- Email: [email protected]
Restenosis after eversion vs patch closure carotid endarterectomy
From the New England Society for Vascular Surgery
Restenosis after eversion vs patch closure carotid endarterectomy Robert S. Crawford, MD, Thomas K. Chung, MA, Thomas Hodgman, BA, Juan D. Pedraza, MD, Michael Corey, BA, and Richard P. Cambria, MD, Boston, Mass Objectives: Recurrent stenosis after carotid endarterectomy (CEA), previously reported to occur in 1%/year after operation, is the finite limitation of CEA. Eversion endarterectomy has a perceived lower incidence of recurrent stenosis, although data to support this contention are conflicting. The goal of the present study was to compare the late anatomic results of patch closure (PC) vs eversion CEA. Methods: Between January 1, 1995 and June 30, 2005, 950 CEA were performed by the senior author with adoption of eversion (EV) as the primary technique as of January 1, 2001. With minimum of 1-year follow-up by study inclusion criteria, complete follow-up data (including a duplex scan) was available for 155 PC and 135 EV patients. Incidence of moderate (50% to 70%) and severe (>70%) restenosis was examined at <2 months and >1 year after operation. Study end-points included late stroke, survival, and freedom from restenosis (moderate and severe) and were assessed by actuarial methods. Results: There were no differences in relevant demographic/clinical parameters, indication for surgery (69% overall asymptomatic) or early perioperative stroke/death (1.1% overall; P ⴝ .25) between PC and EV. After correction for different mean follow-up intervals (PC ⴝ 5.5 years vs EV ⴝ 3.5 years) by actuarial methods, there was no significant difference in late moderate (P ⴝ .91) or severe (P ⴝ .54) recurrent stenosis between PC and EV. In the group of patients with at least 1-year follow-up, 11/290 (3.8%) patients (4/135 EV, 7/155 PC; P ⴝ .39) required reintervention on their operated carotid artery at a cumulative follow-up interval of 4.5 years. Three strokes (3/290; 1.1%) occurred during late follow-up, all in the PC group, with only one related to the operated carotid artery. Late survival was similar between EV and PC, (P ⴝ .86). Female gender (odds ratio [OR] 3.72[1.02-13.5], P ⴝ .046) was associated with severe restenosis irrespective of surgical technique. Univariate analysis also showed that female gender (OR 7.6[CI: 0.88-66.7], P ⴝ .042) was associated with late stroke. Conclusion: These findings indicate that restenosis rates are similar between eversion and patch CEA and likely represent biological remodeling phenomenon rather than technical variations of operations. While EV offers distinct advantages in certain anatomic circumstances, adoption of EV with the hope of decreasing restenosis is not warranted. ( J Vasc Surg 2007;46:41-8.)
Randomized trials have validated the efficacy of carotid endarterectomy (CEA) for stroke prevention in both symptomatic and asymptomatic significant stenosis of the internal carotid artery.1-4 Perioperative stroke rates of 6% in symptomatic patients and 3% in asymptomatic patients are acceptable thresholds according to published guidelines.5 However, perioperative stroke rates of 2% for symptomatic disease and 1% for asymptomatic disease are reported in multiple contemporary surgical series.6-9 CEA is most often performed with a longitudinal arteriotomy typically closed with a patch, as randomized trials demonstrate that relevant outcome measures favor patch over primary closure.10-13 An alternative approach, eversion CEA, was originally described by DeBakey et al in the 1950s14 and involves the oblique transection of the internal carotid artery at its From the Division of Vascular and Endovascular Surgery of the General Surgical Services, Massachusetts General Hospital and Harvard Medical School. Competition of interest: none. Supported in part by the Harold & June Geneen Vascular Research Fund. Presented at the Thirty-third Annual Meeting of the New England Society for Vascular Surgery, Sep 22-24, 2006, Boston, Mass. Reprint requests: Richard P. Cambria, MD, WACC 448, 15 Parkman Street, Boston, MA 02114 (email: [email protected]) 0741-5214/$32.00 Copyright © 2007 by The Society for Vascular Surgery. doi:10.1016/j.jvs.2007.02.055
origin in the carotid bulb, followed by removal of the plaque using the eversion maneuver. The purported or demonstrated advantages of eversion CEA include decreased operative time, avoidance of prosthetic material, facilitating reconstruction of a kinked or coiled internal carotid artery (ICA), and decreased rates of restenosis.15,16 Indeed, recurrent stenosis after CEA, which occurs in 1%/year in our experience,17 is the limitation of CEA, and efforts to minimize it with technical variations are worthy of investigation. Largely as a function of this consideration, the senior author (R.P.C.) adopted eversion as the primary reconstruction mode of CEA in January of 2001. The goal of the present study was to compare the late anatomic results of patch closure vs eversion CEA with the inherent control of all operations being performed by a single surgeon. PATIENTS AND METHODS During the interval January 1, 1995 to June 30, 2005 (to insure minimum 1-year follow-up), 950 CEAs were performed. Procedures performed using primary closure, those performed in conjunction with other surgical procedures such as carotid artery bypass graft, valve surgery, or coronary bypass surgery were excluded from the analysis. After such exclusions, we examined the perioperative out41
JOURNAL OF VASCULAR SURGERY July 2007
42 Crawford et al
come of 636 CEAs (“intention to treat” cohort: patch, n ⫽ 412, eversion, n ⫽ 224) for perioperative death, strokes, and myocardial infarctions. Of these 636, we identified 290 CEAs, 155 done with patch endarterectomy technique (PC) and 135 done using the eversion technique (EV) with at least 1-year duplex ultrasound follow-up. This study was approved by the Institutional Review Board of the Massachusetts General Hospital. Clinical and demographic data were obtained from hospital records and office charts. Additional telephone contact was sought for those patients for whom recent clinical data were not available. Mortality data were obtained using hospital records as well as data from the National Death Index in Bethesda, Maryland. Perioperative complications were obtained from review of prospectively maintained weekly morbidity and mortality statistics reported annually in quality assurance reports. Office records were also quite detailed in this regard. Definitions for clinical variables are as follows: hypertension (HTN), use of anti-hypertensive drugs or BP 165/95 or greater on two separate occasions; renal failure was defined as those patients undergoing dialysis or with serum creatinine ⱖ1.5 mg/dl; diabetes (DM), fasting blood sugar ⬎220 mg/dl or greater or on DM meds; hypercholesterolemia (HC), total cholesterol level of ⬎220 mg/dl or greater or currently on hypercholesterolemia medications; peripheral vascular disease (PVD), intermittent claudication in one or both limbs or prior surgical/ endovascular treatment for PVD, ischemic rest pain, or gangrene of foot; coronary artery disease (CAD), history of myocardial infarction (MI), angina or ventricular ectopy, chronic heart failure or recent angina; atrial fibrillation (AF), diagnosed by history or electrocardiogram. Ipsilateral symptoms were those occurring in the same vascular territory as the operated carotid artery. Primary end-points of the study included restenosis of the operated carotid artery, perioperative and late stroke and death. Restenosis was examined both in the “early” (within 2 months of operation) and “late” (long-term follow-up) time points. If, however, severe stenosis was seen within 2 months, such patients were considered technical failures ie, persistent obstruction. “Overall” restenosis refers to the combined “early” and “late” follow-up components. Moderate (50% to 70% reduction in vessel diameter) restenosis was defined by the following duplex criteria: internal carotid artery (ICA) peak systolic velocity (PSV) ⬎125 cm/sec, ICA/CCA ratio ⬎2 or end-diastolic velocity (EDV) ⬎100 cm/sec and criteria for severe (⬎70% reduction in vessel diameter) restenosis with: (ICA-PSV ⬎250, ICA/CCA ratio ⬎4, or EDV ⬎140).17,18 Duplex exams were typically obtained in the immediate postoperative (within 2 months) period and at ⬎1 year. Follow-up duplex exams (ie, those not conducted in our vascular laboratory) were required to have actual velocity data to be included in the group with “adequate” ultrasound followup. Perioperative and late stroke was defined as a persistent neurological deficit lasting ⬎24 hours and generally confirmed by brain imaging. Freedom from restenosis and death were analyzed using Kaplan-Meier life table analysis.
