Results of carotid artery stenting with distal embolic protection with improved systems: Protected Carotid Artery Stenting in Patients at High Risk for Carotid Endarterectomy (PROTECT) trial

Results of carotid artery stenting with distal embolic protection with improved systems: Protected Carotid Artery Stenting in Patients at High Risk for Carotid Endarterectomy (PROTECT) trial Jon S. Matsumura, MD,a William Gray, MD,b Seemant Chaturvedi, MD,c Dai Yamanouchi, MD, PhD,a Lei Peng, MS,d and Patrick Verta, MD,d Madison, Wisc; New York, NY; Detroit, Mich; and Santa Clara, Calif Objective: The Protected Carotid Artery Stenting in Patients at High Risk for Carotid Endarterectomy (PROTECT) study was performed to evaluate the safety and effectiveness of two devices for carotid artery stenting (CAS) in the treatment of carotid artery stenosis in patients at high risk for carotid endarterectomy (CEA): (1) a new embolic protection device, the Emboshield Pro (Abbott Vascular, Abbott Park, Ill), using the periprocedural composite end point of 30-day death, stroke, and myocardial infarction (DSMI), and (2) a carotid stent in conjunction with an embolic protection device (EPD) using the DSMI periprocedural composite end point plus ipsilateral stroke at up to 3 years for long-term evaluation. Methods: This prospective, multicenter clinical trial enrolled 220 consecutive participants between November 29, 2006, and January 14, 2008, followed by a second cohort of 102 participants between January 14 and June 18, 2008. Enrolled participants had carotid stenosis (symptomatic >50% or asymptomatic >80%). The first 220 subjects underwent distal EPD placement with a new large-diameter filter, and the second cohort of 102 underwent placement of an older EPD that is no longer manufactured. All 322 participants were to be treated with a dedicated carotid stent with a tapered, small, closed-cell design (Xact; Abbott Vascular) and were to be included in the long-term evaluation. Independent neurologic assessment was performed before CAS and at 1 day, 30 days, and annually after CAS. All primary end point events were independently adjudicated by a central committee. Results: The periprocedural composite end point of DSMI (95% confidence interval) in the first 220 participants was 2.3% (0.74%, 5.22%), with a combined death and stroke rate of 1.8% (0.50%, 4.59%) and a rate of death and major stroke of 0.5% (0.01%, 2.51%). As of January 3, 2011, the median follow-up for the entire 322-subject cohort for the long-term evaluation was 2.8 years. Freedom from the periprocedural composite of DSMI plus ipsilateral stroke thereafter was 95.4%, with an annualized ipsilateral stroke rate of 0.4%. Conclusions: CAS outcomes in patients at high risk for CEA have improved from earlier carotid stent trials. With periprocedural rates of DSMI of 2.3%, death or stroke at 1.8%, and death or major stroke rate of 0.5%, PROTECT has the lowest rate of periprocedural complications among other comparable single-arm CAS trials in patients at high risk for CEA. ( J Vasc Surg 2012;55:968-77.)

Legacy trials (North American Symptomatic Carotid Endarterectomy Trial [NASCET],1 European Carotid Surgery Trial [ECST],2 Asymptomatic Carotid Atherosclerosis Study [ACAS],3 and Asymptomatic Carotid Surgery Trial From the Division of Vascular Surgery, University of Wisconsin School of Medicine and Public Health, Madisona; Center for Interventional Vascular Therapy, Columbia University, New Yorkb; Wayne State University, Detroitc; and Abbott Laboratories Ltd, Vascular Endovascular Clinical Science, Santa Clara.d Competition of interest: J.M. has research grants with Abbott, Cook, Covidien, Endologix, and Gore. W.G. has research grants from and provides consulting services for Abbott Vascular, Johnson and Johnson, Medtronic, and W. L. Gore, and consulting services for Boston Scientific, Silk Road, and Contego. S.C. is a consultant for Abbott Vascular and BMS/Sanofi partnership. L.P. and P.V. are employees of Abbott Vascular and hold stock in Abbott Laboratories. Additional material for this article may be found online at www.jvascsurg.org. Reprint requests: Jon S. Matsumura, MD, Division of Vascular Surgery, University of Wisconsin, School of Medicine, Public Health, G5/325 Clinical Science Center, 600 Highland Ave, Madison, WI 53792 (e-mail: [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a competition of interest. 0741-5214/$36.00 Copyright © 2012 by the Society for Vascular Surgery. doi:10.1016/j.jvs.2011.10.120

968

[ACST]4) have established the indications for carotid endarterectomy (CEA) in patients with symptomatic stenosis of ⱖ50% and asymptomatic stenosis of ⱖ60%. The introduction of protected carotid artery stenting (CAS) has been applied primarily to patients at high risk for CEA. Several randomized trials (Carotid and Vertebral Artery Transluminal Angioplasty Study [CAVATAS],5 Stenting and Angioplasty with Protection in Patients at High Risk for Endarterectomy [SAPPHiRE],6,7 Stent-Protected Percutaneous Angioplasty Versus Carotid Endarterectomy [SPACE],8 Endarterectomy Versus Angioplasty in Patients With Symptomatic Severe Carotid Stenosis [EVA-3S],9 International Carotid Stenting Study [ICSS],10 and Carotid Revascularization Endarterectomy Versus Stenting Trial [CREST]11) have directly compared CEA with CAS, but only SAPPHiRE compared the two options in the subgroup of patients who were at high risk for CEA, which is the current indication approved by the U.S. Food and Drug Administration (FDA) for CAS in the United States. The Centers for Medicare and Medicaid Services have issued several CAS national coverage decisions that provide coverage for selected (⬎70% symptomatic) patients at high

