthesiology 45:390-399, 1976. Messick JM Jr, Casement B, Sharbrough FW, Milde LN, Michenfelder JD, Sundt TM Jr: Correlation of regional cerebral blood flow (rCBF) with EEG changes during isoflurane anesthesia for carotid endarterectomy: Critical rCBF. Anesthesiology 66:344-349, 1987.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
1.
28.
30.
COMMENT Modica and coworkers present a carefully done prospective study of selective shunt placement in endarterectomy patients. In 171 endarterectomies, there were 35 instances of ischemia as detected by electroencephalographic (EEG) monitoring. In 12 of these, phenylephrine infusion was used to elevate blood pressure. There were no permanent deficits in this group. In the remaining 23 patients, there were two deficits, neither of which occurred intraoperatively. The criteria used for infusion were strict, and phenylephrine was not given unless the ischemic EEG changes persisted for 2 minutes. These authors counsel against the routine use of alphaadrenergic agents; their 0% fatal myocardial infarction rate strengthens their conclusions. The effect of these agents on cerebral blood flow during ischemia has not been adequately studied, as these authors also point out. This article also brings out the fact that postoperative myocardial infarction is usually silent and painless but carries a high mortality rate. The use of EEG monitoring to indicate the need for a shunt or other intervention is still under review. Theoretically, the scalp recordings recognize cortical events, while at least 20% of the infarcts associated with carotid artery occlusion occur in the depths. Recently, Kresowik et al. (1) reported that the EEG was abnormal in 15% of 458 patients who were monitored during endarterectomy. It failed to diagnose ischemia in 7 of 10 patients who awoke with new permanent deficits and in all 5 who awoke with temporary findings (1). We feel that recognition of the ischemia-prone patient and prophylactic shunting offers him or her the best opportunity to be intact after endarterectomy (2). Robert R. Smith Jackson, Mississippi
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/30/6/842/2751766 by East Carolina University user on 11 January 2019
29.
Anesthesiology 69:846-853, 1988. Steen PA, Tinker JH, Tarhan S: Myocardial reinfarction after anesthesia and surgery. JAMA 239:2566-2570, 1978. Sundt TM Jr, Sharbrough FW, Piepgras DG, Kearns TP, Messick JM Jr, O'Fallon WM: Correlation of cerebral blood flow and electroencephalographic changes during carotid endarterectomy. With results of surgery and hemodynamics of cerebral ischemia. Mayo Clin Proc 56:533-543, 1981. Tempelhoff R, Modica PA, Grubb RL Jr, Rich KM, Holtmann B: Selective shunting during carotid endarterectomy based on two-channel computerized electroencephalographic/compressed spectral array analysis. Neurosurgery 24:339-344, 1989. Zuccarello M, Yeh H, Tew JM: Morbidity and mortality of carotid endarterectomy under local anesthesia: A retrospective study. Neurosurgery 23:445-449, 1988.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
27.
Table 3. Results of 171 Carotid Endarterectomies versus Mayo Clinic Classification
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Table 2. Preoperative Neurological Characteristics in 171 Cases of Carotid Endarterectomy
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/30/6/842/2751766 by East Carolina University user on 11 January 2019
Table 1. Preoperative Medical Characteristics in 171 Cases of Carotid Endarterectomy
PATIENTS AND METHODS Patient characteristics One hundred fifty-nine patients undergoing 171 CEAs were studied prospectively. The study protocol was approved by the Institutional Review Board of the Washington University School of Medicine, and informed consent was obtained. The patient population consisted of 117 men and 54 women, with an average age of 64 ± 9 years (range, 36-82 years).
