Routine Shunting Is a Safe and Reliable Method of Cerebral Protection during Carotid Endarterectomy
Article Outline
The purpose of this report is to describe the perioperative and long-term outcomes of standard carotid endarterectomy (CEA) with general anesthesia, routine shunting, and patching and to show that routine shunting is a safe and reliable method of cerebral protection. Between January 1998 and December 2004, 700 patients attending our Department of Vascular Surgery underwent 786 CEAs performed using a standardized technique. Forty-four patients were excluded from the analysis because they underwent combined CEA and coronary artery bypass grafting, so the analysis is based on the results of 742 CEAs in 656 patients (86 bilateral CEAs). The strict surgical protocol included general anesthesia and standard carotid bifurcation endarterectomy with routine shunting (Javid's shunt) and Dacron patching. The Javid shunts were easily inserted in 738 cases (99.4%) but could not be used in four cases (0.5%) because of the presence of a very small internal carotid artery. The mean ischemic time required to insert the shunt and complete the suture was 4.7 min (±1.15), and the mean time to perform the endarterectomy was 34.3 min (±6.7). The mean follow-up was 24.4 months (±17.3). Overall 30-day mortality was 0.1% (one patient) due to a contralateral major stroke. The 1-month perioperative neurological complication rate was 0.7%, with three major and two minor strokes. The cumulative stroke and death rate was 0.8%. Preoperative symptoms such as hypertension, contralateral occlusion, or an age of more than 80 years were not independent risk factors for perioperative stroke. In the long-term follow-up, Kaplan-Meier analysis indicated an estimated 5-year stroke-free rate of 98.0%. There were eight cases (1%) of >70% restenosis (four cases) or thrombosis (four cases) of the operated internal carotid artery during the follow-up in asymptomatic patients: in four cases, carotid stenting due to >70% restenosis led to good results. The Kaplan-Meier estimate of the restenosis-free rate was 97.8%. The combined stroke and mortality rate of 0.8%, and the restenosis rate of 1% support the argument that standard CEA performed with routine shunting as brain protection leads to excellent early and long-term results.
INTRODUCTION
The aims of carotid surgery are to relieve transient neurological symptoms, prevent strokes (thus improving the patients' quality of life), and, possibly, lengthen survival. Carotid endarterectomy (CEA) is better than medical therapy in treating asymptomatic stenoses of >70% and ipsilateral symptomatic stenoses of ≥50% and preventing major strokes,1, 2 but every carotid surgery team needs to evaluate its rates of operative mortality, postoperative neurological complications, long-term strokes, long-term stroke deaths, and long-term survival.
Carotid surgery raises a number of problems: patient selection, the most appropriate surgical technique (standard versus eversion CEA, direct suture versus patching of the arteriotomy), the use of general or locoregional anesthesia, intra-procedural cerebral protection, and whether or not to use a shunt and, if so, the kind of shunt to use.
The two main areas for which efficacy data are most lacking are shunting and anesthesia. There is some limited evidence that regional anesthesia may lower the risk of nonneurological perioperative complications compared to general anesthesia, but there was no evidence of a reduction in operative stroke risk.3 Moreover, there is still insufficient evidence from randomized controlled trials to support or refute the use of routine or selective shunting during CEA.4, 5
The aim of this report is to describe the perioperative and long-term outcomes of standard CEA performed under general anesthesia with routine shunting and patching to show that routine shunting is a safe and reliable method of cerebral protection.
PATIENTS AND METHODS
Between January 1998 and December 2004, 700 patients attending our Department of Vascular Surgery underwent 786 CEAs performed using a standardized technique; 44 patients were excluded from this analysis because they underwent combined CEA and coronary artery bypass grafting.
