Carotid Endarterectomy: Comparison Between General and Local Anesthesia. Revision of Our Experience With 428 Consecutive Cases
Article Outline
Background
This sequential retrospective monocentric study compares the results between general and local anesthesia for carotid endarterectomy (CEA).
Methods
Between November 2002 and October 2004, 428 CEAs were performed in our vascular unit. Two groups were formed: group GA (general anesthesia): 219 patients operated under general anesthesia; group LA (local anesthesia): 209 patients operated under local anesthesia.
Results
No mortality was found in both groups. After surgery, three strokes were detected in group GA and three in group LA (GA 1.36% vs. LA 1.43%, p = .9540); After CEA, there were three TIAs in GA group and none in LA group (GA 0.42% vs. LA 0%, p = .2634).
Conclusion
The morbi-mortality was not influenced by the type of anesthesia used for carotid surgery. No statistical difference was detected in the perioperative neurological and cardiopulmonary complication rates between GA and LA.
Introduction
This sequential retrospective study compares the results between general and local anesthesia for carotid endarterectomy (CEA), in a consecutive series of groups of patients treated in a single centre.
Methods
Between November 1, 2002 and October 31, 2004, 461 treatments of internal carotid artery (ICA) stenosis were performed in our vascular unit. In November 2003, after the surgical team leader was changed in our Center, the anesthesia to perform CEA was changed; local anesthesia substituted general anesthesia. All patients underwent preoperative diagnostic tests and duplex-scanning was used to evaluate stenosis and carotid plaques. The degree of stenosis was determined by measurements obtained by the European Carotid Surgery Trial method (patients eligible for surgery: grade stenosis of the ICA ≥70% for symptomatic patients and ≥80% for asymptomatic patients). In symptomatic patients, a computed tomography scan or a magnetic resonance scan was performed preoperatively to study cerebral tissue or possible focal lesions associated. A pre- and postoperative neurological consultation was performed on these patients. Symptomatic patients presenting with a 2-3-cm cerebral lesion on computed tomography/magnetic resonance scan and/or a residual neurological deficit underwent CEA under general anesthesia (GA), and the shunt insertion was performed routinary to ensure the best cerebral perfusion. Of the 461 ICA stenosis treated, 21 underwent combined surgery CEA and coronary artery bypass graft (CABG), 12 carotid artery stenting, and the remaining 428 cases underwent CEA alone. Retrospectively, we formed two groups: group GA in which patients were operated under general anesthesia (between November 2002 and October 2003); group LA (local anesthesia) in which patients were operated under local anesthesia (from November 2003 to October 2004). Both groups were homogeneous in terms of age, gender, comorbidities, common risk factors, and specific symptoms (Table I). In group GA, we examined 219 CEAs, in which all the operations were performed under general anesthesia (induction with propophol and fentanyl, maintenance with sevorane; induced hypertension with neosynephrine or ephedrine; induced hypotension with urapidil or nicardipine); the cerebral status was monitored by checking the stump pressure, and in case of stump pressure ≤50 mm Hg, a Pruitt-Inahara shunt was positioned. In group LA, 209 CEA operations were examined: all interventions were conducted under local anesthesia (20 mL carbocaine infiltration, 1% of which without adrenalin, with further 10 mL carbocaine infiltration during surgery; induced hypertension with neosynephrine or ephedrine; hypotension induced with urapidil or nicardipine); the cerebral status monitoring was performed with the evaluation of clinical signs of cerebral ischemia. In case of clamping intolerance, a Pruitt-Inahara shunt was positioned. In LA group, 15 patients needed intraoperative conversion to GA. The reasons were different: 14 cases presented with neurological deterioration on cross-clamping and for protracted or difficult surgery, one case with general discomfort, anxiety, uncontrollable cough, and restlessness. In both groups, an intraoperative duplex scan before the closure of the surgical incision was performed to check the perviousness of carotid vessels. In both groups, blood pressure was recorded directly using a radial artery catheter; the blood pressure was manipulated up in 45% of GA group and in 14% of the LA, it was manipulated down in 12% of GA group and in 33% of the LA, in both groups it was manipulated up and down in 7% of the cases; in the others it was not manipulated. In all cases, heparin was administered before carotid clamping (70 IU/kg) and was not routinely reversed. Statistical analyses were performed using Primer program, and chi-square test with Yates' correction was used to compare groups. The level of significance was taken at p < .05.
