Circulation 60,475 • Volume 14, No. 3 • Fall 1999

Regional Anesthesia Preferable for Carotid Surgery

David A. Zvara, M.D.

In My Opinion

For most patients, regional anesthesia is a superior technique to general anesthesia for carotid endarterectomy (CEA). This bold statement may be supported by some, and bitterly refuted by others. There are proponents who supported both sides of this debate, and there are data to support either position.

Why then, would one make such a provocative statement about a technique?

In 1987, Dr. John Youngberg forwarded the argument that regional anesthesia is superior to general anesthesia for CEA.1,2 The potential benefit of regional anesthesia for CEA stressed the theoretical advantage of monitoring an awake patient for new neurologic deficits compared to the inaccuracies of cerebral monitoring in an anesthetized patient. In an awake patient, intraarterial shunts are placed when indicated by patient assessment. Therefore, shunt use (and the associated risks of dissection, thrombus formation, or atheromatous embolism and resulting neurologic deficits) could be reduced in patients with regional anesthesia. Cardiac morbidity may be reduced with avoidance of general anesthesia. Finally, regional anesthesia would allow quicker postoperative recovery, and shorter intensive care unit and hospital lengths of stay.

Analysis of recent data shows that these theoretical advantages of regional anesthesia may be supported by outcome data.3 Data from 13 manuscripts which compare outcome in a total of 7,619 patients having either regional or general anesthesia for CEA were reviewed. Primary outcome was defined as the incidence of postoperative transient ischemic attack (TIA), stroke, myocardial infarction (MI), and death (Table 1); secondary outcomes included the incidence of shunt placement, hospital and intensive care unit lengths of stay, hospital charges, and postoperative hemodynamic stability and use of vasoactive agents. Two of these studies report data from prospective trials,4,5 while the remainder are retrospective reviews. Three of these studies6-8 show a significant improvement in primary outcome with regional anesthesia and deserve special comment.

Table 1. Primary outcome from the cumulative results of 13 studies of carotid endarterectomy

Regional(n) General (n) P-value
Totals 5362 2257
Stroke/TIA 60(1.1%) 89(3.9%) <0.0001
MI 68(1.3%) 37(1.6%) NS
Death 60(1.1%) 32(1.4%) NS

Allen et al6 reports on 679 consecutive carotid endarterectomies in 584 patients. MI occurred in nine patients in the general anesthesia group (2.5%) and in two patients in the regional anesthesia group (0.6%), p=0.07. Cardiopulmonary complications occurred in 30 operations (8.3%) performed under general anesthesia, and in only 13 (4.1%) of operations with cervical block, p=0.03. Becquemin et al7 retrospectively studied a series of 385 carotid endarterectomies and found 10 MIs documented by ECG or cardiac enzyme changes in the general anesthesia group, and no MIs in the regional anesthesia group, p<0.05. Five of these 10 MIs were fatal. They concluded that regional anesthesia by cervical plexus block was more appropriate than general anesthesia in patients with coronary artery disease. Corson et al8 retrospectively compared 368 patients having 399 CEAs under either general or cervical block anesthesia and found a significantly (p<0.025) higher stroke rate in the general anesthesia group.

The paper by Rockman et al9 reports on the largest number of patients. This study is a retrospective review of a 32-year period in which 3975 carotid operations were performed. There was no difference in primary outcome between general and regional anesthesia groups. One might argue, however, that over these 32 years, patient selection, surgical practice and anesthetic care have changed and, therefore, these results may not be representative of current outcome. The authors recognized this limitation and performed a secondary evaluation of outcome by year of operation. When evaluating the most recent data, e.g., from 1985 to 1994, Rockman reports on 1763 operations (1414 regional and 349 general). The incidence of perioperative MI (1.2% general group vs. 0.6% regional group) and perioperative mortality (0.9% general group vs. 0.9% regional group) are similar in the two groups. The perioperative stroke rate, however, was significantly different between groups (3.2% general group vs. 1.2% regional group; p<0.01). Analysis of this subgroup revealed that the use of general anesthesia (p<0.02) and a history of preoperative stroke (p<0.01) were found to be significantly associated with perioperative stroke. If one combines this accumulated experience, there are data on 7,619 CEA operations. Analysis of the combined data shows no difference in MI and perioperative mortality. There is a threefold difference, however, in the incidence of stroke and TIAs between groups, 3.9% general group vs. 1.1% regional group, p<0.001. There were also significant advantages in the regional anesthesia group for secondary outcomes.3

