From the New England Society for Vascular Surgery

Clinical results of carotid artery stenting compared with carotid endarterectomy

Soma Brahmanandam, MD, MPH,a,b Eric L. Ding, ScD,c,d Michael S. Conte, MD,a,b Michael Belkin, MD,a and Louis L. Nguyen, MD, MBA, MPH,a,b Boston, Mass
Objectives: Carotid artery stenting (CAS) is an alternative to carotid endarterectomy (CEA) for treating carotid artery stenosis. We conducted a systematic review and meta-analysis of the clinical trials to date comparing these two procedures to determine their relative safety and efcacy. Methods: Searches of the Cochrane Controlled Trials Register, MEDLINE, and EMBASE identied two cohort studies and eight randomized, controlled trials (RCTs) comparing CEA and CAS. Meta-analysis was performed for the primary outcome of 30-day stroke or death, using an intention-to-treat analysis. Between-trial heterogeneity was assessed using the 2 test, and xed-effects models were used to pool estimates in the absence of heterogeneity. Meta-regression was conducted to investigate potential effect differences by patient, intervention, and trial characteristics. To evaluate the effect of study design and inclusion criteria, sensitivity and subgroup analyses were performed. Results: Ten trials encompassing 3580 patients were analyzed. Patients who underwent CAS had a higher risk of 30-day stroke/death relative to patients who underwent CEA (risk ratio [RR], 1.30; 95% CI, 1.01-1.67). Meta-analysis and meta-regression demonstrated no between-trial heterogeneity. Sensitivity analysis of only RCTs showed similar higher risk for stroke/death (RR, 1.38; 95% CI, 1.06-1.79) in CAS patients. Subgroup analysis of trials enrolling only symptomatic patients showed higher risk of 30-day stroke/death (RR, 1.63; 95% CI, 1.18-2.25), but trials enrolling both symptomatic and asymptomatic patients showed no signicant differences (RR, 0.89; 95% CI, 0.59-1.35). Conclusions: Meta-analysis of trials to date shows CAS is associated with higher 30-day risk of stroke/death compared with CEA. Thus, for the patient at average surgical risk, the role of CAS is unproven, especially for symptomatic patients. And for the patient at high surgical risk, the role of any intervention is uncertain in the setting of competing comorbidities. The results of ongoing clinical trials in this area will likely provide additional evidence to support treatment choices for carotid artery stenosis. ( J Vasc Surg 2008;47:343-9.)

Large, randomized clinical trials from the early 1990s have shown the superiority of carotid endarterectomy (CEA) and aspirin therapy in preventing stroke compared with aspirin therapy alone for both symptomatic1-5 and asymptomatic6-8 carotid artery stenosis. On the basis of these trials, the most recent American Heart Association (AHA) guidelines recommend CEA for symptomatic patients with a stenosis of 50% to 99% if the perioperative risk of stroke or death is 6%.9,10 In asymptomatic patients, CEA is recommended for a stenosis of 60% 99% if the perioperative risk of stroke or death is 3%.9,10 Endovascular techniques for treating carotid artery stenosis have recently come to the forefront as an alternative to CEA, primarily for patients who are high operative risk. Multiple carotid artery stent (CAS) registries, both where enrollment was voluntary11-13 or standardized,9,14 have documented high levels of technical success with CAS. The
From the Division of Vascular & Endovascular Surgery,a Center for Surgery & Public Health,b Division of Preventive Medicine,c Brigham & Womens Hospital, Harvard Medical School; and Department of Epidemiology, Harvard School of Public Health.d Competition of interest: none. Presented at the Annual Meeting of the New England Society for Vascular Surgery, Ledyard, Conn, Oct 5-7, 2007. Reprint requests: Louis L. Nguyen, MD, MBA, MPH, Division of Vascular & Endovascular Surgery, and the Center for Surgery and Public Health, Brigham & Womens Hospital, 75 Francis St, Boston, MA 02115 (e-mail: llnguyen@partners.org). CME article 0741-5214/$34.00 Copyright 2008 by The Society for Vascular Surgery. doi:10.1016/j.jvs.2007.10.034