JOURNAL OF VASCULAR SURGERY Volume 55, Number 4

risk for CEA using FDA-approved devices and for research subjects in CAS studies. In May 2011, the FDA approved CAS with the Acculink stent (Abbott Vascular, Abbott Park, Ill) with the Accunet embolic protection device (EPD; Abbott Vascular) in standard-risk symptomatic and asymptomatic patients based on the results of CREST, the largest international randomized trial that demonstrated that CAS is not inferior to CEA.11 The Asymptomatic Carotid Trial I (ACTI), another large randomized trial,12 is underway in the United States to compare CAS vs CEA in nonoctogenarian, asymptomatic patients at standard risk for CEA. With increasing experience with dedicated CAS devices in the past decade, patients who are at high risk for CAS have been better understood and defined. Identification of risk predictors for poor outcomes, such as recent symptoms, heavy calcification, arterial tortuosity, intraluminal thrombus, and advanced age, have enabled better selection of patients who may be at higher risk for CAS. Some of these risk factors can be overcome with new techniques or devices, but others interact for compounded risk when present together.13 Some of these risk factors also predict poor outcomes with medical therapy alone or with CEA.1,2,3,4,11 Concurrently, there have been iterative improvements in technology with devices designed for the carotid artery. This trial evaluates modern CAS systems in the hands of trained operators in the background of improving knowledge of the risk factors for CAS. METHODS Study design. The Protected Carotid Artery Stenting in Patients at High Risk for Carotid Endarterectomy (PROTECT) trial is a prospective, multicenter clinical trial that assessed CAS outcomes for patients at high risk for CEA. Study devices. The newer Emboshield Pro EPD (Abbott Vascular) has improved features that were designed specifically for carotid artery application. It has a larger diameter (6.0 or 7.2 mm), a lower crossing profile, enhanced radiopacity, smaller distal pore size, and requires a shorter distal landing site. The older EPD (Emboshield BareWire; Abbott Vascular), which is no longer available, was designed for aortocoronary saphenous vein graft restenosis and was made in diameters of 3.0, 4.0, 5.0, or 6.0 mm. A sample size of 220 participants was deemed statistically sufficient to provide the FDA with a reasonable assurance of safety and efficacy for the new EPD. To evaluate the second purpose of long-term safety and efficacy of the Xact stent system (Abbott Vascular), the FDA required a sample size of 320, irrespective of whether the already cleared Emboshield BareWire EPD or the investigational Emboshield Pro EPD was used. During PROTECT, the Emboshield BareWire was cleared for use, but the Emboshield Pro was still an investigational device. The Xact stent, which has FDA approval, is a selfexpanding, flared, nitinol, small, closed-cell stent available in a tapered configuration to match the anatomy of the

Matsumura et al 969

carotid artery across the bifurcation and also in straight configurations to treat lesions, such as restenosis at a clamp site, that do not involve the bifurcation. It was specifically designed for the carotid artery and deploys accurately with a monorail 6F-compatible deployment catheter. Patient characteristics. Patients were at high risk for CEA and met all of the inclusion criteria to be considered for the study (Appendix Table I, A and B, online only). In brief, a subject must have a carotid stenosis determined by ultrasound imaging or angiography (visual estimate) to be ⱖ50% for symptomatic patients or ⱖ80% for asymptomatic patients. Exclusion criteria included patients with a previously placed stent in the target artery, dementia or neurologic illness confounding neurologic assessment, known cardiac sources of emboli, other abnormal angiographic findings other than in a target lesion placing them at risk for a stroke, stroke ⱕ30 days of the procedure, or a history of an ipsilateral stroke with fluctuating neurologic symptoms ⱕ1 year of the index procedure. An enrolled patient was defined as any patient who was consented and had a study device that entered the artery, whether or not the procedure was ultimately successful. All clinical sites had a study investigational team consisting of at least one interventionist who performed the procedure, a dedicated research coordinator, and a neurologist or personnel trained in National Institutes of Health Stroke Scale (NIHSS) evaluation.14 Before subject participation, investigators had to obtain written Investigational Review Board/Ethics Committee approval for the protocol and the informed consent form. All study subjects had to provide written informed consent using the Investigational Review Board/Ethical Committee and sponsor-approved informed consent form before enrollment in the trial. Preprocedure and postprocedure assessments. All patients underwent laboratory, neurologic, and duplex ultrasound imaging or angiographic evaluations, or both, to determine eligibility (Appendix Table II, A and B, online only). The measurement of degree of stenosis by the angiographic core laboratory was used for final analysis and was based on the NASCET method. A neurologist performed the neurologic examination before the index procedure. The study neurologist or NIHSS-certified personnel performed the postprocedure follow-up neurologic evaluations. Study patients were evaluated at 30 ⫾ 10 days and annually for 3 years (⫾30 days) as reported in Appendix Table II, online only. Any significant neurologic event required examination by a neurologist, and appropriate imaging was obtained. Stroke was defined as the sudden onset of vertigo, numbness, dysphasia, weakness, visual field defects, dysarthria, or other focal neurologic deficits due to vascular lesions of the brain, such as hemorrhage, embolism, thrombosis, or rupturing aneurysm, persisting for ⬎24 hours. Major stroke was defined as an increase in the NIHSS score of ⬎4 points from the preprocedure and/or increase in the modified Rankin Scale (mRS) score of ⬎2 points from the preproce-

JOURNAL OF VASCULAR SURGERY April 2012

970 Matsumura et al

dure score, or any patient with a mRS score of ⱖ5 points.14-16 Baseline cardiac enzyme (creatine kinase [CK]) levels were obtained ⱕ24 hours before the index procedure and postprocedure. CK-MB levels were also required if CK levels were out of the normal reference range. Routine troponin measurement was not required. A 12-lead electrocardiogram (ECG) was required preprocedure (ⱕ24 hours), postprocedure (before discharge), and at the 30day follow-up visit. Treatment. Patients were placed on dual antiplatelet therapy consisting of aspirin (325 mg daily) at least 72 hours before the index procedure and continued indefinitely and clopidogrel (75 mg daily) at least 72 hours before the index procedure and continued for a minimum of 4 weeks. The CAS procedure, including placement of the study device and the EPD or the new-generation EPD, was performed by an interventionist who had documentation of successful training in the use of these devices at the site. Study end points. The primary EPD end point was a composite of death, any stroke, or myocardial infarction (DSMI) ⱕ30 days after CAS for the first 220 consecutively enrolled patients where the new-generation EPD was used. The long-term primary stent end point was DSMI ⱕ30 days postprocedure plus ipsilateral stroke up to 3 years for all 322 carotid stent recipients. The secondary end points included acute device success, procedural success at 30 days, composite of any transient ischemic attack (TIA) and amaurosis fugax at 30 days, and annual rate of clinically driven target lesion revascularization (TLR) through 3 years. Acute device success for the stent was defined as the attainment of ⬍50% residual stenosis covering an area no longer than the original lesion treated with the stent. Placement of an additional stent to treat a dissection or procedural complication as a bailout was considered to be an acute device failure. Determination of residual stenosis was based on angiographic core laboratory results. Acute device success for the distal protection device was defined as successful deployment and retrieval of the protection device before and after stent implantation, in the absence of angiographic distal embolization. Procedural success was defined as the attainment of ⬍50% residual stenosis of the target lesion and the absence of DSMI at 30 days postprocedure. TLR was designated as clinically driven if the patient developed recurrent symptoms and had stenosis ⱖ50% in the stented lesion or was asymptomatic and had a restenosis of ⱖ80% in the stented lesion. An independent Data Safety Monitoring Board (DSMB) reviewed the safety data, including major adverse events. An independent Clinical Events Committee (CEC) adjudicated all deaths and MIs through 30 days and all strokes and suspected strokes through 3 years. Independent core laboratories evaluated all protocol-required angiograms, duplex ultrasound images, and ECGs. Statistical analyses. The trial was powered to show that the periprocedural DSMI rate in the 220 subjects was below a prespecified value, determined from the 30-day