Table I. General patient characteristics
| Group GA | Group LA | p | |
|---|---|---|---|
| Median age | 71.2 | 72.1 | n.s. |
| Gender | Male 157 Female 62 | Male 153 Female 56 | n.s. |
| Coronary artery disease | 70 | 55 | .2388 |
| Peripheral artery disease | 61 | 59 | .9311 |
| Hypertension | 183 | 169 | .5457 |
| Hyperlipidemia | 111 | 95 | .3242 |
| Diabetes mellitus | 28 | 33 | .4530 |
| COPD | 74 | 81 | .3331 |
| Smokers | 107 | 96 | .6107 |
| Asymptomatic | 140 | 146 | .2303 |
Results
In both groups, there was no mortality. After surgery, three strokes were detected in group GA and three in group LA (GA 1.36% vs. LA 1.43%, p = .9540), with no statistical difference. After CEA, we registered three TIAs in GA group and zero in LA group (GA 1.36% vs. LA 0%, p = .2634) without any statistical difference. We registered 27 peripheral neurological deficits (six permanent) in GA group and 26 in LA group (three permanent), but no statistical differences were detected (Table II). Hematomas requiring surgical drainage were more common in LA group (GA 1.82% vs. LA 4.78%, p = .1476), whereas cardiac (GA 1.82% vs. LA 0.47%, p = .3968) and pulmonary (GA 2.28% vs. LA 0.95%, p = .4839) complications were more numerous in GA group, but without any statistically significant difference in both groups. The shunt use was predominant in LA group (GA 5.1% vs. LA 10.1%, p = .0731), but it was statistically irrelevant. When complications in LA arose, we considered the necessity of intraoperative conversion to GA; this sub-group was composed of 15 patients. Comparing the LA not converted with LA converted group, we did not detect statistical difference in terms of neurological deficit (LA not converted 0% vs. LA converted 1.54%, p = .6276). In 11.45% of cases, patients presented with contralateral high-grade stenosis (>70%, 17 cases in GA group and 32 cases in LA group); in 3.9% of cases they presented with a contralateral carotid occlusion (12 cases in GA group and 15 cases in LA group); no patient presented with transient ischemic attack (TIA) or stroke after CEA. We considered the group of patients with contralateral carotid occlusion associated with the group with contralateral high-grade carotid stenosis (76 cases) and compared it with all the others (352 cases) in terms of shunt use; no statistical difference was observed in this case also (11.8 vs. 6.5%, p = .1754). At intraoperative duplex scan control, 11 cases presented with a surgical defect: three cases presented with residual flaps in ICA, six with an occlusion of external carotid artery secondary to a residual flap, two with a turbulent flow into ICA. In all cases, an immediate surgical correction of the defect was performed. Although we corrected 11 surgical defects by duplex scan intraoperative control, there were immediate thrombosis post-CEA in two cases, all of them in LA group, probably as a consequence of surgical mistake. The mean follow-up of 393 patients was 42.2 months (range, 6-79 months), 35 patients were lost to follow-up. At 1-year follow-up, no carotid occlusions were found, no new TIA/stroke/death at 1 year was detected, and the cumulative TIA/stroke/death rate after surgery was 2.29% (9/393); the cumulative TIA/death rate after 1 year was 0% (0/393); the stroke rate in GA group after 1 year was 2.99% (6/201) and in LA group was 1.56% (3/192) without statistical difference (p = .5038). A total of 23 (5.8%) of 393 patients required a new treatment for carotid restenosis; 22 of them agreed and one refused. Of these 22 patients, 19 underwent carotid artery stenting, one underwent bypass, and two underwent CEA; all of them did not present morbi-mortality after second carotid treatment. Statistically significant differences were found on surgical technique in two groups (Table III). Generally, surgical technique was chosen on the basis of vessel anatomy, plaque characteristics, or preference of the surgeon. In presence of kinking or small vessels, the evertion technique was always performed, the patch was used in case of the presence of very small vessels; the bypass was performed when it was not possible to find the end plaque; in all other cases the surgeon chose between conventional or evertion technique. At follow-up, we found more restenosis in CEAs treated by standard technique (7.9%) compared with other techniques (3.1% with evertion, 0.5% with patch, and 0% with bypass).