30-day end points from these registries showed that rates of stroke and death were 2.8% to 4.7%,11-13 and rates of stroke, death and myocardial infarction (MI) were 3.8% to 8.3%,14,15 which were comparable with rates of complications seen in the earlier trials for CEA.1-8 Carotid artery stenting has the added benet of being a less-invasive procedure, potentially minimizing the risks of wound complications and cranial nerve injury. In addition, this may translate to shorter length of hospitalization and less resource utilization. Given the potential safety advantages of CAS in high-risk surgical patients, it is important to determine if it is equivalent to the gold standard of CEA. Although several clinical trials have been conducted to compare these two procedures, there has been no clear consensus among them of whether one is superior or if they are equivalent. To further evaluate this, we conducted a systematic review and meta-analysis of the clinical trials to date comparing CAS and CEA. Our objectives were to determine the end points of 30-day any stroke or death, and 30-day any stroke as measures to compare the safety and efcacy between CAS and CEA. METHODS Searching. Searches were conducted using the Cochrane Controlled Trials Register, MEDLINE, and EMBASE to identify trials comparing CEA and CAS. Relevant clinical trials in the Cochrane Controlled Trials Register were identied with the key words carotid stenting. For both MEDLINE and EMBASE, a combination of key words and MeSH terms were used including carotid, endarterectomy, stenting, clin343

344 Brahmanandam et al

JOURNAL OF VASCULAR SURGERY February 2008

Table I. Study design

Trial Naylor, 199818 Brooks, 200119 Wallstent, 200120 Design Prospective, randomized, single-center trial Prospective, randomized, single-center trial Prospective, randomized, multicenter trial noninferiority trial Prospective, randomized, multicenter trial Prospective, single center trial Prospective, randomized, single-center trial Prospective, randomized, multicenter non-inferiority trial Prospective, nonrandomized, multicenter, equivalence trial Prospective, randomized, multicenter noninferiority trial Prospective, randomized, multicenter non-inferiority trial Allocation Sequentially numbered, sealed envelopes with random tx methods enclosed NK Computerized random number generator; assignment by sequentially numbered sealed envelopes Randomization by computer minimization algorithm; pts. signed consent forms Physicians decided which procedure pt. receives NK Randomization: pseudo-randomnumber generator; numbers distributed by an automated, centralized telephone device Based on physician and patient preference Randomization computer-generated; 8-patient block randomization design per center Randomization by computergenerated sequence Follow-up Duration 30 days 30 days 1 yr Trial Stopped Trial suspended: highly signicant risk of adverse outcome No Trial stopped early due to much higher rates of complications in CAS group; based on futility analysis No NA No Trial stopped early due to decreasing enrollment NA No

Cavatas, 200121 Madyoon, 200216 Brooks, 2004

22

1 yr NK 30 days 1 yr

Sapphire, 200423

Caress, 200517 Space, 200624

1 yr 30 days

Eva-3S, 200625

6 months

Trial stopped early due to safety; higher stroke rates in CAS group. After 30 d.

CARESS, Carotid Revascularization using Endarterectomy or Stenting Systems; CAVATAS, Carotid and Vertebral Artery Transluminal Angioplasty Study; EVA-3S, Endarterectomy Versus Angioplasty in Patients With Symptomatic Severe Carotid Stenosis; NA, not applicable; NK, not known; SAPPHIRE, Stenting and Angioplasty with Protection in Patients at High Risk for Endarterectomy; SPACE, Stent-Protected Percutaneous Angioplasty Versus Carotid Endarterectomy.

ical trial, and randomized, limited to humans. A few background articles and studies were also selected using handsearch methods. The search was conducted in February 2007, and 347 references were found: 86 clinical trials were identied through the Cochrane Controlled Trials Register, 147 citations through MEDLINE, and 114 citations through EMBASE. The rst round of eliminations was based on ve criteria: (1) articles were unrelated to the topic of interest, (2) review articles, (3) articles in languages other than English, (4) articles focused only on the CAS procedure, (5) articles duplicating those found in other databases. The second round of eliminations covered studies that only discussed the design of a trial or if a trial measured outcomes other than the ones of interest. Selection. We selected both randomized, controlled clinical trials (RCTs) and cohort studies that evaluated the outcomes of CAS as compared with CEA. These trials included all patients with symptomatic and asymptomatic internal carotid stenoses who underwent CEA or CAS, including with or without embolic protection device (EPD). The primary outcome of interest was to assess