The Rankin Score for functional assessment after stroke is defined elsewhere.19 Secondary end-points included myocardial infarction (MI), redo CEA, operative site hematomas (not necessarily requiring operation), perioperative transient ischemic attacks or amaurosis fugax (TIA/AFX), cranial nerve injury (persistent deficit at 1 month after surgery), and retinal infarcts. MI was confirmed by positive laboratory enzyme and electrocardiograph (ECG) analysis. Enzyme analysis was obtained only after routine postoperative ECG analysis showed new abnormalities. TIA/AFX was defined as an acute disturbance of focal neurological or monocular function with symptoms lasting ⬍24 hours. Retinal infarct was defined as total or partial monocular loss of vision with symptoms persisting longer than 24 hours, with diagnosis confirmed by an ophthalmologist. Procedures. All procedures were performed generally with patch carotid endarterectomy until the beginning of 2001 when preferentially the eversion technique was used. Operative procedures were most often performed under general anesthesia with electroencephalographic monitoring and selective shunting. Electroencephalogram (EEG) changes dictated a preference towards longitudinal arteriotomy and patch to facilitate shunting. A description of the eversion technique has been published elsewhere.15 Overall, regional anesthesia was performed in seven (2.4%) patients (2/135 for EV, 5/155 for PC). Intraprocedural verification of the technical result of operation is limited to a qualitative hand-held Doppler interrogation. Intraoperative shunting was used in 41 (14.1%) of patients, all in the patch group. EEG slowing was observed in 25 (8.6%) of patients, 22 (14.2%) of them in the patch group, and three (2.2%) in the eversion group. Overall, 30.3 % of patients were operated on for symptomatic disease (see Table I), 29.7% of patients in the patch group and 31.0% of patients for the eversion group. For redo CEA, only preocclusive (⬎90%; residual lumen diameter ⱕ1.5 mm) lesions verified by angiography as critical (in earlier years), or duplex and CTA in more recent years underwent reintervention. Two patients treated for restenosis with carotid stents also had angiography. Statistical methods. Cohort comparisons of demographic data were assessed using two-tailed t-tests for continuous variables, 2 and Fischer exact t tests for categorical data. Late outcomes were assessed using Kaplan-Meier life table analysis and the log-rank test was used when comparing subgroups across comparable time intervals. Univariate analysis of variables potentially associated with study endpoints was performed, and those that were statistically significant (threshold of 0.05) were included in a multiple logistic regression in order to identify predictors of late outcomes. RESULTS The patch group (n ⫽ 155) differed from the eversion (n ⫽ 135) group in four clinical domains as displayed in Table II. The higher incidence of “hypercholesterolemia”
JOURNAL OF VASCULAR SURGERY Volume 46, Number 1
Crawford et al 43
Table I. Preoperative characteristics of patients with complete anatomic follow-up Clinical factor
Patch (n ⫽ 155)
Eversion (n ⫽ 135)
P-value
Symptomatic* TIA/AFX Stroke Retinal Stroke Vertebrobasilar Symptoms Preoperative ultrasound ICA Velocity (cm/sec) ICA/CCA
29.7% (46/155) 22.6% (35/155) 7.1% (11/155) 1.3% (2/155) 1.3% (2/155)
31.0% (42/135) 20.0% (27/135) 11.9% (16/135) 5.2% (7/135) 3.7% (5/135)
.70 .67 .22 .09† .26†
412 ⫾ 131 (13-766) 5.61 ⫾ 2.43 (0.15-16.38)
419 ⫾ 117 (160-730) 5.74 ⫾ 1.98 (1.62-12.2)
.68 .66
TIA/AFX, transient ischemic attacks or amaurosis fugax; ICA, internal carotid artery; ICA/CCA, peak systolic velocity ratio of internal to common carotid artery. *Includes patients with multiple types of symptoms † Fisher exact t test shown
Table II. Demographic and clinical features of patients with complete anatomic follow-up
Age (years) Gender male Smoking HTN Renal failure DM Side (right) Hypercholesterolemia PVD CAD Atrial fibrillation Cholesterol Triglycerides LDL HDL
Patch cohort (n ⫽ 155)
Eversion cohort (n ⫽ 135)
P-value
73.4 ⫾ 10.3 (36-95) 57.4% (89/155) 74.4% (90/121) 85.1% (131/154) 18.7% (29/155) 24.7% (38/154) 54.8% (85/155) 71.4% (110/154) 29.2% (45/154) 69.5% (107/154) 16.9% (26/154) 184 ⫾ 54.1 (0-446) 192 ⫾ 174 (52-1623) 107 ⫾ 37.9 (20-186) 42.3 ⫾ 13.2 (9-84)
74.1 ⫾ 9.7 (41-94) 62.7% (84/134) 74.6% (97/130) 93.3% (126/135) 9.6% (13/135) 23.0% (31/135) 51.1% (69/135) 85.9% (113/135) 25.9% (35/135) 52.6% (70/132) 17.8% (24/135) 178 ⫾ 62.7 (0-303) 174 ⫾ 169 (0-1167) 99 ⫾ 51.2 (0-218) 47.0 ⫾ 24.0 (0-182)
.58 .36 .97 .03 .03 .73 .53 .003 .53 .003 .84 .40 .48 .26 .1
HTN, hypertension; DM, diabetes; PVD, peripheral vascular disease; CAD, coronary artery disease; LDL, low density lipoproteins; HDL, high density lipoproteins.
Table III. Perioperative complications for patients with complete anatomic follow-up
Hematoma MI Stroke CN injury TIA
Patch (n ⫽ 155)
Eversion (n ⫽ 135)
P-value
1.9% (3/155) 0.7% (1/155) 1.9% (3/155) 0% (0/155) 1.3% (2/155)
3.0% (4/135) 1.5% (1/135) 0% (0/135) 0.8% (1/135) 2.2% (3/135)
.71* .59* .25* .47* .66*
MI, myocardial infarction; CN, cranial nerve; TIA, transient ischemic attacks. *Fisher exact t test shown.
in the eversion group likely reflects the more aggressive use of HMG-CoA reductase inhibitors since 2001. Perioperative stroke/death. Combined perioperative stroke/death occurred in 10/636 (1.6%) patients in the intention-to-treat cohort. Strokes occurred in 7/636 (1.1%) patients (5/412 (1.2%) in PC and 2/224 (0.9%) in EV; P ⫽ 1.0). Across all CEAs performed during this timeframe, perioperative death occurred in 3/636 (0.5%) patients, 1/412 (0.2%) in PC and 2/224 (0.9%) in EV; P ⫽ .29. In patients with at least 1-year ultrasound follow-up (n ⫽ 290), perioperative strokes occurred in 3/290 (1.1%) of patients. All of
these occurred in PC, P ⫽ .25 (Table III). With regards to the functional capacity of patients after their stroke, two patients had a Rankin score of 1 (no significant disability despite symptoms; able to carry out all usual duties and activities), and one patient had a Rankin score of 2 (slight disability; unable to carry out all previous activities, but able to look after own affairs without assistance). No patient required re-exploration due to their hematoma. One patient in the eversion cohort experienced transient swallowing difficulty postoperatively with resolution within 1 week. This was considered a cranial nerve injury.
44 Crawford et al
Myocardial infarction (MI). MI occurred in 11/636 (1.7%) of patients in the intention to treat cohort, 6/412 (1.5%) in PC and 5/224 (2.2%) in EV; P ⫽ .53. In patients with at least 1-year ultrasound follow-up (n ⫽ 290), MI occurred in 3/290 (1.0%) of patients, 1/155 (0.7%) in PC, and 2/135 (1.5%) in EV; P ⫽ .6. Restenosis. Preoperative duplex indices of carotid stenosis were similar between PC and EV groups. These results, along with the surgical indications for CEA are summarized in Table I. For the group of patients with long-term follow-up imaging, mean follow-up was 5.5 years (1.0 to 11.7 years; median 5.5 years) for the PC group and 3.5 years (1.0 to 5.1 years; median 3.5 years) for the EV group. Overall, “moderate” restenosis (as defined above) occurred in 21.9% (34/155) patients in PC and 14.1% (19/ 135) in EV; P ⫽ .08. Early (within 2 months) moderate restenosis occurred in 5.8% (9/155) of patients in PC and 6.7% (9/135) of patients in the EV group; P ⫽ .76. Late moderate restenosis occurred in 16.1% (25/155) of patients in PC (mean follow-up: 5.5 years) and 7.4% (10/ 135) of patients in EV (mean follow-up: 3.5 years), P ⫽ .02 and none were considered for reintervention. Further evaluation of moderate stenosis with Kaplan Meier life table analysis showed no significant difference between the two groups (Fig 1); P ⫽ .91. Overall “severe” restenosis (ⱖ70%) occurred in 16 (5.5%) of patients, eight each (5.2%) in PC and (5.9%) EV; P ⫽ .78. Early or persistent severe restenosis occurred in two patients (0.7%), one each for PC (0.6%) and EV (0.7%) groups, (P ⫽ .92) likely representing “persistent” rather than recurrent stenoses. Late severe restenosis occurred in 14 (4.8%) of patients, seven each for PC (4.5%) and EV (5.2%) groups; (P ⫽ .79), with follow-up intervals noted above. Additional analysis by Kaplan-Meier analysis showed no difference (P ⫽ .54) between PC and EV groups (Fig 2). Re-do CEA. In the group of patients with at least 1-year follow-up, 11/290 (3.8%) patients required reintervention on their operated carotid artery; only a single patient had recurrent symptoms (TIA). In EV, four (3.0%) patients needed reintervention at a mean follow-up of 1.57 ⫾ 0.30 years. Two patients had re-do CEA with patch, and two had carotid stents placed. All seven (4.5%) of the patients in PC that needed reintervention had open re-do CEA procedures at a mean follow-up of 2.58 ⫾ 2.18 years. Late stroke. Three strokes (3/290; 1.1%) occurred during late follow-up, all in the PC group. Only one stroke could be ascribed to the operated carotid artery. This patient is a 66-year-old female with heparin-induced thrombocytopenia who had a previous postoperative stroke and on return to the operating room had platelet thrombi forming on the Dacron patch. She eventually recovered and was placed on Warfarin. A subsequent stroke occurred 1.5 years later when she became subtherapeutic on her anticoagulation in the presence of a critical recurrent stenosis. Late survival. In the cohort of patients with anatomic follow-up, crude mortality was 18.2% (28/155) in the PC
JOURNAL OF VASCULAR SURGERY July 2007
Fig 1. Life table analysis of moderate restenosis comparison for patch and eversion cohorts.
group (5.5 year mean follow-up) and 6.7% (9/135) deaths in the EV group (3.5 year mean follow-up); P ⫽ .04. This difference is likely due to differences in follow-up duration as Kaplan-Meier life table analysis for survival showed no difference between EV and PC ( P ⫽ .86). Fig 3 demonstrates a 3- and 5-year survival of 97% ⫾ 1% vs 93% ⫾ 3% and 88% ⫾ 3% vs 92% ⫹ 3% for PC vs EV groups, respectively. Logistic regression analysis was performed to assess variables associated with death, stroke, and moderate and severe restenosis. Increasing age at operation (odds ratio [OR] 1.14 [1.1-1.2], P ⬍ .001); smoking: (OR 5.4 [1.520.0], P ⫽ .01) and diabetes (OR 2.9 [1.1-7.5], P ⫽ .03), were all associated with inferior survival in our cohort. Female gender: (OR 3.2 [CI: 1.61-6.7], P ⫽ .001) and a history of peripheral vascular disease (OR 2.03 [CI: 1.04.1], P ⫽ .048), increased the likelihood of moderate restenosis in our cohort. Female sex (OR 3.72 [1.02-13.5], P ⫽ .046) was associated with severe restenosis irrespective of surgical technique. Univariate analysis (no independent correlates were found) of factors associated with late stroke showed that female gender (OR 7.6 [CI: 0.88-66.7], P ⫽ .042) was significantly associated with stroke in our patients.