The indication for surgery was a >75% carotid stenosis as revealed by a duplex scan evaluated by a vascular specialist in all cases. The velocity criteria used to identify a >75% carotid stenosis were a peak systolic velocity (PSV) of >200 cm/sec, a ratio between the internal carotid artery (ICA) and common carotid artery (CCA) PSV of >4, or an end-diastolic velocity of >100 cm/sec. Angiography was used in 9% of the cases, only when there was suspicion of stenotic lesions involving the origin of the supra-aortic trunks or the intracranial ICA. All patients underwent preoperative brain imaging by means of computed tomography (CT, 87%) or magnetic resonance imaging (MRI, 13%) to show the cerebral status (i.e., ischemic lesions, cortical atrophia) and to detect possible other diseases (e.g., neoplasm); patients received platelet antiaggregation therapy (aspirin or ticlopidine) before and after surgery. The patients' characteristics are shown in Table I.
Table I. Characteristics of the patients undergoing CEA
| Characteristics | Total (742 CEAs in 656 patients) |
|---|---|
| Mean age (years) | 70.5 ± 7.3 |
| Male/female | 519/223 (70%/30%) |
| Hypertension | 371 (50%) |
| Diabetes | 118 (16%) |
| History of smoking | 304 (41%) |
| Coronary heart disease | 400 (54%) |
| Dyslipidemia | 252 (34%) |
| Symptomatic | 482 (65%) |
| Prior stroke | 79 |
| TIA | 242 |
| Amaurosis fugax | 86 |
| Drop attack | 75 |
| >80% Contralateral stenosis | 50 (6.7%) |
| Contralateral carotid occlusion | 36 (4.8%) |
Duplex scans were made 1, 3, and 6 months after the operation and once a year thereafter. Carotid restenosis was defined as a reduction in carotid diameter of ≥50%; reintervention was indicated in the case of a restenosis of ≥70%. The velocity Duplex criteria used to identify a >50% carotid stenosis were a PSV of >150 cm/sec and an ICA:CCA PSV ratio of >3.
The classification, data analysis, and reporting procedures of the study respected the criteria established by the Society for Vascular Surgery/ American Association for Vascular Surgery. Perioperative morbidity was divided into neurological and nonneurological complications occurring within 30 days of surgery. The former were classified as a temporary (<24 hr) lateralizing neurological or ocular event (transient ischemic attack, TIA) or a nondisabling (minor), disabling (major), or fatal stroke. The nonneurological complications were divided into cardiopulmonary complications (postoperative congestive heart failure, myocardial infarction, or respiratory complications requiring intensive care unit admission for >24 hr), wound complications (hemorrhages or hematomas requiring reintervention), or permanent cranial nerve injuries.
Operative Management
All of the operations were performed following a strict protocol involving general anesthesia, routine shunting (Javid shunt; Bard Impra, Tempe, AZ), and a carotid Dacron® patch (Hemashield finesse; Boston Scientific, Wayne, NJ).
The carotid bifurcation was approached by means of a laterocervical incision, with intravenous heparin (5,000 units) being administered before clamping. The CCA was temporarily clamped, and the ICA was closed using a vessel loop tourniquet; a longitudinal arteriotomy was made from the CCA to the ICA as far as necessary in order to ensure direct visualization of the end of the plaque. The shunt (closed with a Kelly clamp) was inserted into the CCA, fixed with a proper clamp, and flushed through with blood from the CCA; the distal shunt was inserted into the ICA, and the occluding Kelly clamp was released.
After the endarterectomy, the arteriotomy was closed using a Dacron impregnated patch and 6/0 polypropylene sutures. The shunt was removed before the completion of the suture, the artery was briefly clamped, and the suture was completed.
We recorded the total ischemic time (to insert and complete the suture when the shunt was removed) and the total time required to perform the endarterectomy. All of the patients were postoperatively monitored in a recovery room for about 2 hr with radial arterial blood pressure and pulse oximetry.
Statistics
The patients' characteristics and risk factors are given as means ± standard deviation (SD), and their perioperative outcomes were compared using the chi-squared test with Yates correction or Student's t-test. The probability of stroke and recurrent stenosis, with and without specific risk factors, was analyzed by means of Kaplan-Meier survival analysis, with p values based on the mean values of the log-rank or Wilcoxon test. The patients' risk factors and characteristics underwent multivariate analysis using logistic regression or Cox's proportional hazards model to obtain odds ratios and 95% confidence intervals. p < 0.05 was considered significant. All data were stored in a computer and statistically analysed using SPSS 11.0 (Chicago, IL).
