- Email: [email protected]
High-risk carotid endarterectomy and high-risk carotid surgery: is surgery or stenting the best choice?
International Congress Series 1247 (2002) 319 – 326
High-risk carotid endarterectomy and high-risk carotid surgery: is surgery or stenting the best choice? Christopher M. Loftus * Department of Neurological Surgery, The University of Oklahoma, College of Medicine, 711 Stanton L. Young Boulevard #206, Oklahoma City, OK 73116, USA
Abstract High-risk carotid surgery means that the likelihood of a complication is increased over the levels that are customarily accepted for carotid reconstruction. There are many instances where a high-risk operation should be performed, and where the increased risk is acceptable to the patient and the physician. In this presentation, I will identify the cases that I consider to be in high-risk groups and will present my evaluation and management strategy for these patients. Clearly, the surgeon considering a high-risk operation should have a good understanding of the natural history of carotid disease and the risks of surgery compared to the expected risks of medical treatment. I divide high-risk operations into several major categories. These include predictably difficult operations, cases with high risk for medical complications, cases with high risk for intraoperative ischemia, and cases with high risk for postoperative occlusion/stroke. Predictably, difficult operations include those with peculiar anatomical variants, reoperation, irradiated carotids, or extreme high exposures. With special emphasis on technique, these cases can be well performed surgically and the presentation will illustrate such cases. Patients with high-risk for medical complications include those with recent myocardial infarction, unstable angina, diabetic patients, patients with severe pulmonary disease, and patients who require continued postoperative anticoagulation (usually for heart valves). In many cases, these patients can undergo successful surgery; some such cases may also be suitable for endovascular treatment, particularly for recent myocardial disease. Patients with high-risk for intraoperative ischemia require that the surgeon have a solid knowledge of shunt placement and other strategies for cerebral protection. These include patients with contralateral occlusion, difficult shunt placements, high or difficult exposures, unstable neurological deficits, and propagating intraluminal thrombus high up the internal carotid.
*
Tel.: +1-405-271-4912; fax: +1-405-271-3091. E-mail address: [email protected] (C.M. Loftus).
0531-5131/02 D 2002 Published by Elsevier Science B.V. PII: S 0 5 3 1 - 5 1 3 1 ( 0 2 ) 0 1 0 4 4 - 0
320
C.M. Loftus / International Congress Series 1247 (2002) 319–326
Patients with high risk for postoperative occlusion or stroke include diabetic patients, high focal lesions in the ICA, patients with postoperative hypotension, and hypertensive patients with headache following opening of a tight stenosis (dysautoregulated brain). In the majority of these cases, the use of a Hemashield patch graft will reduce or eliminate the chances of postoperative problems. Carotid surgery can be well performed by an experienced surgeon even in high-risk groups, and at present it remains the best-proven treatment for these cases. In the future, endovascular treatments may become common for some of these cases, but the scientific evidence to support this must still be proven. The most important factors in assuring a successful operation are careful planning, experience, and a high skill level for the operating surgeon. D 2002 Published by Elsevier Science B.V. Keywords: Carotid endarterectomy; Stenting; Surgery
1. Introduction Let me begin this discussion of high-risk carotid surgery with a review of the indications for carotid treatment in symptomatic or asymptomatic patients. For the purpose of this discussion, I will avoid the use of the term ‘‘carotid surgery’’ and rather say ‘‘carotid treatment’’, although it must be remembered that the cooperative trials that we will discuss here compared surgery versus medical therapy and did not include endovascular treatment as part of their randomization or outcomes-based paradigm. I offer carotid treatment to asymptomatic patients with greater than or equal to 60% stenosis, based on the results of the ACAS trial. I am cognizant of the fact that a tangible benefit was not demonstrated for women in this trial and I have explained this to female patients considering carotid reconstruction. For symptomatic patients, I rely on the combined and corroborating results of the NASCET, ECST, and VASST trials, which determined conclusively that surgery was of benefit for patients with linear angiographic stenosis of greater than or equal to 50%. Surgery was of higher benefit for patients with linear angiographic stenosis of greater than or equal to 70%. Patients with stenosis less than 50% should not be offered surgical and/or endovascular treatment and are best managed with optimal medical therapy. There are a number of other situations in which I offer carotid treatment to patients. The first of these encompasses the group of patients with so-called ‘‘stump syndromes’’ in which an internal carotid occlusion has left behind a residual carotid stump at the takeoff of the internal carotid from the cervical carotid bifurcation. Under these circumstances, if the patient has ipsilateral transient ischemic type symptoms, and if all other potential sources of embolic phenomena are excluded, I offer treatment, customarily consisting of surgical intervention with common to external carotid endarterectomy and ligation of the offending stump with a large Weck clip or with hand-sewn plication of the carotid stump. This is a low-risk procedure that has a very tangible benefit if done properly and in the proper patients. I offer surgery to symptomatic patients with recurrent carotid artery stenosis so long as they understand the higher risk/benefit ratio inherent in such treatment. Asymptomatic patients with recurrent and progressing high-grade stenosis may well be offered treatment
C.M. Loftus / International Congress Series 1247 (2002) 319–326
321