postprocedural rates of any stroke and death 30 days of the intervention. The secondary outcome of interest was postprocedural rates of any stroke 30 days of the intervention. Any disputes were settled by discussion among the coauthors. Data abstraction. Relevant information was abstracted from the 10 selected studies. Study design information included the type of trial, duration of follow-up, randomization and allocation methods, and whether a trial was stopped (Table I). Patient characteristics consisted of inclusion criteria, baseline demographics (mean age, sex), and the numbers of patients in each treatment arm (based on an intention-to-treat analysis; Table II). Intervention characteristics included type of stent and whether an EPD was used for CAS (Table III). The 30-day end points of any stroke and any stroke/death were also extracted from all trials (Table IV). Statistical analysis. A meta-analysis was performed on two cohort studies16,17 and eight RCTs18-25 to evaluate the end points of 30-day stroke/death and 30-day stroke between CAS and CEA. Comparisons were made using risk ratios (RR), and all RCTs were analyzed using an intention-to-treat

JOURNAL OF VASCULAR SURGERY Volume 47, Number 2

Brahmanandam et al 345

Table II. Patient characteristics

Trial Naylor, 199818 Brooks, 200119 Wallstent, 200120 Cavatas, 200121 Madyoon, 200216 Brooks, 200422 Sapphire, 200423 Caress, 200517 Space, 200624 Eva-3S, 2006
25

Inclusion Criteria Sx: ipsilateral stenosis 70%-99% (by duplex) Sx: neuro events within 3 mo. of eval.; 70% stenosis Sx: TIA or stroke within 120 d. of randomization and 60% stenosis (by angio.) Pt. has stenosis that is suitable for either CEA or CAS by CTA/MRA/duplex; both asx & sx included Sx: 50% stenosis Asx 75% stenosis (by angio) Asx: 80% stenosis (by angio) Sx: 50% stenosis Asx: 80% stenosis (by duplex) Asx: 75% stenosis Sx: 50% stenosis (by duplex) Sx: sxs in prior 180 d. & ipsilateral stenosis 70% (by duplex) Sx: TIA /CVA in 120 d. & 60%-99% stenosis in symptomatic side (by duplex & MRA, or by angio

Treatment Arms (CEA/CAS total) 12/11 51/53 112/107 253/251 138/49 42/43 167/167 254/143 595/605 262/265 23 104 219 504 187 85 334 397 200 527

Percent Male 9 (52.9%) NK 140 (64.0%) 352 (69.8%) 187 (58.8%) NK 224 (67.0%) 247 (62.2%) 849 (71.6%) 189 (35.9)

Mean Age (yrs.) 71.3 68.0 68.3 67.0 73.3 68.3 72.6 71.3 67.9 69.7

Asx, Asymptomatic; CARESS, Carotid Revascularization using Endarterectomy or Stenting Systems; CAS, coronary artery stenting; CAVATAS, Carotid and Vertebral Artery Transluminal Angioplasty Study; CEA, carotid endarterectomy; CTA, computed tomography angiography; CVA, cerebrovascular accident; DUS, duplex ultrasound; EVA-3S, Endarterectomy Versus Angioplasty in Patients With Symptomatic Severe Carotid Stenosis; MRA, magnetic resonance angiography; NK, not known; SAPPHIRE, Stenting and Angioplasty with Protection in Patients at High Risk for Endarterectomy; SPACE, Stent-Protected Percutaneous Angioplasty Versus Carotid Endarterectomy; Sx, symptomatic; TIA, transient ischemic attack.

principle because the original numbers of patients allocated to treatment and control groups were available. Between-trial heterogeneity was assessed using the 2 test, and a xedeffects model was used to pool RRs in the absence of heterogeneity (P .05). To evaluate the effect of study design (RCT vs cohort), a sensitivity analysis was conducted which included RCTs only. As an additional check, meta-regression was also conducted for the end point of stroke/death to investigate for potential effect differences by patient, intervention, and trial characteristics. Variables included in the regression model were percentage of men, mean age, duration of postprocedural follow-up, type of stent deployed, whether a trial was stopped before completion, inclusion criteria (symptomatic patients only, asymptomatic patients only, or both symptomatic and asymptomatic patients), and if an EPD was used. An additional subgroup analysis compared both end points between trials that enrolled only symptomatic patients and trials that enrolled both symptomatic and asymptomatic patients. A subgroup analysis on asymptomatic patients alone could not be conducted because only one trial was dened by this inclusion criterion. Publication bias was evaluated for using the Begg test. A continuity correction was used to adjust for trials with sparse outcomes. An 0.05, corresponding to P .05, and 95% condence intervals (CI) were used dene statistical signicance. Analysis was performed using Stata 9 software (StataCorp, College Station, Tex). RESULTS Ten trials encompassing 3580 patients were analyzed. The meta-analysis showed that patients who underwent