JOURNAL OF VASCULAR SURGERY Volume 46, Number 1
Crawford et al 45
Fig 3. Life table analysis of survival comparison for patch and eversion. Fig 2. Life table analysis of severe restenosis comparison for patch and eversion cohorts.
DISCUSSION The safety and efficacy of CEA has been demonstrated in a variety of level 1 data,1-4 administrative and NSQIP database studies9, 20 and large single center series.6, 7 However CEA is complicated by a finite rate of restenosis, which in some cases may require reintervention, although criteria for same are not evidence based. Our published data on CEA analyzed anatomic durability during a median follow-up period of 6 years, and identified a 7.7% failure rate (severe restenosis/occlusion, 4.5%; or reoperative CEA, 3.2%) and an overall perioperative stroke/death rate of 1.4%.6 Further study identified both risk factors for restenosis and the favorable impact of HMG-CoA reductase inhibitors on restenosis.17 Intrigued with the concept that eversion CEA could potentially improve restenosis rates, we adopted eversion endarterectomy in 2001. The current study was designed to assess the impact of eversion on restenosis after CEA with the desirable concept that the “control” patch CEA cohort consisted of our immediately preceding experience. Consequently, our length of follow-up is longer for patients in the patch (5.5 years) vs those in our eversion (3.5 years) CEA cohort. Several advantages have been cited in favor of the eversion technique. Eversion CEA avoids the use of prosthetic material and is particularly useful in situations in
which the internal carotid artery is coiled or kinked.21 Concerns with carotid eversion include difficulty in visualization of the endarterectomy end-point, increased technical demand, difficulty when a shunt needs to be used, and the possibility that a distal flap may remain if the endarterectomy endpoint is not properly performed.16,22 With regards to the use of a shunt, reports from multiple groups with extensive experience have described the safety and feasibility of shunting during carotid eversion.8,23 However, in our opinion, shunt use remains more cumbersome and technically demanding when using the eversion technique. Accordingly, when EEG changes, or the response of the patient who was awake in the case of local anesthesia, mandate the use of a shunt, longitudinal arteriotomy and patch closure is our preferred method. Reports from multiple centers, which include retrospective studies and studies by individual surgeons, show that clinical results with the eversion technique are comparable to those obtained with traditional CEA.8,15,24-27 Some nonrandomized, prospective studies, report improved results for eversion CEA compared with patch CEA in both perioperative and longterm outcomes.22,28-30 Eversion CEA has been further validated in prospective, randomized trials.23,31 The largest of these studies is the Eversion Carotid Endarterectomy Versus Standard Trial (EVEREST). This study included some 1353 patients, 678 for the conventional group and 675 in the eversion group. EVEREST found no statistical difference in
46 Crawford et al
late outcomes (stroke, death, and restenosis) between standard (patch and primary closure) and eversion CEA. Subgroup analysis showed that restenosis rates were statistically comparable (2.8% vs 1.5%) for eversion and patch, respectively, while both had significantly lower restenosis rates than primary closure.31 More recently, a systematic review of the literature by the same authors, which included five controlled trials and almost 2500 patients, found that eversion CEA was associated with a lower risk of restenosis, when compared with conventional (patch) CEA, albeit rates of clinically relevant outcomes such as stroke and death were equal.16 The authors of this review argued that the inability to see clinical differences despite the differences in restenosis rates might be related to sample size. In view of excellent perioperative outcomes achieved in contemporary practice with CEA, we anticipated no discernable impact on same with eversion CEA.6 Our data in this regard are consistent with substantial literature comparing perioperative outcomes with the two surgical techniques. Our 1% combined perioperative stroke/death rate is consistent with that noted in the EVEREST trial (1.3%) and the Cochrane Systematic Review (2.1%).16,31 It is important to note that the Cochrane Review included four trials which compared eversion vs patch angioplasty as in our series, and one, (EVEREST), which included CEA with primary repair in the conventional CEA group.23,26,31-33 Further subgroup analysis of ipsilateral strokes at 4 years in the EVEREST trial, which compared eversion and patch separately, still showed no difference (3.9%, 1.2%, P ⫽ .3) between these two techniques.31 Subgroup analysis of the Cochrane Review again revealed no differences between the patch and eversion groups, in the parameters of perioperative strokes (1.7% vs 2.4%) or perioperative death (2.0% vs 1.9%) for eversion and patch, respectively.34 Even though all of the perioperative strokes in our cohort occurred in the patch group, the stroke rate was not different when comparing the patch and eversion groups (P ⫽ .25). This result is consistent with that found by Ballotta et al in a prospective trial of patients undergoing sequential bilateral CEA with patch and eversion techniques32 as well as in a larger related study published by the same group.23 Comparison with other studies is complicated in some circumstances, since there is variation in the inclusion of primary closure in the conventional CEA groups. What is consistent is that eversion is at least as effective and even superior to conventional CEA in consideration of perioperative stroke and death as an endpoint.8,22,25,26,28,29,33,35 Our analysis demonstrated equivalent late mortality for the two procedures after life table analysis corrected for the discrepancy in follow-up intervals. Restenosis after CEA has been extensively studied, yet will vary as a function of defining criteria and follow-up intervals. It is also recognized that there is a bimodal temporal distribution of restenosis reflecting early intimal hyperplasia (largely within 1.5 years of surgery) and later progressive atherosclerosis, each with likely separate (albeit poorly defined) risk factors. With respect to the technical component of the operation, there is consensus that patch closure is superior to primary closure,23,31 but as noted above, the literature claiming superiority of eversion (vs
JOURNAL OF VASCULAR SURGERY July 2007
patch) is less convincing. In this report, the roughly 1%/ year cumulative incidence of severe recurrent stenosis is consistent with our prior experience6 and compares favorably with the ⱕ3%/year reported in the literature. Furthermore, there was no difference in restenosis between eversion and patch, consistent with both the EVEREST and Cochrane Systematic Review.31,34 This is also consistent with other smaller nonrandomized and single surgeon series that either show no differences in restenosis rates between eversion and patch angioplasty or show superiority of eversion over patch repair.21-23,29,32 Furthermore, it has been suggested that the anatomic localization of recurrent stenosis might differ with the two techniques. Focus on the common carotid artery, (particularly when the latter is heavily diseased) as the site of recurrent stenoses was emphasized by Green et al.22 However, in this study, most severe recurrent lesions occurred in the internal carotid artery with no discernable patterns noted between eversion vs patch CEA. One contemporary study has shown that eversion CEA may have significantly higher rates of restenosis when compared against patch CEA.35 Brothers reported rates of ⬎50% restenosis as 38% and 6% (P ⬍ .001) for eversion and patch CEA at 36 months by life table analysis. Our life table analysis at 36 months show 20% and 16% (P ⫽ .91) moderate (ie, non- flow-limiting) restenosis rates for patch and eversion CEA, likely indicating similar “remodeling biology” for the two operations, and also reflecting the stringent duplex criteria utilized herein. However, Brothers reported ⬎80% restenosis of 6% and 5% for eversion and patch CEA respectively at 36 months by life table analysis; these figures are quite similar to our data. Aside from technical components of CEA, our data are consistent with prior literature identifying female gender as associated with a four-fold increased risk of restenosis.36,37 Others have reported that perioperative outcomes are also inferior in females.38 There are recent data to support the notion that carotid arteries are smaller in women, even after adjusting for body size, age and blood pressure.39 However, at least one study demonstrated equivalent restenosis for men and women after EV, and also noted luminal expansion of the internal carotid artery after eversion endarterectomy.40 Precise data on ICA diameter are not available from this study, and it remains unknown if increased restenosis rates reported in women indeed reflect a gender effect, or if female gender is a surrogate for smaller arteries. Finally, a history of peripheral vascular disease increased the likelihood of moderate restenosis in our cohort, consistent with other reports.37 There are several limitations to our study; first, our sample size is likely not large enough to detect absolute differences between the two surgical techniques. If restenosis rates reported in the EVEREST trial are used as a benchmark, simple power calculations indicate that some 2000 patients per cohort would be required to detect such differences with 80% power. While our study is underpowered with respect to demonstrating statistical differences in restenosis rates, it demonstrates that good results can be obtained with patch and eversion CEA. In addition, our
JOURNAL OF VASCULAR SURGERY Volume 46, Number 1
results demonstrate that previously reported differences between patch and eversion restenosis rates cannot be directly attributed to technical variation across single-surgeon experiences.35 Although we recognize this as a limitation, previous report comparing these two techniques have found significant differences in restenosis rates with comparable numbers of patients.35 A second limitation of our study is the number of patients (approximately 50% of total which includes deceased patients) for whom we did not have available longterm Duplex data. This limitation can be ascribed partly to the retrospective nature of our study, and our requirement for objective ultrasound velocities (not mere qualitative descriptions) for inclusion into the study. Since many of our patients are from remote geographic regions, their follow-up was done with their primary care physicians who referred patients back only when major abnormalities were noted in their ultrasounds. Even with our current follow-up numbers, the EV and PC groups were well matched in terms of the severity of their disease as well as their comorbidities and demographic features. CONCLUSION These data are consistent with prior randomized studies and indicate that restenosis rates are similar between eversion and patch CEA, suggesting that such restenosis rates represent biological remodeling phenomenon rather than technical variations of operation. Accordingly, surgeons obtaining good results with patch closure should not adopt eversion in anticipation of improving restenosis rates. AUTHOR CONTRIBUTIONS Conception and design: RSC, RPC Analysis and interpretation: RSC, TC, TH, JP, MC, RPC Data collection: RSC, TM, JD, MC Writing the article: RSC, TC, RPC Critical revision of the article: RSC, TC, RPC Final approval of the article: RSC, TC, TH, JP, MC, RPC Statistical analysis: RSC, TC Obtained funding: RPC Overall responsibility: RSC REFERENCES 1. Beneficial 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(7):44553. 2. Randomized trial of endarterectomy for recently symptomatic carotid stenosis: final results of the MRC European Carotid Surgery Trial (ECST). Lancet 1998;351(9113):1379-87. 3. Endarterectomy for asymptomatic carotid artery stenosis. Executive Committee for the Asymptomatic Carotid Atherosclerosis Study. JAMA 1995;273(18):1421-8. 4. Halliday A, Mansfield 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: randomized controlled trial. Lancet 2004;363(9420):1491-502. 5. 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. Stroke 1998;29(2):554-62.