also but the risk/benefit ratio in this group is higher still and careful thought needs to be given to how the patient should be counseled. For patients with intraluminal thrombus, as I have published elsewhere, I propose delayed treatment after a period of anticoagulation to dissolve the thrombus and make for a safer treatment, whether surgery or endovascular. In point of fact, I have never offered endovascular treatment to such a patient and all patients after anticoagulation have been surgically reconstructed with excellent results. Surgery for patients with recent stroke is not delayed in my practice unless there is a decreased level of consciousness or a large space-occupying infarct evident on axial cerebral imaging. The old wisdom that 6 weeks should elapse to allow autoregulation to be reestablished in the brain is, in my mind, putting the patient at high risk for recurrent embolic phenomena and I do not delay surgery except when the above criteria are invoked. Likewise, patients who are heparinized because of recent ischemic phenomena are maintained on full heparinization up to and through the time of surgery and this has not in my practice produced any untoward hemorrhagic consequences either in the neck or intracranially. We must then ask ourselves, in considering the dilemma of whether to offer carotid patients surgical reconstruction versus endovascular stenting, whether there is an outcomes-based reason to choose endovascular over surgical treatment. If we are going to make a rational comparison to these two treatment strategies, we need to consider and take into account the standard complications reported in carotid surgery to allow us to evaluate both treatment strategies in an equal light. Outcome analysis in carotid reconstruction and carotid endovascular treatment is divided in the following way. The first consideration is perioperative stroke and this event is customarily stratified into major or minor stroke and likewise ipsilateral or allstroke. The second category to be considered is that of wound complications, meaning primarily hematoma or wound infection. The third category is nerve injury and the fourth category is medical complications, most commonly perioperative myocardial infarction. If we look at the surgical results from randomized cooperative trials, we see that the complication rates are gratifyingly quite low. In the ACAS trial, the perioperative stroke and death rate was 2.3%. In the VAAST trial, the 30-day mortality was 1.9%, all referable to myocardial infarction. The perioperative stroke rate was 2.4%. The nerve injury rate was 3.8% and the rate of nonfatal myocardial infarction was likewise 1.9%. In evaluating the results of symptomatic cooperative trials, we look first at the NASCET trial 9 where the 30-day mortality was 0.6%. The perioperative major stroke and death rate was 2.1%. The all-stroke rate was 5.5%. The nerve injury rate, evaluated by independent examiners, was 7.6%, the wound hematoma rate was 5.5% and the myocardial infarction rate was 0.9%. In the ECST symptomatic cooperative trial 7, it is very difficult to tease out the major stroke or death rate because the data is presented in several different manuscripts. However, it appears to be quite variable depending on the stratification of percent stenosis being considered. In the 0.9% surgery group, the major death or stroke rate was 2.3%. In the 30 – 49% group, it was 8%. In the 50 –69% group, it was 7.9% and in the 70 – 99% group, it was 3.7%. In the VA symptomatic surgery trial (VASST) 6, the 30-day mortality was 3.3%, the perioperative stroke rate was 2.2%, the
322
C.M. Loftus / International Congress Series 1247 (2002) 319–326
nerve injury rate was 5%, the wound hematoma rate was 5%, and the myocardial infarction rate was 2%. If we look at the results, cooperative trials customarily represent fixed observation points and evaluation by nonbiased observers. Randomized trial data is often felt to be the most valid and the highest level of evidence data that we have available to make therapeutic judgements. If we look at the published results of individual skilled carotid surgeons, we come up with data that sheds a bit more of a favorable light on carotid reconstruction. It is to be recalled, nonetheless, that essentially all the major cooperative trials demonstrated that carotid artery surgery was superior to medical therapy despite their higher complication rates as compared to the results of individual surgeons. Returning to the individual surgeon’s published results category, the data published by Spetzler et al. [10] for carotid microsurgery shows that in 200 cases the all-stroke rate was 1%, the mortality was 0.5%, the wound hematoma rate was 1.5% and the nerve injury rate was 5%. The data presented by Findlay [1] in carotid microsurgery document an all-stroke rate of 1.7%, a wound hematoma rate of 3.3% and a nerve injury rate of 5.1%. The data presented by Steiger et al. [11], again considering carotid microsurgery, document a perioperative stroke rate of zero, a myocardial infarction rate of 2%, a mortality of 2% and a nerve injury rate of 2%. The data presented by Harbaugh [2] employing regional anesthesia with loupemagnified carotid reconstruction shows an all-stroke rate of 2.5% with an ipsilateral rate of 1.3% and a major stroke rate of 0.9%. Harbaugh’s myocardial infarction rate is 1.3%. The wound hematoma rate is 2.1% and the nerve injury rate is 3%. Finally, my own data representing conventional loupe-magnified carotid reconstruction, under general anesthesia, with Hemashield patch angioplasty in the second half of the series, shows an all-stroke rate of 1.8% with a major stroke rate of 0.72%. My mortality rate is 0.72% and my wound hematoma rate is 0.4% [5]. If you look then at the data from leading surgeons published on an individual basis with their own observations of complications, we find that the all-stroke rate is customarily less than 2.6%, the nerve injury rate is less than 5.1%, the mortality is less than 2% and the wound hematoma rate varies between a low of 0.35% and a high of 3.35%. We must then ask ourselves the question which carotid lesions do endovascular specialists propose for treatment with stenting as an alternative to open surgical carotid reconstruction? The first manuscript we should consider is that by Theron et al. [12], stenting patients with primary stenosis. This paper discusses a protection scheme in about half the patients with distal balloon protection being employed during the stenting procedure. Two hundred fifty nine patients were treated, of which 136 underwent cerebral protection and 123 did not. In the unprotected cases the rate of carotid dissection was 5% and the rate of embolization was 8%. In the protected cases, no dissection is reported and the embolization rate is 1%. Some of these patients were not stented but merely had angioplasty. In these authors’ experience, stenting reduced the restenosis rate from 16% to 4% and the authors’ conclusion was that all patients undergoing endovascular carotid treatment should be stented and should be treated with cerebral protective strategies during the stenting process. The second endovascular manuscript to be considered is that of Yadav et al. [13], in which 126 arteries were treated, all stented. There were seven minor strokes, two major