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/30/6/842/2751766 by East Carolina University user on 11 January 2019
AUTHOR(S): Modica, Paul A., M.D.; Tempelhoff, Rene, M.D.; Rich, Keith M., M.D.; Grubb, Robert L., Jr., M.D.
endarterectomy (CEA) (3,18,28). The extent to which anesthetic management may influence the incidence of MI after CEA remains unclear. Neurological assessment of the awake patient using local anesthesia provided by a cervical plexus block allows for sensitive detection of cerebral ischemia. In many studies, the use of local anesthesia for CEA has resulted in relatively low rates of postoperative stroke (0.66-2.5%) (9,11,17,22,30) and MI (02.3%) (11,17,18,23,30). Since CEA performed under local anesthesia requires light levels of sedation for optimal neurological monitoring, patient discomfort and anxiety can sometimes occur. Under these conditions, CEA may be poorly tolerated by both the patient and the surgeon, especially during long procedures or in those complicated by more superior carotid bifurcations requiring upper neck exposure. Furthermore, should the airway become compromised by either excessive sedation or neurological insult, proper anesthetic management becomes exceedingly more difficult. On the other hand, general anesthesia, although more acceptable to many patients and surgeons, increases the difficulty of monitoring cerebral function, and has been associated with more variable, and sometimes higher, rates of postoperative stroke (1-6.7%) (15,17,28,29) and MI (1,3,5,16,17). In one series of CEAs performed under general anesthesia, a 12.9% incidence of MI was reported for a high-risk subgroup of patients with symptomatic coronary artery disease (5). Independent of the choice of anesthesia, it is common practice in many institutions to use infusions of alpha-adrenergic agonists (i.e., phenylephrine) to induce hypertension pharmacologically during carotid cross-clamping, in an attempt to increase collateral cerebral blood flow and prevent cerebral ischemia. This practice may, however, increase the incidence of intraoperative myocardial ischemia during CEA when general anesthesia is employed (26). Based on this association between the administration of alpha-adrenergic agonists and myocardial dysfunction, we employed computerized electroencephalographic (CEEG) monitoring and selective shunting during general anesthesia for CEA to restrict the intraoperative use of phenylephrine infusions to specific instances of slowly or poorly reversible cerebral ischemia on the electroencephalogram (EEG). This study attempted to produce a low rate of perioperative MI while limiting the incidence of postoperative stroke by avoiding pharmacologically induced hypertension during CEA.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Neurosurgery 1992-98 June 1992, Volume 30, Number 6 842 Computerized Electroencephalographic Monitoring and Selective Shunting: Influence on Intraoperative Administration of Phenylephrine and Myocardial Infarction after General Anesthesia for Carotid Endarterectomy Experimental and Clinical Study
Intraoperative management: Technique of general anesthesia and criteria for phenylephrine infusion and placement of carotid bypass shunt With each patient receiving 100% oxygen (O2) by mask, general anesthesia was slowly induced with either thiopental (3-4 mg/kg, intravenously) or etomidate (Amidate, Abbott, North Chicago, IL) (0.31.35 mg/kg, intravenously). After facilitation of endotracheal intubation with vecuronium (Norcuron, Organon, West Orange, NJ) (0.1-0.2 mg/kg, intravenously), anesthesia was maintained in all patients with a mixture of 70% nitrous oxide (N2O) in O2, low-concentration (0.1-0.3% end-tidal) isoflurane (Forane, Anaquest, Madison, WI) and a variable-rate (1-2 µg/kg/h) infusion of fentanyl, each titrated so as to achieve sensitive EEG/CSA tracings while maintaining systolic blood pressure (SBP) within 20% of each patient's baseline ward value. Baseline ward SBP was defined as the mean of the values obtained by ward nurses during the first 24 hours after hospital admission. Hypertension (i.e., increase in SBP > 20% of baseline) was treated by increasing isoflurane (to 0.3% end-tidal) and/or fentanyl (to 2 µg/kgh), and administering vasodilators (e.g., nitroglycerin, intravenously) if needed. Conversely, hypotension (i.e., decrease in SBP > 20% of baseline) without signs of a concomitant ischemic EEG event was treated by decreasing isoflurane (to 0.1% endtidal) and/or fentanyl (to 1 µg/kg/h), and