Crawford et al 47
6. LaMuraglia GM, Brewster DC, Moncure AC, Dorer DJ, Stoner MC, Trehan SK, et al. Carotid endarterectomy at the millennium: what interventional therapy must match. Ann Surg 2004;240(3):535-44; discussion 44-6. 7. Ballotta E, Da Giau G, Piccoli A, Baracchini C. Durability of carotid endarterectomy for treatment of symptomatic and asymptomatic stenoses. J Vasc Surg 2004;40(2):270-8. 8. Shah DM, Darling RC, 3rd, Chang BB, Paty PS, Kreienberg PB, Lloyd WE, et al. Carotid endarterectomy by eversion technique: its safety and durability. Ann Surg 1998;228(4):471-8. 9. Stoner MC, Abbott WM, Wong DR, Hua HT, Lamuraglia GM, Kwolek CJ, et al. Defining the high-risk patient for carotid endarterectomy: an analysis of the prospective National Surgical Quality Improvement Program database. J Vasc Surg 2006;43(2):285-95; discussion 95-6. 10. Bond R, Rerkasem K, AbuRahma AF, Naylor AR, Rothwell PM. Patch angioplasty versus primary closure for carotid endarterectomy. Cochrane Database Syst Rev 2004(2):CD000160. 11. Mannheim D, Weller B, Vahadim E, Karmeli R. Carotid endarterectomy with a polyurethane patch versus primary closure: a prospective randomized study. J Vasc Surg 2005;41(3):403-7; discussion 7-8. 12. Counsell C, Salinas R, Warlow C, Naylor R. Patch angioplasty versus primary closure for carotid endarterectomy. Cochrane Database Syst Rev 2000(2):CD000160. 13. Bond R, Rerkasem K, Naylor AR, Aburahma AF, Rothwell PM. Systematic review of randomized controlled trials of patch angioplasty versus primary closure and different types of patch materials during carotid endarterectomy. J Vasc Surg 2004;40(6):1126-35. 14. De Bakey ME, Crawford ES, Cooley DA, Morris GC, Jr.Surgical considerations of occlusive disease of innominate, carotid, subclavian, and vertebral arteries. Ann Surg 1959;149(5):690-710. 15. Darling RC, 3rd, Shah DM, Chang BB, Paty PS, Kreienberg PB, Lloyd WE, et al. Carotid endarterectomy using the eversion technique. Semin Vasc Surg 2000;13(1):4-9. 16. Cao P, De Rango P, Zannetti S. Eversion vs conventional carotid endarterectomy: a systematic review. Eur J Vasc Endovasc Surg 2002; 23(3):195-201. 17. La Muraglia GM, Stoner MC, Brewster DC, Watkins MT, Juhola KL, Kwolek CK, et al. Determinants of carotid endarterectomy anatomic durability: effects of serum lipids and lipid-lowering drugs. J Vasc Surg 2005;41:762-8. 18. Grant EG, Benson CB, Moneta GL, Alexandrov AV, Baker JD, Bluth EI, et al. Carotid artery stenosis: gray-scale and Doppler US diagnosis-Society of Radiologists in Ultrasound Consensus Conference. Radiology 2003;229(2):340-6. 19. Cao P, Giordano G, De Rango P, Zannetti S, Chiesa R, Coppi G, et al. A randomized study on eversion versus standard carotid endarterectomy: study design and preliminary results: the Everest Trial. J Vasc Surg 1998;27(4):595-605. 20. Matsen SL, Chang DC, Perler BA, Roseborough GS, Williams GM. Trends in the in-hospital stroke rate following carotid endarterectomy in California and Maryland. J Vasc Surg 2006;44(3):488-93. 21. Kieny R, Hirsch D, Seiller C, Thiranos JC, Petit H. Does carotid eversion endarterectomy and reimplantation reduce the risk of restenosis? Ann Vasc Surg 1993;7(5):407-13. 22. Green RM, Greenberg R, Illig K, Shortell C, Ouriel K. Eversion endarterectomy of the carotid artery: technical considerations and recurrent stenoses. J Vasc Surg 2000;32(6):1052-61. 23. Ballotta E, Da Giau G, Saladini M, Abbruzzese E, Renon L, Toniato A. Carotid endarterectomy with patch closure versus carotid eversion endarterectomy and reimplantation: a prospective randomized study. Surgery 1999;125(3):271-9. 24. Reigner B, Reveilleau P, Gayral M, Papon X, Enon B, Chevalier JM. Eversion endarterectomy of the internal carotid artery: midterm results of a new technique. Ann Vasc Surg 1995;9(3):241-6. 25. Entz L, Jaranyi S, Nemes A. Eversion endarterectomy in surgery of the internal carotid artery. Cardiovasc Surg 1996;4(2):190-4. 26. Vanmaele RG, Van Schil PE, DeMaeseneer MG, Meese G, Lehert P, Van Look RF. Division-endarterectomy-anastomosis of the internal
48 Crawford et al
27.
28.
29.
30. 31.
32.
33.
carotid artery: a prospective randomized comparative study. Cardiovasc Surg 1994;2(5):573-81. Ballota E, Da Giau G, Guerra M, Toffano M. Carotid eversion endarterectomy and reimplantation: a safe and simple technique to prevent acute thrombosis-occlusion and/or early and late restenosis. Cardiovasc Surg 1997;5(5):473-80. Katras T, Baltazar U, Rush DS, Sutterfield WC, Harvill LM, Stanton PE, Jr.Durability of eversion carotid endarterectomy: comparison with primary closure and carotid patch angioplasty. J Vasc Surg 2001;34(3): 453-8. Radak D, Radevic B, Sternic N, Vucurevic G, Petrovic B, Ilijevski N, et al. Single center experience on eversion versus standard carotid endarterectomy: a prospective nonrandomized study. Cardiovasc Surg 2000; 8(6):422-8. Ballotta E, Da Giau G. Selective shunting with eversion carotid endarterectomy. J Vasc Surg 2003;38(5):1045-50. Cao P, Giordano G, De Rango P, Zannetti S, Chiesa R, Coppi G, et al. Eversion versus conventional carotid endarterectomy: late results of a prospective multicenter randomized trial. J Vasc Surg 2000;31:19-30. Ballotta E, Renon L, Da Giau G, Toniato A, Baracchini C, Abbruzzese E, et al. A prospective randomized study on bilateral carotid endarterectomy: patching versus eversion. Ann Surg 2000;232(1): 119-25. Balzer K, Guds I, Heger J, Jahnel B. Conventional thrombendarterectomy with carotid patch plasty vs. eversion endarterectomy: technique, indications and results. Zentralbl Chir 2000;125(3):228-38.
JOURNAL OF VASCULAR SURGERY July 2007
34. Cao PG, de Rango P, Zannetti S, Giordano G, Ricci S, Celani MG. Eversion versus conventional carotid endarterectomy for preventing stroke. Cochrane Database Syst Rev 2001(1):CD001921. 35. Brothers TE. Initial experience with eversion carotid endarterectomy: absence of a learning curve for the first 100 patients. J Vasc Surg 2005;42(3):429-34. 36. Hugl B, Oldenburg WA, Neuhauser B, Hakaim AG. Effect of age and gender on restenosis after carotid endarterectomy. Ann Vasc Surg 2006. 37. Trisal V, Paulson T, Hans SS, Mittal V. Carotid artery restenosis: an ongoing disease process. Am Surg 2002;68(3):275-9; discussion 9-80. 38. Sarac TP, Hertzer NR, Mascha EJ, O’Hara PJ, Krajewski LP, Clair DG, et al. Gender as a primary predictor of outcome after carotid endarterectomy. J Vasc Surg 2002;35(4):748-53. 39. Krejza J, Arkuszewski M, Kasner SE, Weigele J, Ustymowicz A, Hurst RW, et al. Carotid artery diameter in men and women and the relation to body and neck size. Stroke 2006;37(4):1103-5. 40. Darling RC, 3rd, Mehta M, Roddy SP, Paty PS, Kreienberg PB, Ozsvath KJ, et al. Eversion carotid endarterectomy: a technical alternative that may obviate patch closure in women. Cardiovasc Surg 2003; 11(5):347-52.