DSMI of prior CAS trials in patients at high risk for CEA: A Registry Study to Evaluate the Neuroshield Bare Wire Cerebral Protection System and Xact Stent in Patients at High Risk for Carotid Endarterectomy (SECuRITY),17 SAPPHiRE,18 ACCULINK for Revascularization of Carotids in High-Risk Patients (ARCHeR),19 Boston Scientific EPI: A Carotid Stenting Trial for High Risk Surgical Patients (BEACH),20 and Evaluation of the Medtronic AVE Self-Expanding Carotid Stent System with Distal Protection in the Treatment of Carotid Stenosis (MAVErIC).21 Based on this design, the maximum number of events to meet the composite end point in this cohort was 18 (8.2%). A total of 320 patients were to be enrolled to evaluate the long-term safety and efficacy of the stent device in conjunction with either EPD. Demographic characteristics and cardiac, anatomic, and comorbid risk factors were summarized using descriptive statistics. Continuous variables, such as the age of the patient at the time of enrollment, were summarized using descriptive statistics (number, mean, median, standard deviation, minimum and maximum value, and 95% confidence intervals). Categoric variables, such as sex, were summarized using proportions and 95% Clopper-Pearson confidence intervals. The rates for the periprocedural composite end point of DSMI and for the individual components (death, stroke, and MI) were computed using binary proportions. The rate for the long-term primary end point was estimated using Kaplan-Meier methods. Event rates at 30 days, and at 12, 24, and 36 months after the index procedure were derived from Kaplan-Meier estimates and tabulated. RESULTS Patient enrollment and disposition. The first 220 patients in whom the new-generation EPD was used were enrolled at 34 sites between November 29, 2006, and January 14, 2008. The additional 102 patients that comprised the long-term follow-up cohort were enrolled at 38 sites between January 14 and June 18, 2008. Patient enrollment and disposition is reported in Appendix Table III (online only). The follow-up compliance rate at 30 days was 100%. At the time of data cutoff for analysis, January 3, 2011, of the 322 subjects enrolled, 317 (98%) received the stent device and were eligible for the long-term follow-up phase of the study. Five subjects were not eligible for long-term follow-up because they did not receive the study stent; one minor stroke occurred 2 days after the procedure in these five subjects. At 1 year, 21 patients had died, 10 withdrew consent, and four were lost to follow-up. The 1-year follow-up compliance rate for the 286 subjects eligible to return for their visit was 98.6% (282 of 286). At 2 years, 36 patients had died, 13 withdrew consent, and nine were lost to follow-up. The 2-year follow-up compliance rate for the 268 subjects eligible to return for their visit was 96.6% (259 of 268). At 3 years, 80 subjects had not reached their 3-year study visit date and were not yet eligible for follow-up, 47

JOURNAL OF VASCULAR SURGERY Volume 55, Number 4

Matsumura et al 971

Table I. Baseline patient characteristics

Table II. High risk for carotid endarterectomy (CEA)

Variable

% (n/220) Mean ⫾ SD

95% CI, % or Median (min, max)

Age (year)

72.5 ⫾ 9.7

71.24, 73.81a 73.6 (39.8, 91.5) 23.60, 6.05b 56.90, 70.00b 8.71, 18.03b 11.34, 21.42b 64.90, 77.24b

Age ⱖ80 years Male sex Symptomatic Current smoker Coronary artery disease Previous myocardial infarction Congestive heart failure Known left ventricular dysfunction Hyperlipidemia requiring medication Hypertension requiring medication History of cardiac arrhythmia Aortic or mitral valvular disease Previous carotid endarterectomy Current contralateral disease History of peripheral vascular disease Previous valve replacement Previous CABG History of Transient ischemic attack Stroke Amaurosis fugax Clinical COPD Renal failure Renal insufficiency Diabetes requiring medication Liver failure, bleeding diathesis Gastrointestinal bleeding

29.5 (65) 63.6 (140) 12.8 (28/218) 15.9 (35) 71.4 (157) 30.5 (67) 19.5 (43)

24.45, 37.00b 14.52, 25.41b

25.0 (55)

19.42, 31.26b

89.5 (197)

84.73, 93.26b

88.6 (195)

83.68, 92.51b

19.1 (42)

14.12, 24.92b

14.1 (31)

9.78, 19.40b

23.2 (51)

17.77, 29.33b

54.1 (119)

47.26, 60.81b

35.9 (79) 1.4 (3) 29.1 (64)

29.57, 42.63b 0.28, 3.93b 23.18, 35.57b

18.2 (40) 16.8 (37) 9.1 (20) 18.6 (41) 4.1 (9) 22.3 (49)

13.32, 23.92b 12.13, 22.43b 5.64, 13.69b 13.72, 24.42b 1.89, 7.62b 16.96, 28.33b

30.9 (68)

24.87, 37.47b

0.0 (0) 7.3 (16)

b

0.00, 1.66 4.21, 11.54b

CABG, Coronary artery bypass grafting; CI, confidence interval; COPD, chronic obstructive pulmonary disease; SD, standard deviation. Note. Four patients’ symptomatic status could not be determined due to insufficient data. a By normal approximation method. b By Clopper-Pearson exact method.

patients died, 16 withdrew consent, and three were lost to follow-up (Appendix Fig, online only). The 3-year follow-up compliance rate of the 174 subjects eligible to return for their 3-year visit was 98.3% (171 of 174). Baseline demographics. The baseline patient characteristics are outlined in Table I. Participants were a mean age of 72.5 years, with octogenarians representing 29.5% of the patients, and 63.6% were men. Hyperlipidemia was present in 89.5%, 88.6% were hypertensive requiring medication, 71.4% had coronary artery disease, and 54.1% had current contralateral carotid artery disease. In this cohort, 12.8% of the patients were symptomatic; symptomatic patients were defined as having an ipsilateral TIA, amaurosis