administering infusions of isotonic fluid (e.g., normal saline) and/or colloid (e.g., 5% albumin) if needed. Additional monitoring during anesthesia included an ECG, pulse oximetry, use of an esophageal stethoscope with temperature probe, mass spectrometry, and radial artery cannulation for continuous measurement of arterial blood pressure. Ventilation was mechanically controlled to maintain arterial pCO2 at 35-45 mm, as measured by intermittent blood gas analysis. In all 171 cases, heparin (5,000 units, intravenously) was administered 4 minutes before cross-clamping of the carotid arteries. An infusion of phenylephrine was administered only when the following criteria were met: 1. An ischemic EEG event associated with hypotension occurred and did not resolve within 2
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Electroencephalography Before the induction of anesthesia, a commercially available CEEG monitor (Interspec-Neurotrac, Conshohocken, PA) with compressed spectral array (CSA) capability was employed. This monitor displays two channels (left and right cerebral hemispheres) of raw EEG and CSA data recorded from five subdermal, fine-gauge, platinum needle electrodes (Grass, Quincy, MA). These electrodes were placed in a symmetrical frontal-mastoid montage that closely corresponds to F7T5 and F8-T6, in the international 10-20 System. Details of our method of CEEG monitoring, including electrode placement, have been described previously (14,29) . Throughout each procedure, a continuous on-line, two-channel, raw EEG tracing was obtained with a simultaneous printout of the CSA plus spectral edge frequency (SEF) using a frequency range of 1 to 30 Hz displayed in 2-second epochs. The SEF marker represents the highest significant frequency present in the sampled EEG spectrum. To achieve a sensitive EEG/CSA tracing, the depth of anesthesia was carefully controlled to produce a delta, theta, low alpha pattern with an SEF of 12 to 15 Hz recorded at a gain ranging between 20 to 80 µV. An ischemic EEG event (29) was defined as 1. a more than one-third amplitude attenuation of the raw EEG (by visual inspection of the printed tracing compared with both the other hemispheric
signal and earlier recordings) sustained for more than 30 seconds, and/or 2. a unilateral decrease in the SEF to less than 50% of its average value (over the preceding minute) sustained for more than 30 seconds. All detected ischemic EEG events were subclassified into the following three types (29): 1. Early ischemic EEG event: Any ischemic EEG event detected during the period of carotid crossclamping, when it was still technically possible to place a bypass shunt. 2. Late ischemic EEG event: Any ischemic EEG event detected late in the procedure, when it was no longer technically feasible to place a shunt (e.g., arteriotomy closure, shunt removal). 3. Ischemic EEG event associated with hypotension.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/30/6/842/2751766 by East Carolina University user on 11 January 2019
All patients received detailed medical and neurological evaluation before surgery. The preoperative medical findings are presented in Table 1. Previous history of MI was based upon the clinical history, an electrocardiogram (ECG), or the levels of the cardiac isoenzymes creatinine kinase (CK)-MB or lactic dehydrogenase (LDH). The patients' neurological history before surgery is presented in Table 2. Preoperative history of either a recent (≤6 weeks) or distant (>6 weeks) cerebral infarction was based on the neurological history, clinical examination, and computed tomographic findings for each patient. Cerebral angiography was performed in all patients before surgery. Of the 14 patients studied who had asymptomatic stenosis of the internal carotid artery, 12 had previously treated symptomatic lesions of the contralateral ICA. Grading of the degree of surgical risk was carried out according to the Mayo Clinic classification of Sundt et al. (28) as follows: Grade 1: neurologically stable patients with no major medical or angiographically defined risks and with unilateral or bilateral ulcerative-stenotic carotid disease; Grade 2: neurologically stable patients with no major medical risks but with significant angiographically determined risks; Grade 3: neurologically stable patients with major medical risks with or without significant angiographically determined risks; and Grade 4: major neurological risks, with or without associated major medical or angiographically determined risks.