Submitted Dec 13, 2006; accepted Feb 22, 2007.
Restenosis after eversion vs patch closure carotid endarterectomy Robert S. Crawford, MD, Thomas K. Chung, MA, Thomas Hodgman, BA, Juan D. Pedraza, MD, Michael Corey, BA, and Richard P. Cambria, MD, Boston, Mass Objectives: Recurrent stenosis after carotid endarterectomy (CEA), previously reported to occur in 1%/year after operation, is the finite limitation of CEA. Eversion endarterectomy has a perceived lower incidence of recurrent stenosis, although data to support this contention are conflicting. The goal of the present study was to compare the late anatomic results of patch closure (PC) vs eversion CEA. Methods: Between January 1, 1995 and June 30, 2005, 950 CEA were performed by the senior author with adoption of eversion (EV) as the primary technique as of January 1, 2001. With minimum of 1-year follow-up by study inclusion criteria, complete follow-up data (including a duplex scan) was available for 155 PC and 135 EV patients. Incidence of moderate (50% to 70%) and severe (>70%) restenosis was examined at <2 months and >1 year after operation. Study end-points included late stroke, survival, and freedom from restenosis (moderate and severe) and were assessed by actuarial methods. Results: There were no differences in relevant demographic/clinical parameters, indication for surgery (69% overall asymptomatic) or early perioperative stroke/death (1.1% overall; P ⴝ .25) between PC and EV. After correction for different mean follow-up intervals (PC ⴝ 5.5 years vs EV ⴝ 3.5 years) by actuarial methods, there was no significant difference in late moderate (P ⴝ .91) or severe (P ⴝ .54) recurrent stenosis between PC and EV. In the group of patients with at least 1-year follow-up, 11/290 (3.8%) patients (4/135 EV, 7/155 PC; P ⴝ .39) required reintervention on their operated carotid artery at a cumulative follow-up interval of 4.5 years. Three strokes (3/290; 1.1%) occurred during late follow-up, all in the PC group, with only one related to the operated carotid artery. Late survival was similar between EV and PC, (P ⴝ .86). Female gender (odds ratio [OR] 3.72[1.02-13.5], P ⴝ .046) was associated with severe restenosis irrespective of surgical technique. Univariate analysis also showed that female gender (OR 7.6[CI: 0.88-66.7], P ⴝ .042) was associated with late stroke. Conclusion: These findings indicate that restenosis rates are similar between eversion and patch CEA and likely represent biological remodeling phenomenon rather than technical variations of operations. While EV offers distinct advantages in certain anatomic circumstances, adoption of EV with the hope of decreasing restenosis is not warranted. ( J Vasc Surg 2007;46:41-8.)
Randomized trials have validated the efficacy of carotid endarterectomy (CEA) for stroke prevention in both symptomatic and asymptomatic significant stenosis of the internal carotid artery.1-4 Perioperative stroke rates of 6% in symptomatic patients and 3% in asymptomatic patients are acceptable thresholds according to published guidelines.5 However, perioperative stroke rates of 2% for symptomatic disease and 1% for asymptomatic disease are reported in multiple contemporary surgical series.6-9 CEA is most often performed with a longitudinal arteriotomy typically closed with a patch, as randomized trials demonstrate that relevant outcome measures favor patch over primary closure.10-13 An alternative approach, eversion CEA, was originally described by DeBakey et al in the 1950s14 and involves the oblique transection of the internal carotid artery at its From the Division of Vascular and Endovascular Surgery of the General Surgical Services, Massachusetts General Hospital and Harvard Medical School. Competition of interest: none. Supported in part by the Harold & June Geneen Vascular Research Fund. Presented at the Thirty-third Annual Meeting of the New England Society for Vascular Surgery, Sep 22-24, 2006, Boston, Mass. Reprint requests: Richard P. Cambria, MD, WACC 448, 15 Parkman Street, Boston, MA 02114 (email: [email protected]) 0741-5214/$32.00 Copyright © 2007 by The Society for Vascular Surgery. doi:10.1016/j.jvs.2007.02.055
origin in the carotid bulb, followed by removal of the plaque using the eversion maneuver. The purported or demonstrated advantages of eversion CEA include decreased operative time, avoidance of prosthetic material, facilitating reconstruction of a kinked or coiled internal carotid artery (ICA), and decreased rates of restenosis.15,16 Indeed, recurrent stenosis after CEA, which occurs in 1%/year in our experience,17 is the limitation of CEA, and efforts to minimize it with technical variations are worthy of investigation. Largely as a function of this consideration, the senior author (R.P.C.) adopted eversion as the primary reconstruction mode of CEA in January of 2001. The goal of the present study was to compare the late anatomic results of patch closure vs eversion CEA with the inherent control of all operations being performed by a single surgeon. PATIENTS AND METHODS During the interval January 1, 1995 to June 30, 2005 (to insure minimum 1-year follow-up), 950 CEAs were performed. Procedures performed using primary closure, those performed in conjunction with other surgical procedures such as carotid artery bypass graft, valve surgery, or coronary bypass surgery were excluded from the analysis. After such exclusions, we examined the perioperative out41
JOURNAL OF VASCULAR SURGERY July 2007
42 Crawford et al
come of 636 CEAs (“intention to treat” cohort: patch, n ⫽ 412, eversion, n ⫽ 224) for perioperative death, strokes, and myocardial infarctions. Of these 636, we identified 290 CEAs, 155 done with patch endarterectomy technique (PC) and 135 done using the eversion technique (EV) with at least 1-year duplex ultrasound follow-up. This study was approved by the Institutional Review Board of the Massachusetts General Hospital. Clinical and demographic data were obtained from hospital records and office charts. Additional telephone contact was sought for those patients for whom recent clinical data were not available. Mortality data were obtained using hospital records as well as data from the National Death Index in Bethesda, Maryland. Perioperative complications were obtained from review of prospectively maintained weekly morbidity and mortality statistics reported annually in quality assurance reports. Office records were also quite detailed in this regard. Definitions for clinical variables are as follows: hypertension (HTN), use of anti-hypertensive drugs or BP 165/95 or greater on two separate occasions; renal failure was defined as those patients undergoing dialysis or with serum creatinine ⱖ1.5 mg/dl; diabetes (DM), fasting blood sugar ⬎220 mg/dl or greater or on DM meds; hypercholesterolemia (HC), total cholesterol level of ⬎220 mg/dl or greater or currently on hypercholesterolemia medications; peripheral vascular disease (PVD), intermittent claudication in one or both limbs or prior surgical/ endovascular treatment for PVD, ischemic rest pain, or gangrene of foot; coronary artery disease (CAD), history of myocardial infarction (MI), angina or ventricular ectopy, chronic heart failure or recent angina; atrial fibrillation (AF), diagnosed by history or electrocardiogram. Ipsilateral symptoms were those occurring in the same vascular territory as the operated carotid artery. Primary end-points of the study included restenosis of the operated carotid artery, perioperative and late stroke and death. Restenosis was examined both in the “early” (within 2 months of operation) and “late” (long-term follow-up) time points. If, however, severe stenosis was seen within 2 months, such patients were considered technical failures ie, persistent obstruction. “Overall” restenosis refers to the combined “early” and “late” follow-up components. Moderate (50% to 70% reduction in vessel diameter) restenosis was defined by the following duplex criteria: internal carotid artery (ICA) peak systolic velocity (PSV) ⬎125 cm/sec, ICA/CCA ratio ⬎2 or end-diastolic velocity (EDV) ⬎100 cm/sec and criteria for severe (⬎70% reduction in vessel diameter) restenosis with: (ICA-PSV ⬎250, ICA/CCA ratio ⬎4, or EDV ⬎140).17,18 Duplex exams were typically obtained in the immediate postoperative (within 2 months) period and at ⬎1 year. Follow-up duplex exams (ie, those not conducted in our vascular laboratory) were required to have actual velocity data to be included in the group with “adequate” ultrasound followup. Perioperative and late stroke was defined as a persistent neurological deficit lasting ⬎24 hours and generally confirmed by brain imaging. Freedom from restenosis and death were analyzed using Kaplan-Meier life table analysis.