High risk Anatomic risk factors Previous CEA with significant restenosis (as defined for symptomatic or asymptomatic subjects) Previous radiation treatment to the neck or radical neck dissection Target lesion is at or above the second vertebral body C2 (level of jaw) Inability to extend the head due to cervical arthritis or other cervical disorders Tracheostomy or tracheal stoma Laryngectomy Contralateral laryngeal nerve palsy Severe tandem lesions Comorbid risk factors Age ⱖ years Abnormal stress test. Treadmill, thallium or dobutamine echo are acceptable. The stress tests should be sufficiently abnormal to place this subject at increased risk for CEA Total occlusion of the contralateral carotid artery Left ventricular ejection fraction ⬍35% Congestive heart failure NYHA functional class III or higher Dialysis-dependent renal failure Canadian Cardiovascular Society angina classification III or higher, or unstable angina Requires simultaneous or staged coronary artery bypass surgery, cardiac valve surgery, peripheral vascular surgery, or abdominal aortic aneurysm repair within 60 days Myocardial infarction within previous 6 weeks Severe pulmonary disease, including at least one of the following: requires chronic O2 therapy; resting PO2 ⱕ 60 mm Hg; hematocrit ⱖ50%; FEV1 or DLCO ⱕ50% of normal Summary Subject had only anatomic risk factor(s) Subject had only comorbid risk factors(s) Subject had both anatomic and comorbid risk factors

% (n/220)

16.8 (37) 7.7 (17) 8.2 (18) 7.7 (17) 1.8 (4) 0.5 (1) 0.0 (0) 0.5 (1) 29.5 (65)

14.1 (31) 13.2 (29) 8.2 (18) 3.6 (8) 0.9 (2) 4.1 (9)

7.3 (16) 2.7 (6)

4.1 (9) 28.6 (63) 60.5 (133) 10.9 (24)

DLCO, Diffusion capacity of the lung for carbon monoxide; FEV1, forced expiratory volume in 1 second; NYHA, New York Heart Association; PO2, partial pressure of oxygen.

fugax, or stroke ⱕ180 days before the procedure. The baseline characteristics for long-term follow-up cohort are presented in Appendix Table IV (online only). There was no significant difference between the two cohorts. High risk for CEA. The PROTECT trial enrolled patients considered to be high risk for CEA (Table II). All patients had at least one anatomic or comorbid risk factor listed in the inclusion criteria. Overall, 28.6% of the patients enrolled in PROTECT met the high-risk criteria for anatomic risk factors, 60.5% had at least one comorbid risk factor, and 10.9% met anatomic and comorbid risk factors. Details of patient indication for high risk are summarized in Table II: 65 (29.5%) were aged ⱖ80 years, 37 (16.8%) had restenosis after CEA, 31 (14.1%) had an abnormal stress test result, and 29 (13.2%) presented with total occlusion of the contralateral artery.

JOURNAL OF VASCULAR SURGERY April 2012

972 Matsumura et al

Table III. Lesion characteristics Lesion variables Carotid lesion location Left internal Left internal and common Right internal carotid Right internal and common Lesion length (mm) Eccentric Calcification (moderate or severe) Ulcerated Thrombus Diameter stenosis, %

% (n/N) 95% CI, %a or Mean ⫾ SD (n) Median (min, max) 7.8 (17/217)

4.63, 12.25

39.6 (86/217) 7.4 (16/217)

33.08, 46.48 4.27, 11.70

45.2 (98/217) 18.0 ⫾ 5.6 (218)

38.42, 52.04 17.21, 18.72 17.8 (6.8, 42.0) 10.65, 20.60

15.1 (33/218)

46.1 (100/217) 39.31, 52.96 16.1 (35/218) 11.44, 21.61 0.5 (1/218) 0.01, 2.53 73.5 ⫾ 9.4 (218) 73.4 (50.5, 94.3) 72.22, 74.74

CI, Confidence interval; SD, standard deviation. a By Clopper-Pearson exact method.

Table IV. Periprocedural composite primary end point End point Death, stroke, or MI Death or stroke Death or major stroke Death Stroke Major stroke Minor stroke MI

% (n/220)

95% CI, %a

2.3 (5) 1.8 (4) 0.5 (1) 0.5 (1) 1.8 (4) 0.5 (1) 1.4 (3) 0.5 (1)

0.74, 5.22 0.50, 4.59 0.01, 2.51 0.01, 2.51 0.50, 4.59 0.01, 2.51 0.50, 4.59 0.01, 2.51

CI, Confidence interval; MI, myocardial infarction. a By Clopper-Pearson exact method.

Lesion characteristics. Lesion characteristics are summarized in Table III. The preprocedure stenosis was 73.5% ⫾ 9.4%. The stenosis was 73.7% ⫾ 9.0% in asymptomatic patients and 71.8 ⫾ 11.4% in symptomatic patients. These core laboratory measurements were commonly lower than site visual estimate assessments, which were 85.5% ⫾ 8.5%. The mean target lesion length was 18.0 ⫾ 5.6 mm. Calcification was observed in 46% (moderate) and 1% (severe), ulceration in 16.1%, and thrombus in 0.5%. Primary end point. The primary end point for the composite of DSMI at 30 days for the 220 patients where the new-generation EPD was used is summarized in Table IV. The components of the composite (death or stroke, death or major stroke) are also presented in Table IV. The 30-day primary composite end point of DSMI was 2.3% (five of 220). The combined death or stroke rate was 1.8% (four of 220). One patient (octogenarian and symptomatic) had a major stroke and died ⱕ30 days of the procedure, resulting in a combined death or major stroke rate of 0.5%. Minor stroke occurred in three patients (two octogenarian, one symptomatic). All strokes were located ipsilat-