C.M. Loftus / International Congress Series 1247 (2002) 319–326
323
strokes, and one death in this series. The stroke and death rate was 7.9% with an ipsilateral major stroke risk of 1.6%. The restenosis rate was quoted at 4.9%; all of these patients were asymptomatic. Yadav concluded that it was feasible to stent high-risk carotid patients who were not considered to be surgical candidates. Interesting data is reported by Lanzino et al. [4] regarding recurrent carotid stenosis and we point out that this data is in the neurosurgical literature. Lanzino et al. treated 25 arteries for recurrent stenosis with percutaneous angioplasty alone in 7 and placement of a stent in 18. There were no major neurological or cardiac events. Complications were limited to one TIA and one femoral pseudo aneurysm. Lanzino et al. demonstrated that 60% of angioplasty-alone patients restenosed and only one stent patient restenosed. Accordingly, their recommendation was first that the stenting procedure without protective strategies was safe for recurrent carotid stenosis and that stenting was far superior to angioplasty alone in the treatment of restenotic carotid disease. A contrary viewpoint is presented by Naylor et al. [8] in the United Kingdom randomized trial of angioplasty/stenting versus carotid surgery. In this trial, 23 patients were randomized who had greater than or equal to 70% carotid stenosis. Seventeen of these patients were treated, ten of whom had carotid surgery with no complications. Seven patients underwent either angioplasty or stent and five of these patients had a stroke. This trial was prematurely stopped because of an extremely negative outcome in the endovascular side and has been reported as strong evidence to suggest that carotid surgery is superior to endovascular carotid treatment. Non-atheromatous lesions such as fibromuscular dysplasia have also been studied with endovascular treatment. Hasso et al. [3] report three cases, all of which were symptomatic, which were treated with angioplasty alone in which no complications were evident. This is an extremely small series but it represents essentially the only experience in the literature regarding the endovascular treatment of fibromuscular dysplasia. Diethrich et al. discussed primary stenting for carotid surgery in 110 patients, treating 117 arteries. There were seven strokes (two major) for a stroke rate of 6.4%. Three patients were converted from endovascular treatment to open carotid artery surgery, with a rate of 2.7% for conversion. There were two deaths, one cardiac and one stroke, for a mortality of 1.8%. The 30-day angiographic success rate was 89.1% and there was an 89% patency rate at 7.6 months follow-up in this series. If endovascular treatment is to be considered, we should likewise ask the question ‘‘which lesions are there which the surgeon does not want to operate on?’’ This brings to my mind the central question of what constitutes a high-risk carotid operation. I would put high-risk patients in several categories. The first consists of patients who have predictably difficult operations. The second consists of patients who are at a high risk for medical complications. The third consists of patients who are at a high risk for intraoperative ischemia, and the fourth consists of patients who, for anatomical or other reasons, are at high risk for postoperative occlusion and/or stroke. Considering the first category, predictably difficult operations would include anatomical variants such as a medially directed internal carotid artery, a long and high carotid plaque, a high carotid bifurcation, or other anatomical anomalies such as cervical carotid aneurysm which would render an operation significantly more difficult. Predictably difficult operations also include carotid reoperation, operation upon carotids that have
324
C.M. Loftus / International Congress Series 1247 (2002) 319–326
been previously irradiated for other reasons, and cases in which distal exposure will be extremely high, in the region of C1 or C2 on a lateral angiogram. In the second category, patients at high risk for medical complication, I would include patients with recent myocardial infarction, patients with unstable angina, diabetic patients (who in my experience are at high risk for postoperative occlusion), patients with severe pulmonary disease rendering general anesthesia a major risk, and patients who require preoperative anticoagulation. In the third category, those at high risk for intraoperative ischemia, I would include patients with contralateral occlusion, although in my experience this does not appreciably increase the risk. Patients who will have difficult or impossible shunt placement have a high risk for intraoperative ischemia if a shunt is required. This includes patients who have a right angle bend distal to the carotid plaque and in whom this bend cannot be straightened out through operative exposure, or patients who have a large cervical aneurysm that precludes placement of a shunt. High risk for intraoperative ischemia also includes patients with difficult cervical carotid exposures, patients who are neurologically unstable with a fluctuating deficit, and patients who have a propagating thrombus in the cervical carotid artery distal to an area of high-grade stenosis but who continue to have antegrade flow beyond the region of the thrombus. Patients at high risk for postoperative occlusion and/or stroke include the following categories. Diabetic patients, in my experience, have a higher risk than normal for a postoperative occlusion, patients who experience postoperative hypotension seem to have a higher risk for carotid occlusion, patients in whom a repair is done high up in the internal carotid artery with consequent