The Rankin Score for functional assessment after stroke is defined elsewhere.19 Secondary end-points included myocardial infarction (MI), redo CEA, operative site hematomas (not necessarily requiring operation), perioperative transient ischemic attacks or amaurosis fugax (TIA/AFX), cranial nerve injury (persistent deficit at 1 month after surgery), and retinal infarcts. MI was confirmed by positive laboratory enzyme and electrocardiograph (ECG) analysis. Enzyme analysis was obtained only after routine postoperative ECG analysis showed new abnormalities. TIA/AFX was defined as an acute disturbance of focal neurological or monocular function with symptoms lasting ⬍24 hours. Retinal infarct was defined as total or partial monocular loss of vision with symptoms persisting longer than 24 hours, with diagnosis confirmed by an ophthalmologist. Procedures. All procedures were performed generally with patch carotid endarterectomy until the beginning of 2001 when preferentially the eversion technique was used. Operative procedures were most often performed under general anesthesia with electroencephalographic monitoring and selective shunting. Electroencephalogram (EEG) changes dictated a preference towards longitudinal arteriotomy and patch to facilitate shunting. A description of the eversion technique has been published elsewhere.15 Overall, regional anesthesia was performed in seven (2.4%) patients (2/135 for EV, 5/155 for PC). Intraprocedural verification of the technical result of operation is limited to a qualitative hand-held Doppler interrogation. Intraoperative shunting was used in 41 (14.1%) of patients, all in the patch group. EEG slowing was observed in 25 (8.6%) of patients, 22 (14.2%) of them in the patch group, and three (2.2%) in the eversion group. Overall, 30.3 % of patients were operated on for symptomatic disease (see Table I), 29.7% of patients in the patch group and 31.0% of patients for the eversion group. For redo CEA, only preocclusive (⬎90%; residual lumen diameter ⱕ1.5 mm) lesions verified by angiography as critical (in earlier years), or duplex and CTA in more recent years underwent reintervention. Two patients treated for restenosis with carotid stents also had angiography. Statistical methods. Cohort comparisons of demographic data were assessed using two-tailed t-tests for continuous variables, 2 and Fischer exact t tests for categorical data. Late outcomes were assessed using Kaplan-Meier life table analysis and the log-rank test was used when comparing subgroups across comparable time intervals. Univariate analysis of variables potentially associated with study endpoints was performed, and those that were statistically significant (threshold of 0.05) were included in a multiple logistic regression in order to identify predictors of late outcomes. RESULTS The patch group (n ⫽ 155) differed from the eversion (n ⫽ 135) group in four clinical domains as displayed in Table II. The higher incidence of “hypercholesterolemia”
JOURNAL OF VASCULAR SURGERY Volume 46, Number 1
Crawford et al 43
Table I. Preoperative characteristics of patients with complete anatomic follow-up Clinical factor
Patch (n ⫽ 155)
Eversion (n ⫽ 135)
P-value
Symptomatic* TIA/AFX Stroke Retinal Stroke Vertebrobasilar Symptoms Preoperative ultrasound ICA Velocity (cm/sec) ICA/CCA
29.7% (46/155) 22.6% (35/155) 7.1% (11/155) 1.3% (2/155) 1.3% (2/155)
31.0% (42/135) 20.0% (27/135) 11.9% (16/135) 5.2% (7/135) 3.7% (5/135)
.70 .67 .22 .09† .26†
412 ⫾ 131 (13-766) 5.61 ⫾ 2.43 (0.15-16.38)
419 ⫾ 117 (160-730) 5.74 ⫾ 1.98 (1.62-12.2)
.68 .66
TIA/AFX, transient ischemic attacks or amaurosis fugax; ICA, internal carotid artery; ICA/CCA, peak systolic velocity ratio of internal to common carotid artery. *Includes patients with multiple types of symptoms † Fisher exact t test shown
Table II. Demographic and clinical features of patients with complete anatomic follow-up
Age (years) Gender male Smoking HTN Renal failure DM Side (right) Hypercholesterolemia PVD CAD Atrial fibrillation Cholesterol Triglycerides LDL HDL
Patch cohort (n ⫽ 155)
Eversion cohort (n ⫽ 135)
P-value
73.4 ⫾ 10.3 (36-95) 57.4% (89/155) 74.4% (90/121) 85.1% (131/154) 18.7% (29/155) 24.7% (38/154) 54.8% (85/155) 71.4% (110/154) 29.2% (45/154) 69.5% (107/154) 16.9% (26/154) 184 ⫾ 54.1 (0-446) 192 ⫾ 174 (52-1623) 107 ⫾ 37.9 (20-186) 42.3 ⫾ 13.2 (9-84)
74.1 ⫾ 9.7 (41-94) 62.7% (84/134) 74.6% (97/130) 93.3% (126/135) 9.6% (13/135) 23.0% (31/135) 51.1% (69/135) 85.9% (113/135) 25.9% (35/135) 52.6% (70/132) 17.8% (24/135) 178 ⫾ 62.7 (0-303) 174 ⫾ 169 (0-1167) 99 ⫾ 51.2 (0-218) 47.0 ⫾ 24.0 (0-182)
.58 .36 .97 .03 .03 .73 .53 .003 .53 .003 .84 .40 .48 .26 .1
HTN, hypertension; DM, diabetes; PVD, peripheral vascular disease; CAD, coronary artery disease; LDL, low density lipoproteins; HDL, high density lipoproteins.
Table III. Perioperative complications for patients with complete anatomic follow-up
Hematoma MI Stroke CN injury TIA
Patch (n ⫽ 155)
Eversion (n ⫽ 135)
P-value
1.9% (3/155) 0.7% (1/155) 1.9% (3/155) 0% (0/155) 1.3% (2/155)
3.0% (4/135) 1.5% (1/135) 0% (0/135) 0.8% (1/135) 2.2% (3/135)
.71* .59* .25* .47* .66*
MI, myocardial infarction; CN, cranial nerve; TIA, transient ischemic attacks. *Fisher exact t test shown.
in the eversion group likely reflects the more aggressive use of HMG-CoA reductase inhibitors since 2001. Perioperative stroke/death. Combined perioperative stroke/death occurred in 10/636 (1.6%) patients in the intention-to-treat cohort. Strokes occurred in 7/636 (1.1%) patients (5/412 (1.2%) in PC and 2/224 (0.9%) in EV; P ⫽ 1.0). Across all CEAs performed during this timeframe, perioperative death occurred in 3/636 (0.5%) patients, 1/412 (0.2%) in PC and 2/224 (0.9%) in EV; P ⫽ .29. In patients with at least 1-year ultrasound follow-up (n ⫽ 290), perioperative strokes occurred in 3/290 (1.1%) of patients. All of
these occurred in PC, P ⫽ .25 (Table III). With regards to the functional capacity of patients after their stroke, two patients had a Rankin score of 1 (no significant disability despite symptoms; able to carry out all usual duties and activities), and one patient had a Rankin score of 2 (slight disability; unable to carry out all previous activities, but able to look after own affairs without assistance). No patient required re-exploration due to their hematoma. One patient in the eversion cohort experienced transient swallowing difficulty postoperatively with resolution within 1 week. This was considered a cranial nerve injury.
44 Crawford et al
Myocardial infarction (MI). MI occurred in 11/636 (1.7%) of patients in the intention to treat cohort, 6/412 (1.5%) in PC and 5/224 (2.2%) in EV; P ⫽ .53. In patients with at least 1-year ultrasound follow-up (n ⫽ 290), MI occurred in 3/290 (1.0%) of patients, 1/155 (0.7%) in PC, and 2/135 (1.5%) in EV; P ⫽ .6. Restenosis. Preoperative duplex indices of carotid stenosis were similar between PC and EV groups. These results, along with the surgical indications for CEA are summarized in Table I. For the group of patients with long-term follow-up imaging, mean follow-up was 5.5 years (1.0 to 11.7 years; median 5.5 years) for the PC group and 3.5 years (1.0 to 5.1 years; median 3.5 years) for the EV group. Overall, “moderate” restenosis (as defined above) occurred in 21.9% (34/155) patients in PC and 14.1% (19/ 135) in EV; P ⫽ .08. Early (within 2 months) moderate restenosis occurred in 5.8% (9/155) of patients in PC and 6.7% (9/135) of patients in the EV group; P ⫽ .76. Late moderate restenosis occurred in 16.1% (25/155) of patients in PC (mean follow-up: 5.5 years) and 7.4% (10/ 135) of patients in EV (mean follow-up: 3.5 years), P ⫽ .02 and none were considered for reintervention. Further evaluation of moderate stenosis with Kaplan Meier life table analysis showed no significant difference between the two groups (Fig 1); P ⫽ .91. Overall “severe” restenosis (ⱖ70%) occurred in 16 (5.5%) of patients, eight each (5.2%) in PC and (5.9%) EV; P ⫽ .78. Early or persistent severe restenosis occurred in two patients (0.7%), one each for PC (0.6%) and EV (0.7%) groups, (P ⫽ .92) likely representing “persistent” rather than recurrent stenoses. Late severe restenosis occurred in 14 (4.8%) of patients, seven each for PC (4.5%) and EV (5.2%) groups; (P ⫽ .79), with follow-up intervals noted above. Additional analysis by Kaplan-Meier analysis showed no difference (P ⫽ .54) between PC and EV groups (Fig 2). Re-do CEA. In the group of patients with at least 1-year follow-up, 11/290 (3.8%) patients required reintervention on their operated carotid artery; only a single patient had recurrent symptoms (TIA). In EV, four (3.0%) patients needed reintervention at a mean follow-up of 1.57 ⫾ 0.30 years. Two patients had re-do CEA with patch, and two had carotid stents placed. All seven (4.5%) of the patients in PC that needed reintervention had open re-do CEA procedures at a mean follow-up of 2.58 ⫾ 2.18 years. Late stroke. Three strokes (3/290; 1.1%) occurred during late follow-up, all in the PC group. Only one stroke could be ascribed to the operated carotid artery. This patient is a 66-year-old female with heparin-induced thrombocytopenia who had a previous postoperative stroke and on return to the operating room had platelet thrombi forming on the Dacron patch. She eventually recovered and was placed on Warfarin. A subsequent stroke occurred 1.5 years later when she became subtherapeutic on her anticoagulation in the presence of a critical recurrent stenosis. Late survival. In the cohort of patients with anatomic follow-up, crude mortality was 18.2% (28/155) in the PC
JOURNAL OF VASCULAR SURGERY July 2007
Fig 1. Life table analysis of moderate restenosis comparison for patch and eversion cohorts.
group (5.5 year mean follow-up) and 6.7% (9/135) deaths in the EV group (3.5 year mean follow-up); P ⫽ .04. This difference is likely due to differences in follow-up duration as Kaplan-Meier life table analysis for survival showed no difference between EV and PC ( P ⫽ .86). Fig 3 demonstrates a 3- and 5-year survival of 97% ⫾ 1% vs 93% ⫾ 3% and 88% ⫾ 3% vs 92% ⫹ 3% for PC vs EV groups, respectively. Logistic regression analysis was performed to assess variables associated with death, stroke, and moderate and severe restenosis. Increasing age at operation (odds ratio [OR] 1.14 [1.1-1.2], P ⬍ .001); smoking: (OR 5.4 [1.520.0], P ⫽ .01) and diabetes (OR 2.9 [1.1-7.5], P ⫽ .03), were all associated with inferior survival in our cohort. Female gender: (OR 3.2 [CI: 1.61-6.7], P ⫽ .001) and a history of peripheral vascular disease (OR 2.03 [CI: 1.04.1], P ⫽ .048), increased the likelihood of moderate restenosis in our cohort. Female sex (OR 3.72 [1.02-13.5], P ⫽ .046) was associated with severe restenosis irrespective of surgical technique. Univariate analysis (no independent correlates were found) of factors associated with late stroke showed that female gender (OR 7.6 [CI: 0.88-66.7], P ⫽ .042) was significantly associated with stroke in our patients.