eral to the target lesion. MI occurred in one patient (nonoctogenarian and symptomatic). The primary end point for the freedom from DSMI (to 30 days) or ipsilateral stroke (31 days to 3 years) is presented in Figs 1 and 2. In the 322-patient cohort, 11 primary end point events occurred ⱕ30 days (3.4%), and two minor ipsilateral strokes occurred on days 654 and 869 post-CAS; one of these was associated with restenosis. The freedom from 30-day DSMI or long-term ipsilateral stroke was 95.4% at 3 years. Excluding the periprocedural events, the freedom from long-term ipsilateral stroke was 98.8%. Secondary end point. Two subjects were excluded from this analysis because there were no postprocedural angiographic data and the angiographic core laboratory determination of residual stenosis could not be completed. The analysis excluded two additional subjects because they did not receive a study stent. Acute device success. The acute stent device success was achieved in 213 of 216 subjects (98.6%; Table V). In three cases, the investigator planned to use one stent but needed an additional stent to completely cover the lesion. Acute device success for the new-generation EPD was achieved in 215 of 221 participants (97.3%). There were three patients in whom the new-generation EPD could not be deployed, leading to the use of an additional device in two of these individuals. In three additional patients, the filter could not be retrieved as intended. One TIA and one vasospasm were associated with two of these six patients. Procedural success at 30 days. Procedural success was achieved in 211 of 218 participants (96.8%). Two subjects were removed from the denominator because postprocedure angiographic data were unavailable and thus residual stenosis could not be calculated. Five of seven nonprocedural successes were related to a primary end point event (ie, DSMI to 30 days), and two additional cases were related to the nonplacement of the EPD or stent nondeployment (Table V). Composite of any TIA and amaurosis fugax at 30 days. The composite of any TIA and amaurosis fugax at 30 days after the index procedure was counted per patient (Table VI). Only the first occurrence of TIA or amaurosis fugax ⱕ30 days after the index procedure was counted in each category. Nine events occurred in eight patients. The composite rate of TIA and amaurosis fugax up to 30 days postprocedure in the intention-to-treat analysis population was 3.6% (eight of 220). Separately, in the 200 patients, TIA occurred in eight (3.6%) and amaurosis fugax in one (0.5%). One patient experienced TIA (day 12) and amaurosis fugax (day 5) ⱕ30 days of the procedure. An occlusion of the stented artery 5 days postprocedure was also reported for this patient. All patients with TIA or amaurosis fugax had complete recovery. Annual rate of clinically driven TLR through 3 years. There were eight TLRs, and six were clinically driven: two occurred ⱕ365 days, three between 365 and 730 days, and one between 730 and 1095 days of the index procedure. There was no consensus about the management of asymptomatic restenosis, and the decision for reinterven-

JOURNAL OF VASCULAR SURGERY Volume 55, Number 4

Matsumura et al 973

Fig 1. Kaplan-Meier curve shows freedom from perioperative composite and ipsilateral stroke between 31 days and 3 years. The overall composite event-free rate is 96.6% at 1 year, 95.9% at 2 years, and 95.4% at 3 years.

tion was left at the discretion of the attending physician. Two TLRs were not classified as clinically driven because the core laboratory’s assessment of stenosis was ⬍80% (70.2% and 75.6%). Fig 3 shows freedom from clinically driven TLR up to 3 years. Of the six clinically driven TLR events on five subjects, four were asymptomatic, and two had a TIA. DISCUSSION Large randomized trials established the indications for CEA for selected asymptomatic and symptomatic patients with severe carotid stenosis. SAPPHiRE, the only randomized clinical trial of patients at high risk for CEA, suggests that there are selected patients where CAS is not inferior to CEA.18 One concern of SAPPHiRE was the high absolute 30-day DSMI rate with CEA (9.2%) or CAS (6.0%) in asymptomatic subjects.22 Guidelines advocate periprocedural death and stroke rates of ⱕ3% for intervention in selected asymptomatic patients with severe carotid artery stenosis. The PROTECT study was designed to evaluate an approved carotid stent and a new EPD in patients at high risk for CEA. The overall baseline characteristics reflected

the high-risk nature of the enrolled patients, with 30% octogenarians, 73% with coronary artery disease, 56% with current contralateral carotid artery disease, and 12% symptomatic patients. The 30-day composite end point of DSMI was 2.3%, the combined death or stroke rate was 1.8%, and the combined death or major stroke rate was 0.5%. These outcomes represent an improvement over prior reported trials of CAS with EPD in patients at high risk for CEA and are consistent with the trend of decreased 30-day DSMI rates in recent CAS trials compared with earlier regulatory trials from the first half of the 2000 decade: ●

The ARCHeR trial, started in 2000, had a 30-day DSMI rate of 8.3%.19

●

The SAPPHiRE, started in 2000, had a 30-day DSMI rate of 6.5%.18

●

The Carotid Artery Revascularization Using the Boston Scientific EPI FilterWire EX/EZ and the EndoTex NexStent (CABERNET) evaluated CAS with protection in 488 subjects (enrolled from 2001 to 2007), and the 30-day DSMI rate was 3.8%.23

JOURNAL OF VASCULAR SURGERY April 2012

974 Matsumura et al

Fig 2. Kaplan-Meier curve shows freedom from ipsilateral stroke between 31 days and 3 years for all subjects with available data. The event-free rate is 100.0% at 1 year, 99.3% at 2 years, and 98.8% at 3 years. ●

Table V. Acute device success Acute device success Xact stent Emboshield pro (generation 5) Procedure success

% (n/N) N ⫽ 220

95% CI, %a

98.6 (213/216) 97.3 (215/221) 96.8 (211/218)

95.99, 99.71 94.18, 99.00 93.50, 98.70

●

●

a

Confidence interval (CI) by Clopper-Pearson exact method.

Table VI. Transient ischemic attack (TIA) or amaurosis fugax Variable TIA or amaurosis fugax TIA Amaurosis fugax

% (n/220)

95% CI,%a

3.6 (8) 3.6 (8) 0.5 (1)

1.29, 5.24 1.29, 5.24 0.01, 2.51

a

Confidence interval (CI) by Clopper-Pearson exact method.

●

The BEACH trial, started in 2002, enrolled 747 patients to evaluate the outcomes of CAS in patients at high risk for CEA, and had a 30-day DSMI rate of 5.8% (inclusive of all three cohorts).20

●

The SECuRITY study started in 2003, and the DSMI 30-day rate was 7.5%.17 Carotid Revascularization with ev3 Arterial Technology Evolution (CREATE) enrolled 419 high-risk surgical patients starting in April 2004, and the 30-day DSMI rate was 6.2%.24 In the Embolic Protection with Flow Reversal (EMPIRE) study, 245 patients at high risk for CEA were enrolled starting in 2006, and the 30-day DSMI rate was 3.7%.25 In the Proximal Protection with the Mo.Ma Device During Carotid Artery Stenting (ARMOUR) study, 225 subjects at high risk for CEA were enrolled between 2007 and 2009, and the 30-day DSMI rate was 2.7%.26

This trend may represent not only the improvement of the study device systems but also better understanding of factors associated with high risk for CAS. In PROTECT, the 30-day DSMI rate trended higher in the overall cohort of 322 subjects compared with the 220 patients where the newer-generation EPD was used, even though this EPD was used in the first consecutive 220 patients. The patient demographic characteristics between