RESULTS Table 3 summarizes the morbidity and mortality in this series of 171 CEAs according to the Mayo Clinic grading of risk (28). No fatal MIs (0%) occurred. There were two nonfatal postoperative MIs (1.2%), both in Grade 3 patients, each associated with 1 to 2 mm of ST segment depression on ECG and confirmed by significant increases in CK-MB, as Q waves were never evident on their serial ECGs. The first MI was accompanied by confusion, tachypnea, and mild pulmonary edema on the third postoperative day, whereas the second MI was associated with chest tightness and dyspnea on the first postoperative day. Neither of these 2 patients who suffered a postoperative MI received an intraoperative infusion of phenylephrine, although both suffered an early ischemic EEG event that resolved shortly after shunt placement.
DISCUSSION In this report of 171 CEAs performed with general anesthesia and CEEG-guided shunt placement, selective use of phenylephrine infusion for specific instances of slowly or poorly reversible cerebral ischemia, as shown on CEEG, was associated with only two perioperative MIs, both nonfatal, and a 1.2% incidence of postoperative cerebral infarction. Our 0% rate of fatal MI is identical to that found in previous reports of CEA performed using local anesthesia, including those that also commonly used induced hypertension with phenylephrine during carotid clamping (11,18,30). In contrast to our findings and those for local anesthesia described above, the combined use of general anesthesia and phenylephrine infusion during CEA leads to a greater incidence of intraoperative myocardial ischemia (26). This may explain the often higher rates of MI, especially fatal MI, associated with general anesthetic techniques for CEA (3,13,16,17). Previous studies have reported a decrease in ventricular wall stress when a decrease in blood pressure was associated with an increasing concentration of inhalant anesthetics (24). In a report of 60 CEAs analyzed with transesophageal echocardiography, the maintenance of systemic blood pressure with deep--1.4 minimum alveolar concentration (MAC) (4)--inhalation anesthesia (50% N20 in O2 accompanied by either halothane or isoflurane) and phenylephrine was compared with lighter inhalation anesthesia without phenylephrine with regard to new segmental wall motion and wall thickening abnormalities (26). Increased wall stress was associated with a threefold greater incidence of myocardial ischemia, as evidenced by new segmental
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Diagnosis of new postoperative myocardial infarctions and neurological deficits and statistical analysis Beginning in the recovery room upon emergence from anesthesia, all study patients were questioned about symptoms of chest pain and shortness of breath; observed for hypertension, hypotension, arrhythmias (including tachycardia), and altered mental status. In addition, they were examined for the presence of new postoperative motor, sensory, or speech disorders. Thereafter, these postoperative medical and neurological evaluations were performed two or more times daily until discharge for at least 4 and up to 7 days postoperatively. Serial 12-lead ECGs and CK-MB isoenzymes were obtained daily for 3 postoperative days for any of the above signs or symptoms suggestive of myocardial ischemia (7). The diagnosis of perioperative MI was made by an independent cardiologist and was based on a rise in blood levels of CK-MB isoenzyme greater than 12 IU/L. New neurological deficits persisting for 24 hours or more were classified as permanent deficits, that is, cerebral infarction.