narrowing of the internal carotid are at high risk for occlusion and patients who are hypertensive who experience dysautoregulation type headache following reconstruction of a tight stenosis are at high risk for postoperative intracerebral hemorrhage. In my experience, most of these complications are avoided by the use of the Hemashield patch graft, which I now employ in every case. Let me point out then the cases which I take particular pleasure and pride in reconstructing surgically. These include primary stenosis, recurrent stenosis, irradiated carotids, cases with previous cervical irradiation or radical neck dissection, contralateral occlusion, high exposures, and perhaps the most enjoyable of all, cases which are turned down by other surgeons and referred for tertiary care center specialty attention. In my opinion, all of these cases can be satisfactorily addressed with reasonable surgical morbidity and mortality by an experienced carotid surgeon with a careful management approach. Cases I might send for stenting include the following. Cases of carotid dissection, which we previously treated with anticoagulation alone, may well be managed by a careful cannulation of the vessel and crossing of the lesion followed by placement of a stent. This, in my mind, represents a very significant advance over any surgical treatment or over anticoagulation alone as a therapeutic strategy. Cases with high cervical carotid aneurysms, beyond the reach of the surgeon, are well treated by stenting across the mouth of the lesion. Cases in which extreme high exposures are required at the base of the skull may well be better treated by stenting than by surgical exposure particularly when one considers the risk of cranial nerve injury. Patients who are medically unstable with recent
C.M. Loftus / International Congress Series 1247 (2002) 319–326
325
myocardial infarction are probably better off treated with endovascular treatment than with surgery, which entails an increased risk from general anesthesia. Finally, cases of fibromuscular dysplasia, which I customarily do not treat with surgery, may well be rational cases for an endovascular stenting approach. In conclusion, it is my opinion that the endovascular versus surgical debate for carotid artery therapy and/or asymptomatic patients will continue for certainly the next 5 –10 years, and most likely both treatments will ultimately enjoy some degree of acceptance. I would reiterate that at the present time only carotid artery surgery is proven by level one evidence to be superior to medical therapy and endovascular treatment stands where we stood with carotid artery surgery approximately 15 years ago, leaving us to accept only the results of individual practitioners rather than any randomized data. This will not forever be the case and randomized data will be available for surgery versus endovascular treatment in major centers where presumably the lowest morbidity and mortality can be obtained in each camp by skilled practitioners of the art. Trained, skilled carotid artery surgeons offer an excellent treatment for stroke prevention in symptomatic or asymptomatic patients. There is no documentation in literature for the superiority of endovascular treatment at the present time. However, I am quick to point out that selected cases, as I mentioned in the body of this work, may be more appropriately treated with endovascular strategies and I personally refer patients for such treatment under special circumstances as mentioned such as medically unstable patients, very high cervical carotid exposures, carotid dissection, and/or fibromuscular dysplasia. Individual referral patterns will depend on the surgeon’s comfort with high-risk patients. For surgeons with lesser carotid experience or whose morbidity and mortality data may not be quite satisfactory, endovascular referrals may be an appropriate strategy. For surgeons with a large degree of experience and a high degree of comfort in handling the challenge of complex carotid reconstruction cases, carotid surgery continues to represent the only therapy proven by randomized trials to be superior to medical treatment in specific and selected patients [6,7,9].
References [1] J.M. Findlay, Carotid microendarterectomy, Neurosurgery 32 (1993) 792 – 798. [2] R.E. Harbaugh, Carotid surgery under local anesthesia, Tech. Neurosurg. 3 (1997) 25 – 33. [3] A.N. Hasso, C.R. Bird, D.E. Zinke, J.R. Thompson, Fibromuscular dysplasia of the internal carotid artery: percutaneous transluminal angioplasty, Am. J. Radiol. 136 (1980) 955 – 960. [4] G. Lanzino, R.A. Mericle, D.K. Lopes, A.K. Wakhloo, L.R. Guterman, L.N. Hopkins, Percutaneous transluminal angioplasty and stent placement for recurrent carotid artery stenosis, J. Neurosurg. 90 (1999) 688 – 694. [5] C.M. Loftus, T.F. Kresowik, Anatomical basis and technique of carotid endarterectomy, in: C. Loftus, T. Kresowik (Eds.), Carotid Artery Surgery, Thieme Medical Publishers, New York, 2000. [6] M.R. Mayberg, E. Wilson, F. Yatsu, et al., Carotid endarterectomy and prevention of cerebral ischemia in symptomatic carotid stenosis, JAMA 266 (1991) 3289 – 3294. [7] MRC European Carotid Study Trialists, Interim results for symptomatic patients with severe (70 – 99%) or with mild (0 – 29%) carotid stenosis, Lancet 337 (1991) 1235 – 1243. [8] A.R. Naylor, A. Bolia, R.J. Abbott, et al., Randomized study of carotid angioplasty and stenting versus carotid endarterectomy: a stopped trial, J. Vasc. Surg. 28 (1998) 326 – 334.
326
C.M. Loftus / International Congress Series 1247 (2002) 319–326
[9] North American Symptomatic Carotid Endarterectomy Trial Collaborators, Beneficial effect of carotid endarterectomy in symptomatic patients with high grade stenosis, N. Engl. J. Med. 325 (1991) 445 – 453. [10] R.F. Spetzler, N. Martin, M.N. Hadley, R. Thompson, E. Wilkinson, P. Raudzens, Microsurgical endarterectomy under barbiturate protection: a prospective study, J. Neurosurg. 51 (1986) 147 – 150. [11] H. Steiger, L. Schaffler, S. Liechti, Results of microsurgical carotid endarterectomy, Acta Neurochir. 100 (1989) 31 – 38. [12] J.G. Theron, G.G. Payelle, O. Coskun, H.F. Huet, L. Guimaraens, Carotid artery stenosis: treatment with protected balloon angioplasty and stent placement, Radiology 201 (1996) 627 – 636. [13] J.S. Yadav, G.S. Roubin, S. Iyer, et al., Elective stenting of the extracranial carotid arteries, Circulation 95 (1997) 376 – 381.