CAS had a higher risk of 30-day stroke/death relative to patients who underwent CEA (RR, 1.30; 95% CI, 1.011.67; Fig 1) and that the 30-day risk of stroke was not signicantly different between patients who underwent CAS vs CEA (RR, 1.27; 95% CI, 0.96-1.69; Fig 2). Sensitivity analysis, which only included RCTs, corroborated that patients who underwent CAS had a higher risk of 30-day stroke/death relative to CEA patients (RR, 1.38; 95% CI, 1.06-1.79; Fig 3). In addition, it showed that CAS patients had a higher 30-day stroke risk compared with CEA patients (RR, 1.37; 95% CI, 1.02-1.84; Fig 4). Meta-analysis showed no evidence of between-trial heterogeneity in either the full analysis or RCT analysis for either end point. This nding was conrmed with metaregression, where important factors such as percentage of men, mean age, duration of follow-up, type of stent deployed, completion of trial, inclusion criteria, and use of an EPD were found not to contribute to between-trial heterogeneity. Subgroup analysis of trials enrolling only symptomatic patients showed a higher risk of 30-day stroke/death (RR, 1.63; 95% CI, 1.18-2.25; Fig 5) and 30-day stroke (RR, 1.62; 95% CI, 1.13-2.31; Fig 6) in CAS patients. Trials enrolling both symptomatic and asymptomatic patients, however, showed no signicant differences in 30-day stroke/death rates (RR, 0.89; 95% CI, 0.59-1.35) or 30day stroke rates (RR, 0.84; 95% CI, 0.53-1.35) between both procedures. There was no evidence of publication bias according to the Begg test (P .929). The corresponding funnel plot showed a symmetric distribution of individual trials, conrming the statistical analysis.

346 Brahmanandam et al

JOURNAL OF VASCULAR SURGERY February 2008

Table III. Intervention characteristics

Trial Naylor, 1998 Brooks, 200119 Wallstent, 200120 Cavatas, 200121 Madyoon, 200216 Brooks, 200422 Sapphire, 200418 Caress, 200517 Space, 200624 Eva-3S, 200625
18

Stent used Self-expanding Wallstent* Wallstent* Wallstent* Pre-1994: PTA only after 1994: Wallstent*, Streker, Palmaz Wallstent*, SMART Wallstent*, Dynalink Smart or Precise (self-expanding nitinol stent) Monorail Wallstent* Carotid Wallstent*; Precise ; Acculink 5 types mentioned (1, Carotid Wallstent Monorail*) None None None None

Embolic Protection Device

None None Angioguard or Angioguard XP Guardwire Plus PercuSurge Guardwire; Filterwire EX; Angioguard, Neuroshield , Carotid Trap 7 types mentioned (1: Guardwire Plus )* implemented for all pts. later in trial

CARESS, Carotid Revascularization using Endarterectomy or Stenting Systems; CAVATAS, Carotid and Vertebral Artery Transluminal Angioplasty Study; EVA-3S, Endarterectomy Versus Angioplasty in Patients With Symptomatic Severe Carotid Stenosis; SAPPHIRE, Stenting and Angioplasty with Protection in Patients at High Risk for Endarterectomy; SPACE, Stent-Protected Percutaneous Angioplasty Versus Carotid Endarterectomy. *Wallastent (Boston Scientic Corporation, Natick, MA, USA) Streker (Meditech, Brooklyn, NY, USA). Palmaz (Johnson & Johnson, New Brunswick, NJ, USA). SMART (Cordis Corporation, Miami Lakes, FL, USA). Dynalink (Guidant Corporation, Indianapolis, IN, USA). Precise (Cordis Corporation, Miami Lakes, FL, USA). Acculink (Guidant, Santa Clara, CA, USA). Angioguard & Angioguard XP (Cordis Corporation, Miami Lakes, FL, USA). Guardwire Plus (Medtronic Vascular, Santa Rosa, CA, USA). PercuSurge Guardwire (Medtronic, Minneapolis, MN). Filterwire EX (Boston Scientic Corporation, Natick, MA, USA). Angioguard Neuroshield (MedNova, Horsham, UK). Carotid Trap (Microvena, White Bear Lake, MN, USA).