JOURNAL OF VASCULAR SURGERY Volume 55, Number 4

Matsumura et al 975

Fig 3. Kaplan-Meier curve is shown for freedom from clinically driven reintervention in the stented vessel. The event-free rate is 99.3% at 1 year, 98.2% at 2 years, and 98.2% at 3 years.

these two groups were comparable. Although PROTECT was not designed or powered to detect a difference between the EPD devices, the rates of 2.3% DSMI with the new EPD and 5.9% DSMI with the obsolete EPD are suggestive of improvement in EPD technology. When PROTECT is compared with other studies of CEA patients at high risk, 2.3% is the lowest periprocedural DSMI rate and represents less than one-third of the highest published rate. PROTECT has several limitations and features that are different from other trials. First, although rates are lower than in other comparable trials, the outcome definition and ascertainment methodology are not uniform between trials. Some studies have performed diffusion-weighted magnetic resonance imaging substudies for heightened sensitivity to brain signals, but this was not performed in this trial.27 For the neurologic evaluation, some trials require examination by a neurologist or independent physician and new stroke definitions are emerging. PROTECT used an independent neurologist or NIHSS-certified personnel. Although the NIHSS assessments have been shown to be reliable in general for ascertainment of neurologic deficits, there is some inconsistency among raters that may affect trial results.15

Second, this trial included selected patients at high risk for CEA who are not comparable with standard-risk patients selected for earlier, legacy CEA trials. Specifically, some of the high risk for CEA criteria, such as staged peripheral vascular surgery, abdominal aortic aneurysm repair, and abnormal stress test result, may be considered minimally higher risk in the modern era. Other high-risk criteria for entry into this cohort, such as recurrent carotid stenosis, were exclusion criteria for the legacy trials. Another limitation is that device-sizing guidelines and difficult anatomy for stenting limited the anatomic range of arteries that could be treated in this study in contrast with CEA, which is applicable to these uncommon small or large diameters or severe tortuosity. Although this trial reports relatively low rates of events in octogenarians, indications for carotid intervention for elderly patients are still controversial. Careful patient selection is necessary in these patients to ensure net benefit. In the United States, most carotid revascularizations are performed on patients with asymptomatic severe carotid stenosis, and PROTECT reflects this demographic. Despite several ongoing or soon to be initiated randomized trials, there is ongoing debate about the optimal treatment of

JOURNAL OF VASCULAR SURGERY April 2012

976 Matsumura et al

asymptomatic patients. Future trials of carotid revascularization plus optimal medical therapy vs optimal medical therapy alone are needed. This trial does not specifically address the efficacy of embolic protection vs stenting without embolic protection.

11.

12.

CONCLUSIONS CAS outcomes in patients at high risk for CEA have improved from earlier CAS trials. With periprocedural DSMI rates of 2.3%, death or stroke at 1.8%, and a death or major stroke rate of 0.5%, PROTECT has the lowest rate of periprocedural complications among other comparable single-arm CAS trials in patients at high risk for CEA. We acknowledge Yuan Gao, MS, for assistance for Biostatistics. AUTHOR CONTRIBUTIONS Conception and design: JM, WG, SC, PV Analysis and interpretation: JM, WG, SC, DY, LP, PV Data collection: JM, WG, SC, LP, PV Writing the article: JM, WG, SC, DY, LP, PV Critical revision of the article: JM, WG, SC, DY, LP, PV Final approval of the article: JM, WG, SC, DY, LP, PV Statistical analysis: LP, PV Obtained funding: Not applicable Overall responsibility: JM

13.

14.

15.

16.

17. 18.

19.

20.

REFERENCES 1. North American Symptomatic Carotid Endarterectomy Trial Collaborators. Beneficial effect of carotid endarterectomy in symptomatic patients with high-grade carotid stenosis. N Engl J Med 1991;325:445-53. 2. Randomised trial of endarterectomy for recently symptomatic carotid stenosis: final results of the MRC, European Carotid Surgery Trial (ECST). Lancet 1998;351:1379-87. 3. Executive Committee for the Asymptomatic Carotid Atherosclerosis Study. Endarterectomy for asymptomatic carotid artery stenosis. 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: randomised controlled trial. Lancet 2004;363:1491-502. 5. 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. 6. Flanigan DP, Flanigan ME, Dorne AL, Harward TR, Razavi MK, Ballard JL. Long-term results of 442 consecutive, standardized carotid endarterectomy procedures in standard-risk and high-risk patients. J Vasc Surg 2007;46:876-82. 7. Gurm HS, Yadav JS, Fayad P, Katzen BT, Mishkel GJ, Bajwa TK, et al. Long-term results of carotid stenting versus endarterectomy in high-risk patients. N Engl J Med 2008;358:1572-9. 8. Eckstein HH, Ringleb P, Allenberg JR, Berger J, Fraedrich G, Hacke W, et al. Results of the Stent-Protected angioplasty versus Carotid Endarterectomy (SPACE) study to treat symptomatic stenoses at 2 years: a multinational, prospective, randomised trial. Lancet Neurol 2008;7:893-902. 9. Mas JL, Trinquart L, Leys D, Albucher JF, Rousseau H, Viguier A, Bossavy JP, et al. Endarterectomy versus angioplasty in Patients with Symptomatic Severe Carotid stenosis (EVA-3S) trial: results up to 4 years from a randomised, multicentre trial. Lancet Neurol 2008;7:885-92. 10. Featherstone RL, Brown MM, Coward LJ, ICSS Investigators. International carotid stenting study: protocol for a randomised clinical trial

21.

22.

23.

24.

25.

26.