High-risk carotid endarterectomy and high-risk carotid surgery: is surgery or stenting the best choice? Christopher M. Loftus * Department of Neurological Surgery, The University of Oklahoma, College of Medicine, 711 Stanton L. Young Boulevard #206, Oklahoma City, OK 73116, USA
Abstract High-risk carotid surgery means that the likelihood of a complication is increased over the levels that are customarily accepted for carotid reconstruction. There are many instances where a high-risk operation should be performed, and where the increased risk is acceptable to the patient and the physician. In this presentation, I will identify the cases that I consider to be in high-risk groups and will present my evaluation and management strategy for these patients. Clearly, the surgeon considering a high-risk operation should have a good understanding of the natural history of carotid disease and the risks of surgery compared to the expected risks of medical treatment. I divide high-risk operations into several major categories. These include predictably difficult operations, cases with high risk for medical complications, cases with high risk for intraoperative ischemia, and cases with high risk for postoperative occlusion/stroke. Predictably, difficult operations include those with peculiar anatomical variants, reoperation, irradiated carotids, or extreme high exposures. With special emphasis on technique, these cases can be well performed surgically and the presentation will illustrate such cases. Patients with high-risk for medical complications include those with recent myocardial infarction, unstable angina, diabetic patients, patients with severe pulmonary disease, and patients who require continued postoperative anticoagulation (usually for heart valves). In many cases, these patients can undergo successful surgery; some such cases may also be suitable for endovascular treatment, particularly for recent myocardial disease. Patients with high-risk for intraoperative ischemia require that the surgeon have a solid knowledge of shunt placement and other strategies for cerebral protection. These include patients with contralateral occlusion, difficult shunt placements, high or difficult exposures, unstable neurological deficits, and propagating intraluminal thrombus high up the internal carotid.
*
Tel.: +1-405-271-4912; fax: +1-405-271-3091. E-mail address: [email protected] (C.M. Loftus).
0531-5131/02 D 2002 Published by Elsevier Science B.V. PII: S 0 5 3 1 - 5 1 3 1 ( 0 2 ) 0 1 0 4 4 - 0
320
C.M. Loftus / International Congress Series 1247 (2002) 319–326
Patients with high risk for postoperative occlusion or stroke include diabetic patients, high focal lesions in the ICA, patients with postoperative hypotension, and hypertensive patients with headache following opening of a tight stenosis (dysautoregulated brain). In the majority of these cases, the use of a Hemashield patch graft will reduce or eliminate the chances of postoperative problems. Carotid surgery can be well performed by an experienced surgeon even in high-risk groups, and at present it remains the best-proven treatment for these cases. In the future, endovascular treatments may become common for some of these cases, but the scientific evidence to support this must still be proven. The most important factors in assuring a successful operation are careful planning, experience, and a high skill level for the operating surgeon. D 2002 Published by Elsevier Science B.V. Keywords: Carotid endarterectomy; Stenting; Surgery
1. Introduction Let me begin this discussion of high-risk carotid surgery with a review of the indications for carotid treatment in symptomatic or asymptomatic patients. For the purpose of this discussion, I will avoid the use of the term ‘‘carotid surgery’’ and rather say ‘‘carotid treatment’’, although it must be remembered that the cooperative trials that we will discuss here compared surgery versus medical therapy and did not include endovascular treatment as part of their randomization or outcomes-based paradigm. I offer carotid treatment to asymptomatic patients with greater than or equal to 60% stenosis, based on the results of the ACAS trial. I am cognizant of the fact that a tangible benefit was not demonstrated for women in this trial and I have explained this to female patients considering carotid reconstruction. For symptomatic patients, I rely on the combined and corroborating results of the NASCET, ECST, and VASST trials, which determined conclusively that surgery was of benefit for patients with linear angiographic stenosis of greater than or equal to 50%. Surgery was of higher benefit for patients with linear angiographic stenosis of greater than or equal to 70%. Patients with stenosis less than 50% should not be offered surgical and/or endovascular treatment and are best managed with optimal medical therapy. There are a number of other situations in which I offer carotid treatment to patients. The first of these encompasses the group of patients with so-called ‘‘stump syndromes’’ in which an internal carotid occlusion has left behind a residual carotid stump at the takeoff of the internal carotid from the cervical carotid bifurcation. Under these circumstances, if the patient has ipsilateral transient ischemic type symptoms, and if all other potential sources of embolic phenomena are excluded, I offer treatment, customarily consisting of surgical intervention with common to external carotid endarterectomy and ligation of the offending stump with a large Weck clip or with hand-sewn plication of the carotid stump. This is a low-risk procedure that has a very tangible benefit if done properly and in the proper patients. I offer surgery to symptomatic patients with recurrent carotid artery stenosis so long as they understand the higher risk/benefit ratio inherent in such treatment. Asymptomatic patients with recurrent and progressing high-grade stenosis may well be offered treatment
C.M. Loftus / International Congress Series 1247 (2002) 319–326
321