DISCUSSION Meta-analysis. Until recently, results from completed CAS registries11-15 had been the strongest and most abundant evidence available in favor of CAS as an alternate treatment for carotid artery stenosis. As mentioned previously, these observational studies demonstrated a high level of technical success and 30-day end points of stroke/death and stroke/death/MI of 2.8% to 8.3%.11-15 Although informative, these results have to be interpreted with caution given that the subjects for these registries were not randomized, and the studies may suffer from inclusion bias for lower-risk patients, which would have articially decreased the complication rate for CAS. More recent data from a meta-analysis of six RCTs by Coward et al26 concluded that CAS was equivalent to CEA. The meta-analysis for our study, however, found that 30-day stroke/death across all studies was slightly higher after CAS than after CEA. For the end point of 30-day stroke, the two procedures were found to be statistically equivalent, though trending toward being higher after CAS. The addition of the results from the two latest RCTs, Stent-Protected Angioplasty versus Carotid Endarterectomy (SPACE)24 and Endarterectomy Versus Angioplasty in Patients With Symptomatic Severe Carotid Stenosis (EVA-3S),25 partly explains why our results differ from those of the previous meta-analysis. The SPACE study showed that CEA trended towards better outcomes for

both stroke/death and stroke, though this difference was not of statistical signicance between procedures. In the EVA-3S trial, CAS patients had signicantly higher rates of both outcomes relative to CEA patients, resulting in early termination of the trial. Arguably, improvements in technology, technique, experience, and the use of EPD over time may give greater signicance to more recent trials. However, analysis of the trial results in chronologic order does not indicate a time-related trend for outcomes. In fact, the latest trials have demonstrated poorer outcomes for CAS, indicating that technology and experience alone may not be a determinant for the safety and efcacy of CAS. Although our primary analysis included RCTs and two cohort studies, we conducted a sensitivity analysis for RCTs alone to test if study design affected outcome. The results conrmed that 30-day stroke/death was higher in CAS patients and showed that 30-day stroke was also higher in CAS patients. The magnitudes of the RRs were also slightly larger for both end points in this analysis, indicating that the cohort studies may have diminished the effect of the original estimates in the pooled analysis. Because subjects were not randomized to CAS and CEA in the cohort studies, a greater number of low-risk patients may have selected CAS, improving the results in this group and lowering the overall estimate for both end points. A re-

JOURNAL OF VASCULAR SURGERY Volume 47, Number 2

Brahmanandam et al 347

Table IV. Clinical outcomes at 30 days

30 d any stroke Study Naylor, 199818 Brooks, 200119 Wallstent, 200120 Cavatas, 200121 Madyoon, 200216 Brooks, 200422 Sapphire, 200423 Caress, 200517 Space, 200624 Eva-3S, 200625 CEA 0/12 (0%) 0/51 (0%) NK 21/253 (8.3%) 8/138 (5.8%) 0/42 (0%) 5/167 (3.0%) 9/254 (3.6%) 36/595 (6%) 9/262 (3.4%) CAS 5/11 (45.5%) 0/53 (0%) NK 18/251 (7.2%) 2/49 (4.1%) 0/43 (0%) 6/167 (3.65) 3/143 (2.1%) 45/605 (7.5%) 24/265 (9.1%) 30 d any stroke/death CEA 0/12 (0%) 1/51 (2.0%) 5/112 (4.5%) 25/253 (9.9%) 8/138 (5.8%) 0/42 (0%) 9/167 (5.4%) 9/254 (3.6%) 38/595 (6.5%) 10/262 (3.8%) CAS 5/11 (45.5%) 0/53 (0%) 13/107 (12.1%) 25/251 (10.0%) 2/49 (4.1%) 0/43 (0%) 8/167 (4.8%) 3/143 (2.1%) 46/605 (7.7%) 25/265 (9.4%)

Fig 2. Thirty-day stroke in all trials. CI, Condence interval.