RESULTS
Perioperative (30 Days)
We performed 742 CEAs in 656 patients (86 bilateral CEAs), excluding combined CEA and coronary artery bypass grafting. The presenting symptoms, risk factors, and comorbidities in patients aged >80 years (73 CEAs) were not significantly different from those of the younger patients. Twenty percent of the preoperative CT or MRI scans were positive for ispilateral ischemic lesions (152 patients, including 45 who were clinically asymptomatic; these patients were classified in the asymptomatic group for the analysis). The mean PSV duplex scan carotid stenosis value was 269 cm/sec (±54.1), and the mean ICA:CCA PSV ratio was 5.3 (± 1.3). A contralateral ICA stenosis of 75–99% was present in 50 cases (6.7%) and a contralateral occlusion in 36 (4.8%).
The shunt was inserted easily in 738 cases (99.4%) but could not be used in four cases (0.6%) because of the small size of the ICA. The mean ischemic time was 4.7 min (± 1.1), and the mean time to perform the endarterectomy was 34.3 min (±6.7). In the only one case of shunt thrombosis (0.1%), the shunt was changed and the patient remained asymptomatic.
Overall 30-day mortality was 0.1% (one patient) related to a contralateral major stroke occurring 14 days after CEA in a patient with an 80% contralateral stenosis. The 30-day perioperative nonfatal neurological complication rate was 0.9%, with three major contralateral strokes in symptomatic patients with contralateral carotid artery occlusion, two minor ipsilateral strokes (one probably due to embolic debridements from the common carotid plaque through the shunt; the cause of the other one is unknown but the patient died 3 months after the operation due to brain metastasis), and two TIAs. Two strokes (one major and one minor) occurred when patients woke up (intraoperative stroke); the other three strokes occurred within 24 hr (postoperative strokes). In all cases, duplex scans revealed a patent ICA endarterectomized. The postoperative CT scan of three major strokes revealed a new brain infarction contralateral to the operated carotid artery. CT scan of minor strokes was negative.
Statistical analysis showed that preoperative symptoms, hypertension, contralateral asymptomatic occlusion, or age of >80 years were not independent risk factors for perioperative stroke; mean ischemic time was not significantly different between patients with or without postoperative neurological symptoms (4.7 vs. 4.5 min). However, contralateral symptomatic ICA occlusion was a significant independent risk factor for postoperative neurological events (p < 0.001) (Table II).
Table II. Risk factors for perioperative stroke
| Stroke as predicted by | p | OR | 95% CI |
|---|---|---|---|
| Preoperative symptoms | 0.28 | 0.89 | 0.72–1.10 |
| Hypertension | 0.56 | 1.05 | 0.89–1.24 |
| Female | 0.22 | 1.34 | 0.84–2.12 |
| Age >80 years | 0.33 | 1.12 | 0.89–1.42 |
| Asymptomatic contralateral carotid occlusion | 0.06 | 1.74 | 0.96–3.12 |
| Symptomatic contralateral carotid occlusion | 0.001 | 0.72 | 0.65–0.85 |
There were eight nonneurological complications (1%): two were due to congestive heart failure, four (0.5%) to myocardial infarction, and two to respiratory complications requiring admission to the intensive care unit. All patients with nonneurological complications were asymptomatic when discharged. Cranial nerve injuries occurred in nine cases (1.2%): six cases of recurrent laryngeal nerve paralysis and three cases of hypoglossal nerve injury. Eight cases (1%) required reintervention because of wound hematoma.
Long-Term
The mean follow-up was 24.4 months (±17.3). No patient was lost to follow-up during the first 6 months; three were lost between 6 and 12 months after surgery; and another 15 were lost thereafter.
The 5-year Kaplan-Meier estimated stroke-free rate was 98.0% (Fig. 1).
Fig. 1.
The Kaplan-Meier estimated stroke-free rate.