also but the risk/benefit ratio in this group is higher still and careful thought needs to be given to how the patient should be counseled. For patients with intraluminal thrombus, as I have published elsewhere, I propose delayed treatment after a period of anticoagulation to dissolve the thrombus and make for a safer treatment, whether surgery or endovascular. In point of fact, I have never offered endovascular treatment to such a patient and all patients after anticoagulation have been surgically reconstructed with excellent results. Surgery for patients with recent stroke is not delayed in my practice unless there is a decreased level of consciousness or a large space-occupying infarct evident on axial cerebral imaging. The old wisdom that 6 weeks should elapse to allow autoregulation to be reestablished in the brain is, in my mind, putting the patient at high risk for recurrent embolic phenomena and I do not delay surgery except when the above criteria are invoked. Likewise, patients who are heparinized because of recent ischemic phenomena are maintained on full heparinization up to and through the time of surgery and this has not in my practice produced any untoward hemorrhagic consequences either in the neck or intracranially. We must then ask ourselves, in considering the dilemma of whether to offer carotid patients surgical reconstruction versus endovascular stenting, whether there is an outcomes-based reason to choose endovascular over surgical treatment. If we are going to make a rational comparison to these two treatment strategies, we need to consider and take into account the standard complications reported in carotid surgery to allow us to evaluate both treatment strategies in an equal light. Outcome analysis in carotid reconstruction and carotid endovascular treatment is divided in the following way. The first consideration is perioperative stroke and this event is customarily stratified into major or minor stroke and likewise ipsilateral or allstroke. The second category to be considered is that of wound complications, meaning primarily hematoma or wound infection. The third category is nerve injury and the fourth category is medical complications, most commonly perioperative myocardial infarction. If we look at the surgical results from randomized cooperative trials, we see that the complication rates are gratifyingly quite low. In the ACAS trial, the perioperative stroke and death rate was 2.3%. In the VAAST trial, the 30-day mortality was 1.9%, all referable to myocardial infarction. The perioperative stroke rate was 2.4%. The nerve injury rate was 3.8% and the rate of nonfatal myocardial infarction was likewise 1.9%. In evaluating the results of symptomatic cooperative trials, we look first at the NASCET trial 9 where the 30-day mortality was 0.6%. The perioperative major stroke and death rate was 2.1%. The all-stroke rate was 5.5%. The nerve injury rate, evaluated by independent examiners, was 7.6%, the wound hematoma rate was 5.5% and the myocardial infarction rate was 0.9%. In the ECST symptomatic cooperative trial 7, it is very difficult to tease out the major stroke or death rate because the data is presented in several different manuscripts. However, it appears to be quite variable depending on the stratification of percent stenosis being considered. In the 0.9% surgery group, the major death or stroke rate was 2.3%. In the 30 – 49% group, it was 8%. In the 50 –69% group, it was 7.9% and in the 70 – 99% group, it was 3.7%. In the VA symptomatic surgery trial (VASST) 6, the 30-day mortality was 3.3%, the perioperative stroke rate was 2.2%, the
322
C.M. Loftus / International Congress Series 1247 (2002) 319–326
nerve injury rate was 5%, the wound hematoma rate was 5%, and the myocardial infarction rate was 2%. If we look at the results, cooperative trials customarily represent fixed observation points and evaluation by nonbiased observers. Randomized trial data is often felt to be the most valid and the highest level of evidence data that we have available to make therapeutic judgements. If we look at the published results of individual skilled carotid surgeons, we come up with data that sheds a bit more of a favorable light on carotid reconstruction. It is to be recalled, nonetheless, that essentially all the major cooperative trials demonstrated that carotid artery surgery was superior to medical therapy despite their higher complication rates as compared to the results of individual surgeons. Returning to the individual surgeon’s published results category, the data published by Spetzler et al. [10] for carotid microsurgery shows that in 200 cases the all-stroke rate was 1%, the mortality was 0.5%, the wound hematoma rate was 1.5% and the nerve injury rate was 5%. The data presented by Findlay [1] in carotid microsurgery document an all-stroke rate of 1.7%, a wound hematoma rate of 3.3% and a nerve injury rate of 5.1%. The data presented by Steiger et al. [11], again considering carotid microsurgery, document a perioperative stroke rate of zero, a myocardial infarction rate of 2%, a mortality of 2% and a nerve injury rate of 2%. The data presented by Harbaugh [2] employing regional anesthesia with loupemagnified carotid reconstruction shows an all-stroke rate of 2.5% with an ipsilateral rate of 1.3% and a major stroke rate of 0.9%. Harbaugh’s myocardial infarction rate is 1.3%. The wound hematoma rate is 2.1% and the nerve injury rate is 3%. Finally, my own data representing conventional loupe-magnified carotid reconstruction, under general anesthesia, with Hemashield patch angioplasty in the second half of the series, shows an all-stroke rate of 1.8% with a major stroke rate of 0.72%. My mortality rate is 0.72% and my wound hematoma rate is 0.4% [5]. If you look then at the data from leading surgeons published on an individual basis with their own observations of complications, we find that the all-stroke rate is customarily less than 2.6%, the nerve injury rate is less than 5.1%, the mortality is less than 2% and the wound hematoma rate varies between a low of 0.35% and a high of 3.35%. We must then ask ourselves the question which carotid lesions do endovascular specialists propose for treatment with stenting as an alternative to open surgical carotid reconstruction? The first manuscript we should consider is that by Theron et al. [12], stenting patients with primary stenosis. This paper discusses a protection scheme in about half the patients with distal balloon protection being employed during the stenting procedure. Two hundred fifty nine patients were treated, of which 136 underwent cerebral protection and 123 did not. In the unprotected cases the rate of carotid dissection was 5% and the rate of embolization was 8%. In the protected cases, no dissection is reported and the embolization rate is 1%. Some of these patients were not stented but merely had angioplasty. In these authors’ experience, stenting reduced the restenosis rate from 16% to 4% and the authors’ conclusion was that all patients undergoing endovascular carotid treatment should be stented and should be treated with cerebral protective strategies during the stenting process. The second endovascular manuscript to be considered is that of Yadav et al. [13], in which 126 arteries were treated, all stented. There were seven minor strokes, two major