In the 171 CEAs analyzed, there were 35 patients in whom ischemic EEG events occurred. Of these 35 patients, 12 were treated with an infusion of phenylephrine at some time intraoperatively, and none suffered a permanent deficit. In the 23 remaining patients in whom ischemic EEG events were detected, treatment with phenylephrine was not required at any time intraoperatively. In 1 of these 23 cases, a late EEG event accompanied by significant hypertension (i.e., SBP > 20% of baseline) occurred shortly after release of carotid cross-clamping. The patient who suffered this ischemic EEG event awoke with a deficit that was permanent and did not recover. He was classified as a Grade 4 patient, and died 3 days later from an autopsy-proven pulmonary embolism. Overall, two patients (1.2%) were left with permanent deficits postoperatively and three (1.8%) died, including the Grade 4 patient described above. The second permanent deficit and death occurred in a Grade 1 patient who initially was neurologically intact. After approximately 12 hours, however, he suffered a cerebral infarction due to acute carotid occlusion and died 4 days later. Finally, the third death occurred 6 days postoperatively as the result of an acute subdural hematoma in a Grade 4 patient treated with heparin for 48 hours for unstable angina. Serial ECGs and cardiac (CK-MB) isoenzymes were not significant for MI in this patient.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/30/6/842/2751766 by East Carolina University user on 11 January 2019
minutes after isoflurane and fentanyl were decreased and isotonic fluid and/or colloid treatment was instituted. 2. An early ischemic EEG event that did not resolve within 2 minutes after shunt placement occurred, accompanied by either hypotension (i.e., a decrease in SBP > 20% of the baseline that responded poorly to decreasing isoflurane and fentanyl and instituting treatment with fluid and/or colloid) or SBP no higher than 20% above the baseline. The decision to place a bypass shunt during the period of carotid cross-clamping was based on the presence or absence of an early ischemic EEG event on the two-channel EEG/CSA monitor. 3. A late ischemic EEG event that did not resolve within 2 minutes of cross-clamp removal occurred, accompanied by either hypotension (i.e., a decrease in SBP > 20% of baseline that responded poorly to decreasing isoflurane and fentanyl and instituting treatment with fluid and/or colloid) or SBP no higher than 20% above the baseline.
Received for publication, August 14, 1991; accepted, September 18, 1991. Presented in part at the 65th Annual Meeting of the International Anesthesia Research Society, San Antonio, Texas, February 9, 1991. Reprint requests: R. Tempelhoff, M.D., Department of Anesthesiology, Division of Neuroanesthesia, Washington University School of Medicine, Box 8054, 660 S. Euclid Avenue, St. Louis, MO 63110. REFERENCES: (1-30)
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
symptomatic (17-55%) and asymptomatic (49-66%) coronary artery disease in patients scheduled for CEA (5,8,16) . Some recent investigations implicate isoflurane and/or N2O, the inhalation anesthetics used in our study, as causes of myocardial ischemia in patients with coronary artery disease (20,21), leading us to question whether their use during CEA is appropriate. Yet, various investigations, including this study, do not document an increase in adverse myocardial effects during CEA employing N2O and isoflurane (3, 26) . In a review of 2,223 CEAs, a fatal MI rate of 0.25% was found in the 784 CEA patients receiving a 1.0-MAC mixture of isoflurane in 50% N2O and O2, a significantly lower incidence as compared with the 1,439 other CEAs performed with different anesthetic agents (3). Using transesophageal echocardiography during CEA, Smith et al. (26) found no difference in the incidence of myocardial ischemia between patients anesthetized with isoflurane and those receiving halothane, each at similar MAC levels with 50% N2O and O2. Also, the addition of N2O to isoflurane did not increase the incidence of myocardial ischemia, as measured by transesophageal echocardiography and 12-lead ECG, in patients undergoing carotid artery surgery (10). In view of these reports, as well as our finding of only two MIs in 171 CEAs, both of which were nonfatal, it appears that whatever effect N2O and isoflurane might have in patients with coronary artery disease, it is not clinically evident in patients undergoing