CARESS, Carotid Revascularization using Endarterectomy or Stenting Systems; CAS, coronary artery stenting; CAVATAS, Carotid and Vertebral Artery Transluminal Angioplasty Study; CEA, carotid endarterectomy; EVA-3S, Endarterectomy Versus Angioplasty in Patients With Symptomatic Severe Carotid Stenosis; SAPPHIRE, Stenting and Angioplasty with Protection in Patients at High Risk for Endarterectomy; SPACE, Stent-Protected Percutaneous Angioplasty Versus Carotid Endarterectomy.

Fig 3. Thirty-day stroke/death in randomized, controlled trials only. CI, Condence interval.

Fig 1. Thirty-day stroke/death in all trials. CI, Condence interval.

cently published meta-analysis using only RCTs also showed increased risk of 30-day death/stroke.27 Heterogeneity. The 2 test in meta-analysis showed no evidence of heterogeneity in all trials or in RCTs alone. Despite these results, potential effect differences by clinically relevant factors were further investigated to ensure a thorough analysis of heterogeneity. Meta-regression corroborated the absence of between-trial heterogeneity due to sex, mean age, duration of follow-up, type of stent

deployed, completion of trial, inclusion criteria, or use of an EPD. These results demonstrate that our pooled estimate from the meta-analysis is fairly reliable because all trials were statistically similar across patient, trial, and intervention characteristics. Subgroup analysis. According to guidelines established by the AHA/ASA (American Stroke Association), symptomatic patients who have a low operative risk with either moderate (50%-69%) or severe (70%-99%) stenosis are recommended to undergo CEA, whereas patients who are symptomatic with severe stenosis ( 70%) and who are high risk for CEA are recommended to undergo CAS.10 The ndings from our subgroup analysis of trials enrolling only symptomatic patients, with a threshold stenosis of 60%, were not consistent with the AHA/ASA recommendations. We found that the 30-day risks for both any stroke/death and stroke were signicantly higher in CAS patients, making it a suboptimal choice for symptomatic patients with moderate to severe stenosis. Many of the trials included in this meta-analysis were not performed on pa-

348 Brahmanandam et al

JOURNAL OF VASCULAR SURGERY February 2008

Fig 4. Thirty-day stroke in randomized, controlled trials only. CI, Condence interval.

Fig 6. Thirty-day stroke in trials with only symptomatic patients.

Fig 5. Thirty-day stroke/death in trials with only symptomatic patients. CI, Condence interval.

tients with high surgical risk, however; therefore, we cannot conclude whether CAS or CEA is better for the truly high-risk surgical patient. Guidelines for management of asymptomatic carotid stenosis, on the contrary, are currently not well dened. For the asymptomatic patient who has a low operative risk, two areas of controversy exist: rst, is intervention needed in these patients; and second, if recommending intervention, what should be the threshold stenosis to intervene?15 The AHA recommends CEA for these patients with stenosis of 60% to 99%, and a perioperative risk of stroke or death that is 3%.9,10 The role for CAS in the patient with asymptomatic severe carotid stenosis is less clear. Although it would seem to be an appropriate intervention for patients who have high operative risk, no level I data are currently available to support this. Our second subgroup analysis looked at trials enrolling both symptomatic and asymptomatic patients. Although isolated analysis of asymptomatic trials would have been more helpful, this was not possible owing to the small

number of trials with this inclusion criterion. Our analysis of these combined trials showed that CAS trended toward a better outcome for both end points (RR 1), but this difference was not statistically signicant. Because results from (only) symptomatic patients statistically favor CEA over CAS, it is tempting to speculate that analysis of only asymptomatic patients would favor CAS. Limitations. Several potential limitations are inherent in an analysis of this nature. Although meta-analysis and meta-regression showed no evidence of between-trial heterogeneity, there may have been other potential sources of heterogeneity that could not be measured. One example was the disparity between trials in criteria used to assess the qualications of an interventionalist. Certain trials provided clear denitions of their criteria, but others were not as explicit; therefore, we could not assess if the differential training between interventionalists was a true source of heterogeneity. This ultimately affects our condence in the pooled estimate from meta-analysis. A second limitation was that most trials did not stratify their patients by level of operative risk. As mentioned previously, the AHA guidelines for CEA are based on three criteria: whether the patient is symptomatic or asymptomatic, whether the patient is considered high or low risk for an operation, and the degree of stenosis. Although we could evaluate the outcomes of CAS in symptomatic patients, we did not have information about whether they were considered high or low risk for surgery. Furthermore, there is no widely accepted consensus on the denition of high risk in CEA patients other than anatomic high risk. Because patients in RCTs must be eligible to receive both CEA and CAS, we can assume that these patients do not have a prohibitive risk for CEA. Nevertheless, operative risk has gradation, and we believe that more efforts to dene and characterize operative risk in future trials may help answer the question of whether CAS is benecial in truly high risk CEA patients. Finally, this meta-analysis builds on the ndings of previous analyses by including two contemporary RCTs. And although meta-analyses can be a useful tool in summa-