JOURNAL OF VASCULAR SURGERY Volume 46, Number 1
Crawford et al 45
Fig 3. Life table analysis of survival comparison for patch and eversion. Fig 2. Life table analysis of severe restenosis comparison for patch and eversion cohorts.
DISCUSSION The safety and efficacy of CEA has been demonstrated in a variety of level 1 data,1-4 administrative and NSQIP database studies9, 20 and large single center series.6, 7 However CEA is complicated by a finite rate of restenosis, which in some cases may require reintervention, although criteria for same are not evidence based. Our published data on CEA analyzed anatomic durability during a median follow-up period of 6 years, and identified a 7.7% failure rate (severe restenosis/occlusion, 4.5%; or reoperative CEA, 3.2%) and an overall perioperative stroke/death rate of 1.4%.6 Further study identified both risk factors for restenosis and the favorable impact of HMG-CoA reductase inhibitors on restenosis.17 Intrigued with the concept that eversion CEA could potentially improve restenosis rates, we adopted eversion endarterectomy in 2001. The current study was designed to assess the impact of eversion on restenosis after CEA with the desirable concept that the “control” patch CEA cohort consisted of our immediately preceding experience. Consequently, our length of follow-up is longer for patients in the patch (5.5 years) vs those in our eversion (3.5 years) CEA cohort. Several advantages have been cited in favor of the eversion technique. Eversion CEA avoids the use of prosthetic material and is particularly useful in situations in
which the internal carotid artery is coiled or kinked.21 Concerns with carotid eversion include difficulty in visualization of the endarterectomy end-point, increased technical demand, difficulty when a shunt needs to be used, and the possibility that a distal flap may remain if the endarterectomy endpoint is not properly performed.16,22 With regards to the use of a shunt, reports from multiple groups with extensive experience have described the safety and feasibility of shunting during carotid eversion.8,23 However, in our opinion, shunt use remains more cumbersome and technically demanding when using the eversion technique. Accordingly, when EEG changes, or the response of the patient who was awake in the case of local anesthesia, mandate the use of a shunt, longitudinal arteriotomy and patch closure is our preferred method. Reports from multiple centers, which include retrospective studies and studies by individual surgeons, show that clinical results with the eversion technique are comparable to those obtained with traditional CEA.8,15,24-27 Some nonrandomized, prospective studies, report improved results for eversion CEA compared with patch CEA in both perioperative and longterm outcomes.22,28-30 Eversion CEA has been further validated in prospective, randomized trials.23,31 The largest of these studies is the Eversion Carotid Endarterectomy Versus Standard Trial (EVEREST). This study included some 1353 patients, 678 for the conventional group and 675 in the eversion group. EVEREST found no statistical difference in
46 Crawford et al
late outcomes (stroke, death, and restenosis) between standard (patch and primary closure) and eversion CEA. Subgroup analysis showed that restenosis rates were statistically comparable (2.8% vs 1.5%) for eversion and patch, respectively, while both had significantly lower restenosis rates than primary closure.31 More recently, a systematic review of the literature by the same authors, which included five controlled trials and almost 2500 patients, found that eversion CEA was associated with a lower risk of restenosis, when compared with conventional (patch) CEA, albeit rates of clinically relevant outcomes such as stroke and death were equal.16 The authors of this review argued that the inability to see clinical differences despite the differences in restenosis rates might be related to sample size. In view of excellent perioperative outcomes achieved in contemporary practice with CEA, we anticipated no discernable impact on same with eversion CEA.6 Our data in this regard are consistent with substantial literature comparing perioperative outcomes with the two surgical techniques. Our 1% combined perioperative stroke/death rate is consistent with that noted in the EVEREST trial (1.3%) and the Cochrane Systematic Review (2.1%).16,31 It is important to note that the Cochrane Review included four trials which compared eversion vs patch angioplasty as in our series, and one, (EVEREST), which included CEA with primary repair in the conventional CEA group.23,26,31-33 Further subgroup analysis of ipsilateral strokes at 4 years in the EVEREST trial, which compared eversion and patch separately, still showed no difference (3.9%, 1.2%, P ⫽ .3) between these two techniques.31 Subgroup analysis of the Cochrane Review again revealed no differences between the patch and eversion groups, in the parameters of perioperative strokes (1.7% vs 2.4%) or perioperative death (2.0% vs 1.9%) for eversion and patch, respectively.34 Even though all of the perioperative strokes in our cohort occurred in the patch group, the stroke rate was not different when comparing the patch and eversion groups (P ⫽ .25). This result is consistent with that found by Ballotta et al in a prospective trial of patients undergoing sequential bilateral CEA with patch and eversion techniques32 as well as in a larger related study published by the same group.23 Comparison with other studies is complicated in some circumstances, since there is variation in the inclusion of primary closure in the conventional CEA groups. What is consistent is that eversion is at least as effective and even superior to conventional CEA in consideration of perioperative stroke and death as an endpoint.8,22,25,26,28,29,33,35 Our analysis demonstrated equivalent late mortality for the two procedures after life table analysis corrected for the discrepancy in follow-up intervals. Restenosis after CEA has been extensively studied, yet will vary as a function of defining criteria and follow-up intervals. It is also recognized that there is a bimodal temporal distribution of restenosis reflecting early intimal hyperplasia (largely within 1.5 years of surgery) and later progressive atherosclerosis, each with likely separate (albeit poorly defined) risk factors. With respect to the technical component of the operation, there is consensus that patch closure is superior to primary closure,23,31 but as noted above, the literature claiming superiority of eversion (vs
JOURNAL OF VASCULAR SURGERY July 2007
patch) is less convincing. In this report, the roughly 1%/ year cumulative incidence of severe recurrent stenosis is consistent with our prior experience6 and compares favorably with the ⱕ3%/year reported in the literature. Furthermore, there was no difference in restenosis between eversion and patch, consistent with both the EVEREST and Cochrane Systematic Review.31,34 This is also consistent with other smaller nonrandomized and single surgeon series that either show no differences in restenosis rates between eversion and patch angioplasty or show superiority of eversion over patch repair.21-23,29,32 Furthermore, it has been suggested that the anatomic localization of recurrent stenosis might differ with the two techniques. Focus on the common carotid artery, (particularly when the latter is heavily diseased) as the site of recurrent stenoses was emphasized by Green et al.22 However, in this study, most severe recurrent lesions occurred in the internal carotid artery with no discernable patterns noted between eversion vs patch CEA. One contemporary study has shown that eversion CEA may have significantly higher rates of restenosis when compared against patch CEA.35 Brothers reported rates of ⬎50% restenosis as 38% and 6% (P ⬍ .001) for eversion and patch CEA at 36 months by life table analysis. Our life table analysis at 36 months show 20% and 16% (P ⫽ .91) moderate (ie, non- flow-limiting) restenosis rates for patch and eversion CEA, likely indicating similar “remodeling biology” for the two operations, and also reflecting the stringent duplex criteria utilized herein. However, Brothers reported ⬎80% restenosis of 6% and 5% for eversion and patch CEA respectively at 36 months by life table analysis; these figures are quite similar to our data. Aside from technical components of CEA, our data are consistent with prior literature identifying female gender as associated with a four-fold increased risk of restenosis.36,37 Others have reported that perioperative outcomes are also inferior in females.38 There are recent data to support the notion that carotid arteries are smaller in women, even after adjusting for body size, age and blood pressure.39 However, at least one study demonstrated equivalent restenosis for men and women after EV, and also noted luminal expansion of the internal carotid artery after eversion endarterectomy.40 Precise data on ICA diameter are not available from this study, and it remains unknown if increased restenosis rates reported in women indeed reflect a gender effect, or if female gender is a surrogate for smaller arteries. Finally, a history of peripheral vascular disease increased the likelihood of moderate restenosis in our cohort, consistent with other reports.37 There are several limitations to our study; first, our sample size is likely not large enough to detect absolute differences between the two surgical techniques. If restenosis rates reported in the EVEREST trial are used as a benchmark, simple power calculations indicate that some 2000 patients per cohort would be required to detect such differences with 80% power. While our study is underpowered with respect to demonstrating statistical differences in restenosis rates, it demonstrates that good results can be obtained with patch and eversion CEA. In addition, our
JOURNAL OF VASCULAR SURGERY Volume 46, Number 1
results demonstrate that previously reported differences between patch and eversion restenosis rates cannot be directly attributed to technical variation across single-surgeon experiences.35 Although we recognize this as a limitation, previous report comparing these two techniques have found significant differences in restenosis rates with comparable numbers of patients.35 A second limitation of our study is the number of patients (approximately 50% of total which includes deceased patients) for whom we did not have available longterm Duplex data. This limitation can be ascribed partly to the retrospective nature of our study, and our requirement for objective ultrasound velocities (not mere qualitative descriptions) for inclusion into the study. Since many of our patients are from remote geographic regions, their follow-up was done with their primary care physicians who referred patients back only when major abnormalities were noted in their ultrasounds. Even with our current follow-up numbers, the EV and PC groups were well matched in terms of the severity of their disease as well as their comorbidities and demographic features. CONCLUSION These data are consistent with prior randomized studies and indicate that restenosis rates are similar between eversion and patch CEA, suggesting that such restenosis rates represent biological remodeling phenomenon rather than technical variations of operation. Accordingly, surgeons obtaining good results with patch closure should not adopt eversion in anticipation of improving restenosis rates. AUTHOR CONTRIBUTIONS Conception and design: RSC, RPC Analysis and interpretation: RSC, TC, TH, JP, MC, RPC Data collection: RSC, TM, JD, MC Writing the article: RSC, TC, RPC Critical revision of the article: RSC, TC, RPC Final approval of the article: RSC, TC, TH, JP, MC, RPC Statistical analysis: RSC, TC Obtained funding: RPC Overall responsibility: RSC REFERENCES 1. Beneficial 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(7):44553. 2. Randomized trial of endarterectomy for recently symptomatic carotid stenosis: final results of the MRC European Carotid Surgery Trial (ECST). Lancet 1998;351(9113):1379-87. 3. Endarterectomy for asymptomatic carotid artery stenosis. Executive Committee for the Asymptomatic Carotid Atherosclerosis Study. JAMA 1995;273(18):1421-8. 4. Halliday A, Mansfield 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: randomized controlled trial. Lancet 2004;363(9420):1491-502. 5. 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. Stroke 1998;29(2):554-62.