There were 13 cases (1.7%) of >50% restenosis or thrombosis of the operated ICA during the follow-up period. Four cases of asymptomatic ICA thrombosis (0.5%) occurred after 1, 3, and 6 months (mean 4 months), three of which had not received a shunt because of their very small ICAs. The nine cases of >50% restenosis (1.2%) occurred in asymptomatic patients after a mean follow-up of 20 months (± 23.1): in four cases (0.5%) with >70% restenosis, carotid stenting led to good results. The 5-year Kaplan-Meier estimated recurrent stenosis-free rate was 97.8% (Fig. 2).
Fig. 2.
The 5-year Kaplan-Meier estimated recurrent stenosis-free rate.
No patch or neck infections occurred during the perioperative or follow-up periods. One patient (0.1%) experienced a pseudoaneurysm after 42 months due to chronic patch disruption and underwent a successful reoperation with prosthetic reconstruction of the carotid artery.
DISCUSSION
Most of the published data indicate that restenosis rates are generally lower with patch rather than primary closure of the arteriotomy (0.1–5.8% vs. 1–14%).6, 7, 8 Our results showing a 1% rate of >70% restenosis or thrombosis, with a 5-year Kaplan-Meier stenosis-free rate of 97.8%, confirm that carotid patch closure leads to excellent long-term results.
Shunting during CEA remains controversial. The first use of a temporary shunt was reported by Cooley et al. in 1956,9 and the elective application of temporary shunting was advocated by Thompson et al. in the early 1960s.10 The practice of intraoperative cerebral monitoring varies between surgeons and institutions depending on local tradition, surgeons' preferences, technical resources, economy, and the fact that no single method is superior in all aspects.11 Some surgeons have used it routinely, such as Thompson;12 some use it selectively on the basis of the assessment of cerebral collateral circulation, and a few state they rarely or never use it.
A Cochrane review published in 2002 concluded that the available data were too limited to support or refute the use of routine or selective shunting in CEA, and no one method of monitoring selective shunting has been shown to produce better outcomes.4
Ten years ago, Riles et al.13 showed that most perioperative strokes were due to technical errors made during CEA: they reported that 8/10 perioperative strokes were related to difficulties in shunt placement, three of which involved patients with contralateral carotid artery occlusion, and concluded that preoperative stroke was a predictor of perioperative stroke. Some years later, other authors demonstrated worse results when shunting was used in selected patients with encephalic ischemic signs during various cerebral protection systems.14, 15, 16, 17 All of these authors suggest that neurological complications are directly attributable to intraluminal shunting.
In our study, we found only one case of a neurological complication that may have been due to embolic debridements through the shunt; the three major perioperative strokes were contralateral in patients with contralateral ICA occluded and preoperative ischemic symptoms due to this occlusion. The cause of the other minor stroke was a metastatic brain cancer.
There are some reasons for believing that the shunt complications reported in the literature may be related to placement difficulties on the part of a hurried and imprecise surgeon who does not routinely perform it. This is particularly true when an awake patient under locoregional anesthesia manifests dramatic ischemic symptoms such as coma or convulsions.
We believe that, if the patient is sleeping, a shunt can be placed as a routine maneuver quickly, calmly, and more accurately, thus visualizing the end of the plaque and avoiding its dissection.
In our series, contralateral symptomatic carotid artery occlusion was significantly associated with risk of postoperative stroke. The postoperative CT scan of these patients revealed a new brain infarction contralateral to the operated carotid artery: we postulate that the ischemic time to insert the shunt was the cause of stroke.
In this subgroup of patients, the reduced ischemic time of carotid stenting may lead to better results in terms of postoperative neurological complications.
A recent randomized trial comparing carotid artery stenting to endarterectomy showed a 4.8% 30-day incidence of stroke and death in the carotid stenting arm; the authors concluded that carotid stenting with the use of an emboli-protection device is not inferior to CEA (cumulative stroke and death 5.6% in the endarterectomy arm).18 This is lower than that described in previous reports of carotid artery stenting but higher than the 1–2.6% described in recently published reports of CEA.19, 20, 21 Our data demonstrating a 5-year >70% restenosis or thrombosis rate of 1 % confirm that the results of CEA are long-lasting, so clinical trials comparing CEA and carotid stenting require a sufficiently long follow-up in order to assess long-term restenosis rates and functional outcomes. Furthermore, although less invasive, carotid stenting must demonstrate equivalent robustness before it can be considered a viable alternative to surgery.