JOURNAL OF VASCULAR SURGERY Volume 47, Number 2

Brahmanandam et al 349

rizing the results of several trials, the meta-analysis can only reect the currently available data. With the completion of ongoing and potentially future trials, the conclusions about CAS vs CEA in the treatment of cerebrovascular disease may be changing. CONCLUSIONS Meta-analysis of trials to date shows CAS is associated with a higher 30-day risk of stroke/death compared with CEA. Thus, the role of CAS is unproven for the patient at average surgical risk, especially for symptomatic patients. And for the patient at high surgical risk, the role of any intervention is uncertain in the setting of competing comorbidities. The results of ongoing clinical trials in this area will likely provide additional evidence to support treatment choices for carotid artery stenosis. AUTHOR CONTRIBUTIONS Conception and design: SB, ED, LN Analysis and interpretation: SB, ED, MB, MC, LN Data collection: SB, ED, LN Writing the article: SB, LN Critical revision of the article: SB, ED, MB, MC, LN Final approval of the article: SB, ED, MB, MC, LN Statistical analysis: SB, ED, LN Obtained funding: MB, LN Overall responsibility: SB, LN
REFERENCES
1. Benecial effect of carotid endarterectomy in symptomatic patients with high-grade carotid stenosis. North American Symptomatic Carotid Endarterectomy Trial Collaborators. N Engl J Med 1991;325:445-53. 2. Barnett HJ, Taylor DW, Eliasziw M, Fox AJ, Ferguson GG, Haynes RB, et al. Benet of carotid endarterectomy in patients with symptomatic moderate or severe stenosis. North American Symptomatic Carotid Endarterectomy Trial Collaborators. N Engl J Med 1998; 339:1415-25. 3. Randomised trial of endarterectomy for recently symptomatic carotid stenosis: nal results of the MRC European Carotid Surgery Trial (ECST). Lancet 1998;351:1379-87. 4. MRC European Carotid Surgery Trial: interim results for symptomatic patients with severe (70-99%) or with mild (0-29%) carotid stenosis. European Carotid Surgery Trialists Collaborative Group. Lancet 1991; 337:1235-43. 5. Mayberg MR, Wilson SE, Yatsu F, Weiss DG, Messina L, Hershey LA, et al. Carotid endarterectomy and prevention of cerebral ischemia in symptomatic carotid stenosis. Veterans Affairs Cooperative Studies Program 309 Trialist Group. JAMA 1991;266:3289-94. 6. Hobson RW 2nd, Weiss DG, Fields WS, Goldstone J, Moore WS, Towne JB, et al. Efcacy of carotid endarterectomy for asymptomatic carotid stenosis. The Veterans Affairs Cooperative Study Group. N Engl J Med 1993;328:221-7. 7. Endarterectomy for asymptomatic carotid artery stenosis. Executive Committee for the Asymptomatic Carotid Atherosclerosis Study. Jama 1995;273:1421-8. 8. Halliday A, Manseld A, Marro J, Peto C, Peto R, Potter J, et al. Prevention of disabling and fatal strokes by successful carotid endarterectomy in patients without recent neurological symptoms: randomised controlled trial. Lancet 2004;363:1491-502. 9. Biller J, Feinberg WM, Castaldo JE, Whittemore AD, Harbaugh RE, Dempsey RJ, et al. Guidelines for carotid endarterectomy: a statement for healthcare professionals from a Special Writing Group of the Stroke Council, American Heart Association. Circulation 1998;97:501-9.