27.

comparing carotid stenting with endarterectomy in symptomatic carotid artery stenosis. Cerebrovasc Dis 2004;18:69-74. Brott TG, Hobson RW 2nd, Howard G, Roubin GS, Clark WM, Brooks W, et al. Stenting versus endarterectomy for treatment of carotid-artery stenosis. N Engl J Med 2010;363:11-23. ClinicalTrials.gov. Carotid Stenting vs. Surgery of Severe Carotid Artery Disease and Stroke Prevention in Asymptomatic Patients (ACT I). http://www.clinicaltrials.gov/ct2/show/NCT00106938?term⫽ACT⫹ and⫹stent&rank⫽2. Matsumura JS, Gray W, Chaturvedi S, Gao X, Cheng J, Verta P, et al. CAPTURE 2 risk-adjusted stroke outcome benchmarks for carotid artery stenting with distal embolic protection. J Vasc Surg 2010;52: 576-83, 583 e1-583 e2. Lyden P, Lu M, Jackson C, Marler J, Kothari R, Brott T, et al. Underlying structure of the National Institutes of Health Stroke Scale: results of a factor analysis. NINDS tPA Stroke Trial Investigators. Stroke 1999;30:2347-54. Josephson SA, Hills NK, Johnston SC. NIH Stroke Scale reliability in ratings from a large sample of clinicians. Cerebrovasc Dis 2006; 22:389-95. Quinn TJ, Lees KR, Hardemark HG, Dawson J, Walters MR. Initial experience of a digital training resource for modified Rankin scale assessment in clinical trials. Stroke 2007;38:2257-61. Gray WA. A cardiologist in the carotids. J Am Coll Cardiol 2004;43: 1602-5. 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. 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. Iyer SS, White CJ, Hopkins LN, Katzen BT, Safian R, Wholey MH, et al. Carotid artery revascularization in high-surgical-risk patients using the Carotid WALLSTENT and FilterWire EX/EZ: 1-year outcomes in the BEACH Pivotal Group. J Am Coll Cardiol 2008;51:427-34. Higashida RT, Popma JJ, Apruzzese P, Zimetbaum P; MAVErIC I and II Investigators. Evaluation of the Medtronic exponent self-expanding carotid stent system with the Medtronic guardwire temporary occlusion and aspiration system in the treatment of carotid stenosis: combined from the MAVErIC (Medtronic AVE self-expanding CaRotid Stent System with distal protection in the treatment of Carotid stenosis) I and MAVErIC II trials. Stroke 2010;41:e102-9. U.S. Food and Drug Administration. Circulatory Systems Devices Panel April 21, 2004. Available at: http://www.fda.gov/ohrms/ dockets/ac/04/briefing/4033b1.htm. Accessed September 12, 2011. Hopkins LN, Myla SV, Grube E, Eles G, Dave R, Jaff MR, et al. Carotid artery revascularisation in high-surgical-risk patients with the NexStent and the FilterWire EX/EZ: 3-year results from the CABERNET trial. EuroIntervention 2010;5:917-24. Safian RD, Bresnahan JF, Jaff MR, Foster M, Bacharach JM, Maini B, et al. Protected carotid stenting in high-risk patients with severe carotid artery stenosis. J Am Coll Cardiol 2006;47:2384-9. Clair DG, Hopkins LN, Mehta M, Kasirajan K, Schermerhorn M, Schonholz C, et al. Neuroprotection during carotid artery stenting using the GORE flow reversal system: 30-day outcomes in the EMPiRE Clinical Study. Catheter Cardiovasc Intervent 2011;77:420-9. Ansel GM, Hopkins LN, Jaff MR, Rubino P, Bacharach JM, Scheinert D, et al. Safety and effectiveness of the INVATEC MO.MA proximal cerebral protection device during carotid artery stenting: results from the armour pivotal trial. Catheter Cardiovasc Intervent 2010;76:1-8. Bonati LH, Jongen LM, Haller S, Flach HZ, Dobson J, Nederkoorn PJ, et al. New ischaemic brain lesions on MRI after stenting or endarterectomy for symptomatic carotid stenosis: a substudy of the international carotid stenting study (ICSS). Lancet Neurol 2010;9:353-62.

Submitted Apr 1, 2011; accepted Oct 17, 2011.

Additional material for this article may be found online at www.jvascsurg.org.

976.e1 Matsumura et al

JOURNAL OF VASCULAR SURGERY April 2012

Appendix Table I (online only). A, Inclusion criteria 1. Patient must be ⱖ18 years of age. 2. Patient has the ability to cooperate with study procedures and agrees to return for all required follow-up visits, tests, and examinations. 3. Female patients of childbearing potential must have a documented negative pregnancy test within 30 days prior to the index procedure. 4. Patients taking warfarin may be included if their dosage is reduced before the index procedure to result in an international normalized ratio of ⱕ1.5. Warfarin may be restarted to therapeutic dose after the index procedure. 5. The patient must sign a written informed consent prior to the initiation of any study procedures, using a form that is approved by the Investigational Review Board or Medical Ethics Committee. 6. The life expectancy of the patient is at least 2 years. 7. The patient has a lesion located in the internal carotid artery (ICA); the carotid bifurcation may be involved. 8. The target lesion is intended to be treated with a single stent. 9. Target ICA vessel diameter must be visually estimated to be: ⬎2.5 mm and ⬍7.0 for the Emboshield pro, ⬎2.8 mm and ⬍6.2 for the Emboshield (generation 3); and ⬎4.0 mm and ⬍9.0 mm for the Xact stent treatment area. 10. The patient has a carotid artery stenosis determined by ultrasound or angiography (visual estimate) to be: ⱖ50% for symptomatic patients; or ⱖ80% for asymptomatic patients 11. High-risk inclusion criteria for either anatomical or comorbid risk factors are present. The patient must fulfill at least one of the criteria, category I or II risk factors, listed below to meet the inclusion criteria: Category I: Anatomic risk factors ● Previous carotid endarterectomy (CEA) with significant restenosis (as defined for symptomatic or asymptomatic patients) ● Previous radiation treatment to the neck or radical neck dissection ● Target lesion is at or above the second vertebral body C2 (level of jaw) ● Inability to extend the head due to cervical arthritis or other cervical disorders ● Tracheostomy or tracheal stoma ● Laryngectomy ● Contralateral laryngeal nerve palsy Category II: Comorbid risk factors ● Total occlusion of the contralateral carotid artery ● Left ventricular ejection fraction ⬍35% ● Congestive heart failure NYHA functional class III or higher ● Dialysis-dependent renal failure ● Canadian Cardiovascular Society angina classification III or higher or unstable angina ● Requires simultaneous or staged coronary artery bypass surgery, cardiac valve surgery, peripheral vascular surgery, or abdominal aortic aneurysm repair within 60 days ● Aged ⱖ 80 years ● Myocardial infarction within previous 6 weeks ● Abnormal stress test. Treadmill, thallium or dobutamine echo are acceptable. The stress tests should be sufficiently abnormal to place the patient at increased risk for CEA ● Severe pulmonary disease, including at least one of the following: requires chronic oxygen therapy; resting PO2 ⱕ60 mm Hg, hematocrit ⱖ50%, FEV1 or DLCO ⱕ50% of normal DLCO, Diffusion capacity of the lung for carbon monoxide; FEV1, forced expiratory volume in 1 second; NYHA, New York Heart Association; PO2, partial pressure of oxygen.