The combined stroke and mortality rate of 0.8% observed in this study supports the argument that standard CEA performed under general anesthesia with routine shunting and patch reconstruction leads to excellent early and long-term results and is a safe and cheap means of performing carotid surgery. Nevertheless, a randomized trial is mandatory to prove that routine shunting is better than other cerebral protection systems.
We are aware that shunting is not always necessary: however, its routine use not only avoids the need for sometimes very expensive or complicated means of cerebral monitoring but also makes the insertion maneuvers easy and safe, thus practically eliminating complications.
REFERENCES
- National Institute of Neurological Disorders and Stroke, Stroke and Trauma Division. North American Symptomatic Carotid Endarterectomy Trial (NASCET) investigators. Clinical alert: benefit of carotid endarterectomy for patients with high-grade stenosis of the internal carotid artery . Stroke . 1991;22:816–817
- . Prevention of disabling and fatal strokes by successful carotid endaretrectomy in patients without recent neurological symptoms: randomized controlled trial . Lancet . 2004;363:1491–1502
- . Local versus general anaesthesia for carotid endarterectomy . Cochrane Database Syst Rev . 2004;2:CD000126
- . Routine or selective shunting for carotid andarterectomy (and different methods of monitoring in selective shunting) . Cochrane Database Syst Rev . 2002;2:CD000190
- . Group-Interim results for symptomatic patients with severe (70–99%) or with mild (0–29%) carotid stenosis . Lancet . 1991;337:1235–1243
- . A fifteen-year experience with carotid endarterectomy after a formal protocol requiring highly frequent patch angioplasty . J Vasc Surg . 2000;31:724–735
- . Carotid artery restenosis: an ongoing disease process . Am Surg . 2002;68:275–280
- Eversion versus conventional carotid endarterectomy: late results of a prospective multicenter randomized trial . J Vasc Surg . 2000;31:19–30
- . Surgical treatment of arteriosclerotic occlusion of common carotid artery . J Neurosurg . 1956;13:500–506
- . Endarterectomy of the totally occluded carotid artery for stroke . Arch Surg . 1967;95:791–801
- . Carotid endarterectomy. A survey of UK anaesthetic practice . Anaesthesia . 2000;55:481–485
- . Carotid surgery: the past is prologue . J Vasc Surg . 1997;25:131–138
- The cause of perioperative stroke after carotid endarterectomy . J Vasc Surg . 1994;19:206–216
- . Surgical therapy for the patient with internal carotid artery occlusion and contralateral stenosis . J Vasc Surg . 1987;5:856–861
- Neurologic changes during carotid endarterectomy under cervical block predict a high risk of postoperative stroke . Anaesthesiology . 1993;78:829–833
- . Technique and clinical results of carotid stump back-pressure to determine selective shunting during carotid endarterectomy . J Vasc Surg . 1991;13:319–327
- . Hemodynamic ischemic stroke during carotid endarterectomy: an appraisal of risk and cerebral protection . J Vasc Surg . 1997;25:611–619
- (Sapphire trial). Protected carotid artery stenting versus endarterectomy in high-risk patients . N Engl J Med . 2004;351:1493–1501
- Carotid endarterectomy by eversion technique: its safety and durability . Ann Surg . 1998;228:471–478
- . Durability of carotid endarterectomy . Stroke . 2003;34:2941–2944
- . Early outcome assessment for 2228 consecutive carotid endarterectomy procedures: the Cleveland Clinic experience from 1989 to 1995 . J Vasc Surg . 1997;26:1–10
PII: S0890-5096(06)61466-0
doi:10.1007/s10016-006-9037-8
© 2006 Annals of Vascular Surgery, Inc. Published by Elsevier Inc All rights reserved.