10. Sacco RL, Adams R, Albers G, Alberts MJ, Benavente O, Furie K, et al. Guidelines for prevention of stroke in patients with ischemic stroke or transient ischemic attack: a statement for healthcare professionals from the American Heart Association/American Stroke Association Council on Stroke: co-sponsored by the Council on Cardiovascular Radiology and Intervention: the American Academy of Neurology afrms the value of this guideline. Circulation 2006;113:e409-49. 11. Wholey MH, Al-Mubarek N, Wholey MH. Updated review of the global carotid artery stent registry. Catheter Cardiovasc Interv 2003; 60:259-66. 12. Theiss W, Hermanek P, Mathias K, Ahmadi R, Heuser L, Hoffmann FJ, et al. Pro-CAS: a prospective registry of carotid angioplasty and stenting. Stroke 2004;35:2134-9. 13. Bosiers M, Peeters P, Deloose K, Verbist J, Sievert H, Sugita J, et al. Does carotid artery stenting work on the long run: 5-year results in high-volume centers (ELOCAS Registry). J Cardiovasc Surg (Torino) 2005;46:241-7. 14. Gray WA, Hopkins LN, Yadav S, Davis T, Wholey M, Atkinson R, et al. Protected carotid stenting in high-surgical-risk patients: the ARCHeR results. J Vasc Surg 2006;44:258-68. 15. Bates ER, Babb JD, Casey DE Jr, Cates CU, Duckwiler GR, Feldman TE, et al. ACCF/SCAI/SVMB/SIR/ASITN 2007 clinical expert consensus document on carotid stenting: a report of the American College of Cardiology Foundation Task Force on Clinical Expert Consensus Documents (ACCF/SCAI/SVMB/SIR/ASITN Clinical Expert Consensus Document Committee on Carotid Stenting). J Am Coll Cardiol 2007;49:126-70. 16. Madyoon H, Braunstein E, Callcott F, Oshtory M, Gurnsey L, Croushore L, et al. Unprotected carotid artery stenting compared to carotid endarterectomy in a community setting. J Endovasc Ther 2002; 9:803-9. 17. Carotid Revascularization Using Endarterectomy or Stenting Systems (CaRESS) phase I clinical trial: 1-year results. J Vasc Surg 2005;42: 213-9. 18. Naylor AR, Bolia A, Abbott RJ, Pye IF, Smith J, Lennard N, et al. Randomized study of carotid angioplasty and stenting versus carotid endarterectomy: a stopped trial. J Vasc Surg 1998;28:326-34. 19. Brooks WH, McClure RR, Jones MR, Coleman TC, Breathitt L. Carotid angioplasty and stenting versus carotid endarterectomy: randomized trial in a community hospital. J Am Coll Cardiol 2001;38: 1589-95. 20. Alberts MJ. Results of a multicenter prospective randomized trial of carotid artery stenting vs carotid endarterectomy. Stroke 2001;32:325. 21. Endovascular versus surgical treatment in patients with carotid stenosis in the Carotid and Vertebral Artery Transluminal Angioplasty Study (CAVATAS): a randomised trial. Lancet 2001;357:1729-37. 22. Brooks WH, McClure RR, Jones MR, Coleman TL, Breathitt L. Carotid angioplasty and stenting versus carotid endarterectomy for treatment of asymptomatic carotid stenosis: a randomized trial in a community hospital. Neurosurgery 2004;54:318-24; discussion 324-5. 23. Yadav JS, Wholey MH, Kuntz RE, Fayad P, Katzen BT, Mishkel GJ, et al. Protected carotid-artery stenting versus endarterectomy in high-risk patients. N Engl J Med 2004;351:1493-501. 24. SPACE Collaborative Group, Ringleb PA, Allenberg J, Brckmann H, Eckstein HH, Fraedrich G, et al. 30 day results from the SPACE trial of stent-protected angioplasty versus carotid endarterectomy in symptomatic patients: a randomised non-inferiority trial. Lancet 2006;368: 1239-47. 25. Mas JL, Chatellier G, Beyssen B, Branchereau A, Moulin T, Becquemin JP, et al. Endarterectomy versus stenting in patients with symptomatic severe carotid stenosis. N Engl J Med 2006;355:1660-71. 26. Coward LJ, Featherstone RL, Brown MM. Safety and efcacy of endovascular treatment of carotid artery stenosis compared with carotid endarterectomy: a Cochrane systematic review of the randomized evidence. Stroke 2005;36:905-11. 27. Luebke T, Aleksic M, Brunkwall J. Meta-analysis of Randomized Trials Comparing Carotid Endarterectomy and Endovascular Treatment. Eur J Vasc Endovasc Surg 2007;34:470-9. Submitted Aug 26, 2007; accepted Oct 20, 2007.