C.M. Loftus / International Congress Series 1247 (2002) 319–326
323
strokes, and one death in this series. The stroke and death rate was 7.9% with an ipsilateral major stroke risk of 1.6%. The restenosis rate was quoted at 4.9%; all of these patients were asymptomatic. Yadav concluded that it was feasible to stent high-risk carotid patients who were not considered to be surgical candidates. Interesting data is reported by Lanzino et al. [4] regarding recurrent carotid stenosis and we point out that this data is in the neurosurgical literature. Lanzino et al. treated 25 arteries for recurrent stenosis with percutaneous angioplasty alone in 7 and placement of a stent in 18. There were no major neurological or cardiac events. Complications were limited to one TIA and one femoral pseudo aneurysm. Lanzino et al. demonstrated that 60% of angioplasty-alone patients restenosed and only one stent patient restenosed. Accordingly, their recommendation was first that the stenting procedure without protective strategies was safe for recurrent carotid stenosis and that stenting was far superior to angioplasty alone in the treatment of restenotic carotid disease. A contrary viewpoint is presented by Naylor et al. [8] in the United Kingdom randomized trial of angioplasty/stenting versus carotid surgery. In this trial, 23 patients were randomized who had greater than or equal to 70% carotid stenosis. Seventeen of these patients were treated, ten of whom had carotid surgery with no complications. Seven patients underwent either angioplasty or stent and five of these patients had a stroke. This trial was prematurely stopped because of an extremely negative outcome in the endovascular side and has been reported as strong evidence to suggest that carotid surgery is superior to endovascular carotid treatment. Non-atheromatous lesions such as fibromuscular dysplasia have also been studied with endovascular treatment. Hasso et al. [3] report three cases, all of which were symptomatic, which were treated with angioplasty alone in which no complications were evident. This is an extremely small series but it represents essentially the only experience in the literature regarding the endovascular treatment of fibromuscular dysplasia. Diethrich et al. discussed primary stenting for carotid surgery in 110 patients, treating 117 arteries. There were seven strokes (two major) for a stroke rate of 6.4%. Three patients were converted from endovascular treatment to open carotid artery surgery, with a rate of 2.7% for conversion. There were two deaths, one cardiac and one stroke, for a mortality of 1.8%. The 30-day angiographic success rate was 89.1% and there was an 89% patency rate at 7.6 months follow-up in this series. If endovascular treatment is to be considered, we should likewise ask the question ‘‘which lesions are there which the surgeon does not want to operate on?’’ This brings to my mind the central question of what constitutes a high-risk carotid operation. I would put high-risk patients in several categories. The first consists of patients who have predictably difficult operations. The second consists of patients who are at a high risk for medical complications. The third consists of patients who are at a high risk for intraoperative ischemia, and the fourth consists of patients who, for anatomical or other reasons, are at high risk for postoperative occlusion and/or stroke. Considering the first category, predictably difficult operations would include anatomical variants such as a medially directed internal carotid artery, a long and high carotid plaque, a high carotid bifurcation, or other anatomical anomalies such as cervical carotid aneurysm which would render an operation significantly more difficult. Predictably difficult operations also include carotid reoperation, operation upon carotids that have
324
C.M. Loftus / International Congress Series 1247 (2002) 319–326
been previously irradiated for other reasons, and cases in which distal exposure will be extremely high, in the region of C1 or C2 on a lateral angiogram. In the second category, patients at high risk for medical complication, I would include patients with recent myocardial infarction, patients with unstable angina, diabetic patients (who in my experience are at high risk for postoperative occlusion), patients with severe pulmonary disease rendering general anesthesia a major risk, and patients who require preoperative anticoagulation. In the third category, those at high risk for intraoperative ischemia, I would include patients with contralateral occlusion, although in my experience this does not appreciably increase the risk. Patients who will have difficult or impossible shunt placement have a high risk for intraoperative ischemia if a shunt is required. This includes patients who have a right angle bend distal to the carotid plaque and in whom this bend cannot be straightened out through operative exposure, or patients who have a large cervical aneurysm that precludes placement of a shunt. High risk for intraoperative ischemia also includes patients with difficult cervical carotid exposures, patients who are neurologically unstable with a fluctuating deficit, and patients who have a propagating thrombus in the cervical carotid artery distal to an area of high-grade stenosis but who continue to have antegrade flow beyond the region of the thrombus. Patients at high risk for postoperative occlusion and/or stroke include the following categories. Diabetic patients, in my experience, have a higher risk than normal for a postoperative occlusion, patients who experience postoperative hypotension seem to have a higher risk for carotid occlusion, patients in whom a repair is done high up in the internal carotid artery with consequent narrowing of the internal carotid are at