Table II. Results in group GA and LA
| Group GA (219) | Group LA (209) | p | |
|---|---|---|---|
| Mortality | 0 | 0 | n.s. |
| Major/minor stroke | 3 | 3 | .9540 |
| TIA | 3 | 0 | .2634 |
| Total deficit peripheral nervous | 27 | 26 | .9721 |
| Permanent deficit peripheral nervous | 6 | 3 | .5464 |
| Hematomas drainage | 4 | 10 | .1476 |
| Heart complications | 4 | 1 | .3968 |
| Pulmonary complications | 5 | 2 | .4839 |
| Shunt use | 11 | 21 | .0731 |
Table III. Surgical technique in group GA and LA
| Group GA (219) | Group LA (209) | p | |
|---|---|---|---|
| Standard | 52 | 136 | <.0001 |
| Evertion | 133 | 60 | <.0001 |
| Patch | 29 | 11 | .0076 |
| Bypass | 5 | 2 | .4839 |
Discussion
The choice between LA and GA for CEA is still a matter of debate in literature.1, 2, 3, 4, 5, 6 During CEA, LA offers a better perioperative hemodynamic stability and a direct evaluation of the patient's neurological status; in contrast, GA is indicated for noncooperating patients, it is more comfortable for the surgeon, and allows increase of cerebral flow and perfusion (by modulating CO2 and arterial pressure). For this reason, there is an open debate in literature to establish which type of anesthesia presents better results in CEA surgery. Rerkasem et al.7 assert that nonrandomized studies suggest potential benefits from the use of local anesthetic. Lutz et al.1 suggest that LA is safer for CEA; Love and Hollyoak8 assert that with LA there is a reduction in mortality and major stroke events, Watts et al.9 in a retrospective study mentioned that there was no difference between LA and GA with regard to neurologic complications among 582 patients, and Forsell et al.10 confirmed that LA is comparable with GA regarding perioperative complications. In our study, there were no statistical differences between stroke or morbi-mortality in both groups (Table II) and never in sub-group of LA converted. In a study by Mofidi et al.,4 LA appears to offer a reduction in intraoperative shunting and perioperative stroke, the duration of hospital stay, clinical and possible cost advantages, as compared with GA; Gabelman et al.11 confirmed the benefits of performing CEA in conscious patients, resulting in less operative time and in less intensive care unit stay compared with those operated on under GA. In the studies of Watts et al.,9 Santamaria et al.,5 and Kalko et al.,2 the shunt use was significantly more common in GA group, whereas in Lutz et al.1 study there was no difference in shunt use. In our report the shunt use was selective in both groups without any statistical difference (Table II). In a study by Sbarigia et al.,12 the different hemodynamic effect of the two anesthetic techniques was confirmed: patients who received LA had a rate of myocardial ischemia that was half to that of patients who had GA. They asserted that LA can be used safely, even in high-risk patients undergoing CEA. On the contrary, in a study by Kasprzak et al.,3 no significant differences could be observed in the perioperative neurological and cardiopulmonary complication rates between GA and LA in patients undergoing CEA. Recently, the GALA Trial13 study has not shown a definite difference in outcomes between GA and LA for carotid surgery with regard to major perioperative complications of stroke, myocardial infarction, and death; this is apparently in contrast with our series in which the incidence of cardiac complication appears to be 4 times higher in GA and the incidence of lung complication is 2.5 times greater in the GA, but at statistical analysis we observe that there are no differences between both groups (Table II). In our series, the variation of two groups consisted in surgical technique with statistical difference: in GA group 23.7% standard, 60.7% evertion, 13.3% patch, and 2.3% bypass; in LA group 65.1% standard, 28.7% evertion, 5.3% patch, and 0.9% bypass. The short-term results show that there are no statistical differences in terms of patency between the various surgical techniques, whereas the long-term results show that the standard technique is more susceptible to restenosis requiring re-treatment.
Conclusion
The morbi-mortality was not influenced by the type of anesthesia used. No statistical difference was detected in the perioperative neurological and cardiopulmonary complication rates between GA and LA in patients undergoing CEA. The intraoperative monitoring by duplex scan is an important system to check the quality of the CEA. In a centre with high volumes of carotid surgery, the results, in terms of outcomes, are independent of the anesthetic technique used. At long-term follow-up patency of the vessels undergoing CEA depends on the surgical technique used and not on the type of anesthetic.
References
- . Local versus general anaesthesia for carotid endarterectomy—improving the gold standard?. Eur J Vasc Endovasc Surg. 2008;36:145–149disussion 150-151
- Surgery of the carotid artery: local anaesthesia versus general anaesthesia. Acta Chir Belg. 2007;107:53–57
- . General versus locoregional anaesthesia in carotid surgery: a prospective randomised trial. Vasa. 2006;35:232–238
- . Regional versus general anaesthesia for carotid endarterectomy: impact of change in practice. Surgeon. 2006;4:158–162
- . Comparison between local and general anaesthesia for carotid endarterectomy. A retrospective analysis. Minerva Anestesiol. 2004;70:771–778
- . Patient and hospital benefits of local anaesthesia for carotid endarterectomy. Eur J Vasc Endovasc Surg. 2001;22:13–18
- . Local versus general anaesthesia for carotid endarterectomy. Cochrane Database Syst Rev. 2003;CD000126
- . Carotid endarterectomy and local anaesthesia: reducing the disasters. Cardiovasc Surg. 2000;8:429–435
- The impact of anaesthetic modality on the outcome of carotid endarterectomy. Am J Surg. 2004;188:741–747
- . Local versus general anaesthesia in carotid surgery. A prospective, randomised study. Eur J Vasc Surg. 1989;3:503–509
- . One hundred consecutive carotid reconstructions: local versus general anaesthesia. Am J Surg. 1983;145:477–482
- Locoregional versus general anaesthesia in carotid surgery: is there an impact on perioperative myocardial ischemia? Results of a prospective monocentric randomized trial. J Vasc Surg. 1999;30:131–138
- General anaesthesia versus local anaesthesia for carotid surgery (GALA): a multicentre, randomised controlled trial. Lancet. 2008;372:2132–2142
PII: S0890-5096(10)00218-9
doi:10.1016/j.avsg.2010.04.001
© 2010 Annals of Vascular Surgery Inc. Published by Elsevier Inc All rights reserved.