CEA. During both normocapnic and hypocapnic states under general anesthesia for CEA, pharmacologically induced hypertension to a level approximately 20% higher than preoperative SBP produced regional cerebral blood flow and carotid artery back (stump) pressure responses during carotid clamping that could not be predicted in individual cases (2). These findings indicate that in patients with cerebrovascular disease, pressure cannot be equated with flow (12). Thus, the routine use of induced hypertension without concomitant EEG signs of cerebral ischemia cannot be advocated, especially in view of its potentially hazardous consequences of myocardial ischemia during general anesthesia in this population. Based on these previous findings and on the results of our study, in order to decrease the incidence of both MI and cerebral infarction after general anesthesia for CEA with EEG monitoring, we recommend the restrictive use of phenylephrine-induced hypertension for specific instances of slowly or poorly reversible cerebral ischemia as detected by EEG.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/30/6/842/2751766 by East Carolina University user on 11 January 2019
wall motion and wall thickening abnormalities, and was found in the patients who received phenylephrine (26) . An additional factor that may also contribute to the development of myocardial ischemia in patients treated with phenylephrine during CEA is the effect of alpha-adrenergic coronary vasoconstriction (6). In view of these findings and our results, the restrictive use of a phenylephrine infusion for specific instances of slowly or poorly reversible cerebral ischemia as detected by EEG appears to decrease the occurrence of perioperative MI during general anesthesia for CEA. It is possible that the low rate of perioperative MI in this study may have been the result of a lower severity of preoperative cardiovascular and neurological characteristics in our patient population; however, the mean age of our patients, 64 years, is close to the range (62-64 years) of other reports (1,5,17, 18) . Also, only 8.2% of the CEAs in this series were performed for asymptomatic carotid stenosis, a finding similar to or less than that of previous reports (1,11,17,18) , and most were performed as a second procedure in patients with previously treated symptomatic lesions of the contralateral internal carotid artery. The incidence of preoperative hypertension (71.9%) in our study is also close to or higher than that of earlier reports (1,5,11,18,23). A previous MI was reported in 20.5% of our patients, a finding similar to the range (7-24%) reported in other, comparable series (1,5,11,18). In addition, 42.7% of our CEAs were associated with a history of angina, a percentage almost identical to that recorded in earlier studies (41-44%) (16,18). Finally, over three fourths (76%) of our CEAs were performed in patients in Mayo Clinic Grades 3 and 4, a much higher incidence of patients with major medical and/or neurological risks as compared with the CEA patients reported on by Sundt et al. (28). Overall, when compared with patients in other reports, the patients in our CEA series appear to have been at the same or higher risk for postoperative complications. In this study, one of the two MIs presented without pain, a finding consistent with those of earlier reports, in which approximately 50% of the MIs following surgery were also painless (19,25,27). There is still the possibility that undiagnosed silent MIs may have occurred in some of the remaining 169 patients we studied. Based on the recommendation of Goldman (7) , we were alert for signs and symptoms other than chest pain suggestive of myocardial ischemia and obtained serial ECGs and CK-MB isoenzymes for 3 postoperative days whenever the possibility of an MI was felt to be present. Also, our study patients were monitored for 4 to 7 days postoperatively, which is consistent with the findings of the most recent prospective studies showing that perioperative MIs always occur within 4 postoperative days after noncardiac surgery (19,25). The high mortality associated with postoperative MI, ranging from 23% (16,25) to 69% (27), also makes it unlikely that many painless MIs could have occurred in our series without there being a considerable increase in the number of postoperative deaths. It is well known that there is a high incidence of
2.
4.
5.
6. 7. 8.
9. 10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22. 23.
24.
25.
26.