high risk for occlusion and patients who are hypertensive who experience dysautoregulation type headache following reconstruction of a tight stenosis are at high risk for postoperative intracerebral hemorrhage. In my experience, most of these complications are avoided by the use of the Hemashield patch graft, which I now employ in every case. Let me point out then the cases which I take particular pleasure and pride in reconstructing surgically. These include primary stenosis, recurrent stenosis, irradiated carotids, cases with previous cervical irradiation or radical neck dissection, contralateral occlusion, high exposures, and perhaps the most enjoyable of all, cases which are turned down by other surgeons and referred for tertiary care center specialty attention. In my opinion, all of these cases can be satisfactorily addressed with reasonable surgical morbidity and mortality by an experienced carotid surgeon with a careful management approach. Cases I might send for stenting include the following. Cases of carotid dissection, which we previously treated with anticoagulation alone, may well be managed by a careful cannulation of the vessel and crossing of the lesion followed by placement of a stent. This, in my mind, represents a very significant advance over any surgical treatment or over anticoagulation alone as a therapeutic strategy. Cases with high cervical carotid aneurysms, beyond the reach of the surgeon, are well treated by stenting across the mouth of the lesion. Cases in which extreme high exposures are required at the base of the skull may well be better treated by stenting than by surgical exposure particularly when one considers the risk of cranial nerve injury. Patients who are medically unstable with recent
C.M. Loftus / International Congress Series 1247 (2002) 319–326
325
myocardial infarction are probably better off treated with endovascular treatment than with surgery, which entails an increased risk from general anesthesia. Finally, cases of fibromuscular dysplasia, which I customarily do not treat with surgery, may well be rational cases for an endovascular stenting approach. In conclusion, it is my opinion that the endovascular versus surgical debate for carotid artery therapy and/or asymptomatic patients will continue for certainly the next 5 –10 years, and most likely both treatments will ultimately enjoy some degree of acceptance. I would reiterate that at the present time only carotid artery surgery is proven by level one evidence to be superior to medical therapy and endovascular treatment stands where we stood with carotid artery surgery approximately 15 years ago, leaving us to accept only the results of individual practitioners rather than any randomized data. This will not forever be the case and randomized data will be available for surgery versus endovascular treatment in major centers where presumably the lowest morbidity and mortality can be obtained in each camp by skilled practitioners of the art. Trained, skilled carotid artery surgeons offer an excellent treatment for stroke prevention in symptomatic or asymptomatic patients. There is no documentation in literature for the superiority of endovascular treatment at the present time. However, I am quick to point out that selected cases, as I mentioned in the body of this work, may be more appropriately treated with endovascular strategies and I personally refer patients for such treatment under special circumstances as mentioned such as medically unstable patients, very high cervical carotid exposures, carotid dissection, and/or fibromuscular dysplasia. Individual referral patterns will depend on the surgeon’s comfort with high-risk patients. For surgeons with lesser carotid experience or whose morbidity and mortality data may not be quite satisfactory, endovascular referrals may be an appropriate strategy. For surgeons with a large degree of experience and a high degree of comfort in handling the challenge of complex carotid reconstruction cases, carotid surgery continues to represent the only therapy proven by randomized trials to be superior to medical treatment in specific and selected patients [6,7,9].
References [1] J.M. Findlay, Carotid microendarterectomy, Neurosurgery 32 (1993) 792 – 798. [2] R.E. Harbaugh, Carotid surgery under local anesthesia, Tech. Neurosurg. 3 (1997) 25 – 33. [3] A.N. Hasso, C.R. Bird, D.E. Zinke, J.R. Thompson, Fibromuscular dysplasia of the internal carotid artery: percutaneous transluminal angioplasty, Am. J. Radiol. 136 (1980) 955 – 960. [4] G. Lanzino, R.A. Mericle, D.K. Lopes, A.K. Wakhloo, L.R. Guterman, L.N. Hopkins, Percutaneous transluminal angioplasty and stent placement for recurrent carotid artery stenosis, J. Neurosurg. 90 (1999) 688 – 694. [5] C.M. Loftus, T.F. Kresowik, Anatomical basis and technique of carotid endarterectomy, in: C. Loftus, T. Kresowik (Eds.), Carotid Artery Surgery, Thieme Medical Publishers, New York, 2000. [6] M.R. Mayberg, E. Wilson, F. Yatsu, et al., Carotid endarterectomy and prevention of cerebral ischemia in symptomatic carotid stenosis, JAMA 266 (1991) 3289 – 3294. [7] MRC European Carotid Study Trialists, Interim results for symptomatic patients with severe (70 – 99%) or with mild (0 – 29%) carotid stenosis, Lancet 337 (1991) 1235 – 1243. [8] A.R. Naylor, A. Bolia, R.J. Abbott, et al., Randomized study of carotid angioplasty and stenting versus carotid endarterectomy: a stopped trial, J. Vasc. Surg. 28 (1998) 326 – 334.
326
C.M. Loftus / International Congress Series 1247 (2002) 319–326
[9] North American Symptomatic Carotid Endarterectomy Trial Collaborators, Beneficial effect of carotid endarterectomy in symptomatic patients with high grade stenosis, N. Engl. J. Med. 325 (1991) 445 – 453. [10] R.F. Spetzler, N. Martin, M.N. Hadley, R. Thompson, E. Wilkinson, P. Raudzens, Microsurgical endarterectomy under barbiturate protection: a prospective study, J. Neurosurg. 51 (1986) 147 – 150. [11] H. Steiger, L. Schaffler, S. Liechti, Results of microsurgical carotid endarterectomy, Acta Neurochir. 100 (1989) 31 – 38. [12] J.G. Theron, G.G. Payelle, O. Coskun, H.F. Huet, L. Guimaraens, Carotid artery stenosis: treatment with protected balloon angioplasty and stent placement, Radiology 201 (1996) 627 – 636. [13] J.S. Yadav, G.S. Roubin, S. Iyer, et al., Elective stenting of the extracranial carotid arteries, Circulation 95 (1997) 376 – 381.