Proponents of general anesthesia argue that regional anesthesia is associated with a lack of airway control, conversion to general anesthesia is often required for a failed block, and that cervical block is associated with potentially dangerous hemidiaphragmatic paralysis. Although these concerns are real, data suggest they are not common. The conversion from regional to general anesthesia is low. Shah et al10 report that seven of 654 CEAs under regional anesthesia were converted to general (1.1%). Four of these were for inadequate anesthesia, and three were for airway control. Corson et al8 report that five of 157 patients (3.2%) converted to general anesthesia secondary to inadequate anesthesia. Castresana et al11 studied diaphragmatic motion in 28 patients having deep cervical plexus block for CEA and report abnormalities in hemidiaphragmatic motion in 61% of patients by fluoroscopy. These abnormalities were associated with a statistically significant rise in PaCO2 from 36+ 1 mmHg preblock to 40+ 1 mmHg postblock. Although this is a difference, the clinical physiologic consequence of these changes are insignificant. Monitoring of cerebral function under general anesthesia is also fraught with inconsistency. In a recent report by Stoughton et al12 208 consecutive carotid endarterectomies were prospectively evaluated for cerebral function during surgery under regional anesthesia with simultaneous mental status evaluation and intraoperative electroencephalographic (EEG) monitoring. The authors found a high incidence of false positive (6.7%) and false negative (4.5%) EEG results in the detection of neurologic deficits when compared to an awake mental status evaluation. If the correlation is so bad in an awake patient, one must wonder what the utility is in an anesthetized patient. Assessment of neurologic function in the awake individual avoids all ambiguity in other surrogate testing modalities.

Clearly, there are instances when general anesthesia is the preferred technique. Obviously, in uncooperative patients, claustrophobic patients, patients with tremor or cough, and some others, general anesthesia may be required. In addition, institutional data should be a primary consideration when evaluating a change in anesthetic technique. For example, many anesthesiologists and surgeons have a long institutional record of successful carotid operations under general anesthesia with or without shunt. If results are good, then a change in technique may not improve outcome.

As a general statement, regional anesthesia is associated with a reduced incidence of perioperative neurologic deficit. Additionally, regional anesthesia results in reduced shunt placement, reduced hospital utilization, reduced hospital charges and improved postoperative hemodynamics. The technique is safe, and there are few conversions to general anesthesia for failed block or airway control.

Finally, regional anesthesia allows for the best intraoperative monitoring of neurologic status—an awake patient. Therefore, in my opinion, for most patients, regional anesthesia is superior to general anesthesia for patients having CEA.

Dr. Zvara is from the Department of Anesthesiology, Wake Forest University School of Medicine, WinstonSalem, NC.


  1. Youngberg JA: Regional Anesthesia is preferable to general anesthesia for carotid artery surgery. Pro: Regional anesthesia is preferable to general anesthesia for carotid artery surgery. Wynands JE (ed.) J Cardiothorac Anesth 1987;1:47982.
  2. Trop D: Con: Carotid endarterectomy: General is safer than regional anesthesia. J Cardiothorac Anesth 1987;1:4838.
  3. Zvara DA: Pro: Regional anesthesia is the best technique for carotid endarterectomy. J Cardiothorac Vasc Anesth 1998;12(1):1114.
  4. McCleary AJ, Dearden NM, Dickson DH, et al: The differing effects of regional and general anaesthesia on cerebral metabolism during carotid endarterectomy. Eur J Vasc Endovasc Surg 1996;12:17381.
  5. Forssell C, Takolander R, Bergqvist D, et al: Local versus general anaesthesia in carotid surgery. A prospective, randomised study. Eur J Vasc Surg 1989;3:5039.
  6. Allen BT, Anderson CB, Rubin BG, et al: The influence of anesthetic technique on perioperative complications after carotid endarterectomy. J Vasc Surg 1994;19:83442.
  7. Becquemin JP, Paris E, Valverde A, et al: Carotid surgery. Is regional anesthesia always appropriate? J Cardiovasc Surg 1991;32:5928.
  8. Corson JD, Chang BB, Shah DM et al: The influence of anesthetic choice on carotid endarterectomy outcome. Arch Surg 1987;122:80712.
  9. Rockman CB, Riles TS, Gold M, et al: A comparison of regional and general anesthesia in patients undergoing carotid endarterectomy. J Vasc Surg 1996;24:94656.
  10. Shah DM, Darling III RC, Chang BB, et al: Carotid endarterectomy in awake patients: Its safety, acceptability, and outcome. J Vasc Surg 1994;19:10151019.
  11. Castresana MR, Masters RD, Castresana EJ, et al: Incidence and clinical significance of hemidiaphragmatic paresis in patients undergoing carotid endarterectomy during cervical plexus block anesthesia. J Neurosurg Anesthesiol 1994;6:213.
  12. Stoughton J, Nath RL, Abbott WM: Comparison of simultaneous electroencephalographic and mental status monitoring during carotid endarterectomy with regional anesthesia. J Vasc Surg 1998;28;101423.