Modica PA, Tempelhoff R: A comparison of computerized EEG with internal carotid artery stump pressure for detection of ischemia during carotid endarterectomy. J Neurosurg Anesth 1:211-218, 1989. Morawetz RB, Zeiger HE, McDowell HA Jr, McKay RD, Warner PD, Gelman S, Halsey JH: Correlation of cerebral blood flow and EEG during carotid occlusion for endarterectomy (without shunting) and neurologic outcome. Surgery 96:184-189, 1984. O'Donnell TF Jr, Callow AD, Willet C, Payne D, Cleveland RJ: The impact of coronary artery disease on carotid endarterectomy. Ann Surg 198:705-712, 1983. Peitzman AB, Webster MW, Loubeau JM, Grundy BL, Bahnson HT: Carotid endarterectomy under regional (conductive) anesthesia. Ann Surg 196:59-64, 1982. Prough DS, Scuderi PE, Stullken E, Davis CH Jr: Myocardial infarction following regional anaesthesia for carotid endarterectomy. Can Anaesth Soc J 31:192-196, 1984. Rao TLK, Jacobs KH, El-Etr AA: Reinfarction following anesthesia in patients with myocardial infarction. Anesthesiology 59:499505, 1983. Reiz S, Balfors E, Sorensen MB, Ariola S Jr, Friedman A, Truedsson H: Isoflurane-a powerful coronary vasodilator in patients with coronary artery disease. Anesthesiology 59:9197, 1983. Reiz S: Nitrous oxide augments the systemic and coronary haemodynamic effects of isoflurane in patients with ischaemic heart disease. Acta Anaesthesiol Scand 27:464-469, 1983. Rich NM, Hobson RW II: Carotid endarterectomy under regional anesthesia. Am Surg 41:253-259, 1975. Riles TS, Kopelman I, Imparato AM: Myocardial infarction following carotid endarterectomy: A review of 683 operations. Surgery 85:249-252, 1979. Roizen MF, Hamilton WK, Sohn YJ: Treatment of stress-induced increases in pulmonary capillary wedge pressure using volatile anesthetics. Anesthesiology 55:446450, 1981. Shah KB, Kleinman BS, Sami H, Patel J, Rao TLK: Reevaluation of perioperative myocardial infarction in patients with prior myocardial infarction undergoing noncardiac operations. Anesth Analg 71:231-235, 1990. Smith JS, Roizen MF, Cahalan MK, Benefiel DJ, Beaupre PN, Sohn YJ, Byrd BF, Schiller NB, Stoney RJ, Ehrenfeld WK, Ellis JE, Aronson S: Does anesthetic technique make a difference? Augmentation of systolic blood pressure during carotid endarterectomy: Effects of phenylephrine versus light anesthesia and of isoflurane versus halothane on the incidence of myocardial ischemia.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/30/6/842/2751766 by East Carolina University user on 11 January 2019
3.
Asiddao BB, Donegan JH, Whitesell RC, Kalbfleisch JH: Factors associated with perioperative complications during carotid endarterectomy. Anesth Analg 61:631-637, 1982. Boysen G, Engell HC, Henriksen H: The effect of induced hypertension on internal carotid artery pressure and regional cerebral blood flow during temporary carotid clamping for endarterectomy. Neurology 22:1133-1144, 1972. Cucchiara RF, Sundt TM Jr, Michenfelder JD: Myocardial infarction in carotid endarterectomy patients anesthetized with halothane, enflurane, or isoflurane. Anesthesiology 69:783-784, 1988. Eger EI II, Saidman LJ, Brandstater B: Minimum alveolar anesthetic concentration, a standard of anesthetic potency. Anesthesiology 26:756-763, 1965. Ennix CL Jr, Lawrie GM, Morris GC Jr, Crawford ES, Howell JF, Reardon MJ, Weatherford SC: Improved results of carotid endarterectomy in patients with symptomatic coronary artery disease: An analysis of 1,546 consecutive carotid operations. Stroke 10:122125, 1979. Fiegl EO: The paradox of adrenergic vasoconstriction. Circulation 76:737-745, 1987. Goldman L: Cardiac risks and complications of noncardiac surgery. Ann Intern Med 98:504513, 1983. Hertzer NR, Lees CD: Fatal myocardial infarction following carotid endarterectomy: Three hundred thirty-five patients followed 611 years after operations. Ann Surg 194:212218, 1981. Imparato AM, Ramirez A, Riles T, Mintzer R: Cerebral protection in carotid surgery. Arch Surg 117:1073-1078, 1982. Kozmary SV, Lampe GH, Benefiel DJ, Cahalan MK, Strum D, Wauk LZ, Whitendale T, Eger EI II: Does addition of N2O to isoflurane increase myocardial ischemia in carotid artery surgery patients? Anesth Analg 68:S152, 1989 (abstr). Lee KS, Davis CH Jr, McWhorter JM: Low morbidity and mortality of carotid endarterectomy performed with regional anesthesia. J Neurosurg 69:483-487, 1988. McKay RD, Sundt TM, Michenfelder JD, Gronert GA, Messick JM, Sharbrough FW, Piepgras DG: Internal carotid artery stump pressure and cerebral blood flow during carotid endarterectomy: Modification by halothane, enflurane, and innovar. Anesthesiology 45:390-399, 1976. Messick JM Jr, Casement B, Sharbrough FW, Milde LN, Michenfelder JD, Sundt TM Jr: Correlation of regional cerebral blood flow (rCBF) with EEG changes during isoflurane anesthesia for carotid endarterectomy: Critical rCBF. Anesthesiology 66:344-349, 1987.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
1.