Crawford et al 47
6. LaMuraglia GM, Brewster DC, Moncure AC, Dorer DJ, Stoner MC, Trehan SK, et al. Carotid endarterectomy at the millennium: what interventional therapy must match. Ann Surg 2004;240(3):535-44; discussion 44-6. 7. Ballotta E, Da Giau G, Piccoli A, Baracchini C. Durability of carotid endarterectomy for treatment of symptomatic and asymptomatic stenoses. J Vasc Surg 2004;40(2):270-8. 8. Shah DM, Darling RC, 3rd, Chang BB, Paty PS, Kreienberg PB, Lloyd WE, et al. Carotid endarterectomy by eversion technique: its safety and durability. Ann Surg 1998;228(4):471-8. 9. Stoner MC, Abbott WM, Wong DR, Hua HT, Lamuraglia GM, Kwolek CJ, et al. Defining the high-risk patient for carotid endarterectomy: an analysis of the prospective National Surgical Quality Improvement Program database. J Vasc Surg 2006;43(2):285-95; discussion 95-6. 10. Bond R, Rerkasem K, AbuRahma AF, Naylor AR, Rothwell PM. Patch angioplasty versus primary closure for carotid endarterectomy. Cochrane Database Syst Rev 2004(2):CD000160. 11. Mannheim D, Weller B, Vahadim E, Karmeli R. Carotid endarterectomy with a polyurethane patch versus primary closure: a prospective randomized study. J Vasc Surg 2005;41(3):403-7; discussion 7-8. 12. Counsell C, Salinas R, Warlow C, Naylor R. Patch angioplasty versus primary closure for carotid endarterectomy. Cochrane Database Syst Rev 2000(2):CD000160. 13. Bond R, Rerkasem K, Naylor AR, Aburahma AF, Rothwell PM. Systematic review of randomized controlled trials of patch angioplasty versus primary closure and different types of patch materials during carotid endarterectomy. J Vasc Surg 2004;40(6):1126-35. 14. De Bakey ME, Crawford ES, Cooley DA, Morris GC, Jr.Surgical considerations of occlusive disease of innominate, carotid, subclavian, and vertebral arteries. Ann Surg 1959;149(5):690-710. 15. Darling RC, 3rd, Shah DM, Chang BB, Paty PS, Kreienberg PB, Lloyd WE, et al. Carotid endarterectomy using the eversion technique. Semin Vasc Surg 2000;13(1):4-9. 16. Cao P, De Rango P, Zannetti S. Eversion vs conventional carotid endarterectomy: a systematic review. Eur J Vasc Endovasc Surg 2002; 23(3):195-201. 17. La Muraglia GM, Stoner MC, Brewster DC, Watkins MT, Juhola KL, Kwolek CK, et al. Determinants of carotid endarterectomy anatomic durability: effects of serum lipids and lipid-lowering drugs. J Vasc Surg 2005;41:762-8. 18. Grant EG, Benson CB, Moneta GL, Alexandrov AV, Baker JD, Bluth EI, et al. Carotid artery stenosis: gray-scale and Doppler US diagnosis-Society of Radiologists in Ultrasound Consensus Conference. Radiology 2003;229(2):340-6. 19. Cao P, Giordano G, De Rango P, Zannetti S, Chiesa R, Coppi G, et al. A randomized study on eversion versus standard carotid endarterectomy: study design and preliminary results: the Everest Trial. J Vasc Surg 1998;27(4):595-605. 20. Matsen SL, Chang DC, Perler BA, Roseborough GS, Williams GM. Trends in the in-hospital stroke rate following carotid endarterectomy in California and Maryland. J Vasc Surg 2006;44(3):488-93. 21. Kieny R, Hirsch D, Seiller C, Thiranos JC, Petit H. Does carotid eversion endarterectomy and reimplantation reduce the risk of restenosis? Ann Vasc Surg 1993;7(5):407-13. 22. Green RM, Greenberg R, Illig K, Shortell C, Ouriel K. Eversion endarterectomy of the carotid artery: technical considerations and recurrent stenoses. J Vasc Surg 2000;32(6):1052-61. 23. Ballotta E, Da Giau G, Saladini M, Abbruzzese E, Renon L, Toniato A. Carotid endarterectomy with patch closure versus carotid eversion endarterectomy and reimplantation: a prospective randomized study. Surgery 1999;125(3):271-9. 24. Reigner B, Reveilleau P, Gayral M, Papon X, Enon B, Chevalier JM. Eversion endarterectomy of the internal carotid artery: midterm results of a new technique. Ann Vasc Surg 1995;9(3):241-6. 25. Entz L, Jaranyi S, Nemes A. Eversion endarterectomy in surgery of the internal carotid artery. Cardiovasc Surg 1996;4(2):190-4. 26. Vanmaele RG, Van Schil PE, DeMaeseneer MG, Meese G, Lehert P, Van Look RF. Division-endarterectomy-anastomosis of the internal
48 Crawford et al
27.
28.
29.
30. 31.
32.
33.
carotid artery: a prospective randomized comparative study. Cardiovasc Surg 1994;2(5):573-81. Ballota E, Da Giau G, Guerra M, Toffano M. Carotid eversion endarterectomy and reimplantation: a safe and simple technique to prevent acute thrombosis-occlusion and/or early and late restenosis. Cardiovasc Surg 1997;5(5):473-80. Katras T, Baltazar U, Rush DS, Sutterfield WC, Harvill LM, Stanton PE, Jr.Durability of eversion carotid endarterectomy: comparison with primary closure and carotid patch angioplasty. J Vasc Surg 2001;34(3): 453-8. Radak D, Radevic B, Sternic N, Vucurevic G, Petrovic B, Ilijevski N, et al. Single center experience on eversion versus standard carotid endarterectomy: a prospective nonrandomized study. Cardiovasc Surg 2000; 8(6):422-8. Ballotta E, Da Giau G. Selective shunting with eversion carotid endarterectomy. J Vasc Surg 2003;38(5):1045-50. Cao P, Giordano G, De Rango P, Zannetti S, Chiesa R, Coppi G, et al. Eversion versus conventional carotid endarterectomy: late results of a prospective multicenter randomized trial. J Vasc Surg 2000;31:19-30. Ballotta E, Renon L, Da Giau G, Toniato A, Baracchini C, Abbruzzese E, et al. A prospective randomized study on bilateral carotid endarterectomy: patching versus eversion. Ann Surg 2000;232(1): 119-25. Balzer K, Guds I, Heger J, Jahnel B. Conventional thrombendarterectomy with carotid patch plasty vs. eversion endarterectomy: technique, indications and results. Zentralbl Chir 2000;125(3):228-38.
JOURNAL OF VASCULAR SURGERY July 2007
34. Cao PG, de Rango P, Zannetti S, Giordano G, Ricci S, Celani MG. Eversion versus conventional carotid endarterectomy for preventing stroke. Cochrane Database Syst Rev 2001(1):CD001921. 35. Brothers TE. Initial experience with eversion carotid endarterectomy: absence of a learning curve for the first 100 patients. J Vasc Surg 2005;42(3):429-34. 36. Hugl B, Oldenburg WA, Neuhauser B, Hakaim AG. Effect of age and gender on restenosis after carotid endarterectomy. Ann Vasc Surg 2006. 37. Trisal V, Paulson T, Hans SS, Mittal V. Carotid artery restenosis: an ongoing disease process. Am Surg 2002;68(3):275-9; discussion 9-80. 38. Sarac TP, Hertzer NR, Mascha EJ, O’Hara PJ, Krajewski LP, Clair DG, et al. Gender as a primary predictor of outcome after carotid endarterectomy. J Vasc Surg 2002;35(4):748-53. 39. Krejza J, Arkuszewski M, Kasner SE, Weigele J, Ustymowicz A, Hurst RW, et al. Carotid artery diameter in men and women and the relation to body and neck size. Stroke 2006;37(4):1103-5. 40. Darling RC, 3rd, Mehta M, Roddy SP, Paty PS, Kreienberg PB, Ozsvath KJ, et al. Eversion carotid endarterectomy: a technical alternative that may obviate patch closure in women. Cardiovasc Surg 2003; 11(5):347-52.
Submitted Dec 13, 2006; accepted Feb 22, 2007.