28.
30.
COMMENT Modica and coworkers present a carefully done prospective study of selective shunt placement in endarterectomy patients. In 171 endarterectomies, there were 35 instances of ischemia as detected by electroencephalographic (EEG) monitoring. In 12 of these, phenylephrine infusion was used to elevate blood pressure. There were no permanent deficits in this group. In the remaining 23 patients, there were two deficits, neither of which occurred intraoperatively. The criteria used for infusion were strict, and phenylephrine was not given unless the ischemic EEG changes persisted for 2 minutes. These authors counsel against the routine use of alphaadrenergic agents; their 0% fatal myocardial infarction rate strengthens their conclusions. The effect of these agents on cerebral blood flow during ischemia has not been adequately studied, as these authors also point out. This article also brings out the fact that postoperative myocardial infarction is usually silent and painless but carries a high mortality rate. The use of EEG monitoring to indicate the need for a shunt or other intervention is still under review. Theoretically, the scalp recordings recognize cortical events, while at least 20% of the infarcts associated with carotid artery occlusion occur in the depths. Recently, Kresowik et al. (1) reported that the EEG was abnormal in 15% of 458 patients who were monitored during endarterectomy. It failed to diagnose ischemia in 7 of 10 patients who awoke with new permanent deficits and in all 5 who awoke with temporary findings (1). We feel that recognition of the ischemia-prone patient and prophylactic shunting offers him or her the best opportunity to be intact after endarterectomy (2). Robert R. Smith Jackson, Mississippi
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/30/6/842/2751766 by East Carolina University user on 11 January 2019
29.
Anesthesiology 69:846-853, 1988. Steen PA, Tinker JH, Tarhan S: Myocardial reinfarction after anesthesia and surgery. JAMA 239:2566-2570, 1978. Sundt TM Jr, Sharbrough FW, Piepgras DG, Kearns TP, Messick JM Jr, O'Fallon WM: Correlation of cerebral blood flow and electroencephalographic changes during carotid endarterectomy. With results of surgery and hemodynamics of cerebral ischemia. Mayo Clin Proc 56:533-543, 1981. Tempelhoff R, Modica PA, Grubb RL Jr, Rich KM, Holtmann B: Selective shunting during carotid endarterectomy based on two-channel computerized electroencephalographic/compressed spectral array analysis. Neurosurgery 24:339-344, 1989. Zuccarello M, Yeh H, Tew JM: Morbidity and mortality of carotid endarterectomy under local anesthesia: A retrospective study. Neurosurgery 23:445-449, 1988.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
27.
Table 3. Results of 171 Carotid Endarterectomies versus Mayo Clinic Classification
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Table 2. Preoperative Neurological Characteristics in 171 Cases of Carotid Endarterectomy
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/30/6/842/2751766 by East Carolina University user on 11 January 2019
Table 1. Preoperative Medical Characteristics in 171 Cases of Carotid Endarterectomy