JOURNAL OF VASCULAR SURGERY Volume 55, Number 4

Matsumura et al 976.e2

Appendix Table I (online only). B, Exclusion criteria 1. The patient is participating in another investigational trial that would interfere with the conduct or result of this study. 2. The patient has dementia or a neurological illness that may confound the neurological evaluation. 3. Total occlusion of the target vessel. 4. There is an existing, previously placed stent in the target artery. 5. The patient has a known life-threatening allergy to the contrast media that cannot be treated. 7. The patient has a gastrointestinal bleed that would interfere with antiplatelet therapy. 8. The patient has known cardiac sources of emboli. 9. The patient has hemoglobin less than 8 g/dL (unless on dialysis), platelet count ⬍50,000/mm3, or known heparin-associated thrombocytopenia. 10. The patient has a history of bleeding diathesis or coagulopathy, including thrombocytopenia or an inability to receive heparin or bivalirudin (Angiomax) in amounts sufficient to maintain an activated clotting time of ⬎250, or will refuse blood transfusion. 11. The patient has atherosclerotic disease involving adjoining vessels that precludes safe placement of the guiding catheter or sheath. 12. The patient has other abnormal angiographic findings that indicate the patient is at risk for a stroke due to a problem other than that of the target lesion, such as ipsilateral arterial stenosis greater in severity than the target lesion, cerebral aneurysm, or arteriovenous malformation of the cerebral vasculature. 13. There is evidence of a carotid artery dissection prior to the initiation of the index procedure. 14. There is an angiographically visible thrombus. 15. There is any condition that precludes proper angiographic assessment, placement of the cerebral protective system, or makes percutaneous arterial access unsafe, eg, morbid obesity, sustained systolic blood pressure ⬎180 mm Hg, tortuosity, occlusive disease, vessel anatomy, or aortic arch anatomy. 16. Occlusion (TIMI 0 flow), or string sign of the ipsilateral common or internal carotid artery. 17. There is evidence of bilateral carotid stenosis that would require intervention within 30 days of the index procedure. 18. There is evidence of a stroke within the previous 30 days of the index procedure. 19. There is a planned treatment of a nontarget lesion within 30 days after the index procedure. 20. There is a history of intracranial hemorrhage within the previous 3 months, including hemorrhagic transformation of an ischemic stroke. 21. There is a history of an ipsilateral stroke with fluctuating neurologic symptoms within 1 year of the index procedure. TIMI, Thrombolysis in myocardial infarction.

JOURNAL OF VASCULAR SURGERY April 2012

976.e3 Matsumura et al

Appendix Table II (online only). A, Preprocedural assessments Test/evaluation General examination Medical history with medication information physical examination Neurologic examinations for each of the following: National Institutes of Health Stroke Scalea Barthel Activities of Daily Living Indexa Modified Rankin scalea Laboratory assessments Creatine kinaseb,c Red blood cell Hemoglobin and hematocrit Platelet count Creatinine Fasting glucose (fasting is defined as a minimum of 8 hours without consuming anything except water) Prothrombin time (must include international normalized ratio) Pregnancy testd 12-lead electrocardiogram Duplex ultrasounde (bilateral) and/or angiogram a

Time frame before index procedure Within 30 days Within 30 days

Within 24 hours Within 30 days

Test/evaluation Neurologic examination National Institutes of Health Stroke Scalea Barthel Activities of Daily Living Indexa Modified Rankin scalea Laboratory tests Red blood cells Hemoglobin and hematocrit Platelet count Creatinine Creatine kinaseb,c 12-lead electrocardiogram Bilateral carotid duplex ultrasoundd Medication review Adverse event documentation and review

Within 24 hours Within 60 days

Must be performed by a neurologist. Cardiac enzyme testing may be done per institution standard of care (ie, troponin). c If the creatine kinase (CK) result is outside of the normal levels, CK-MB levels are required. Samples should be collected at intervals consistent with the institution standard of care. d For women of childbearing potential. e Should be performed at an approved ultrasound laboratory according to the Ultrasound Core Lab Protocol. b

Appendix Table II (online only). B, Postprocedural assessments Time frame Within 24 hours

Before hospital discharge

Before hospital discharge Before hospital discharge Within 7 days after index procedure Before hospital discharge As required

a Must be performed by a neurologist or National Institutes of Health Stroke Scale-certified personnel. b Cardiac enzyme testing may be done per institution standard of care (ie, troponin). c The CK-MB is not required if the total CK value is within normal limits of the site’s laboratory values. If the CK-MB is required, collect samples at 6 to 9 hours, 12 to 16 hours, and 20 to 24 hours (or discharge). The CK-MB should also be obtained for chest pain lasting ⬎15 minutes. d Should be performed at an approved ultrasound laboratory according to the Ultrasound Core Lab Protocol.

JOURNAL OF VASCULAR SURGERY Volume 55, Number 4

Matsumura et al 976.e4

Appendix Table III (online only). Patient enrollment and follow-up Variable Patients’ follow-up completed Patients who did not receive a stent Patients who dieda Patients withdrawn Patients’ disposition unknown Patients lost to follow-up Patients not yet eligible

30 day (n ⫽ 322)

1 year (n ⫽ 317)

2 years (n ⫽ 317)

3 years (n ⫽ 317)

320 5 2 0 0 0 0

282

259

171

21 10 4 2 0

36 13 9 3 0

47 16 3 3 80

a Patients who died and were withdrawn are cumulative counts. “Patients who died” includes patients who died by the end of the visit window and did not have follow-up data for the visit before.

Appendix Table IV (online only). Baseline patient characteristics for long-term follow-up cohort % (n/332) Mean ⫾ SD

95% CI% or Median (min, max)

Age (years)

72.7 ⫾ 9.8 (322)

Age ⱖ 80 years Gender Male Symptomatic Current smoker Coronary artery disease Previous myocardial infarction Congestive heart failure Known left ventricular dysfunction Hyperlipidemia requiring medication Hypertension requiring medication History of cardiac arrhythmia Aortic or mitral valvular disease Previous carotid endarterectomy Current contralateral disease Clinical COPD Diabetes requiring medication History of Peripheral vascular disease Valve replacement CABG Transient ischemic attack Stroke Amaurosis fugax Renal failure Renal insufficiency Liver failure with bleeding diathesis Gastrointestinal bleeding

30.1 (97/322)

71.64, 73.78a 73.6 (39.8, 92.6) 25.16, 35.46b

64.3 (207/322) 11.6 (37/318) 16.8 (54/322) 73.3 (236/322) 30.7 (99/322) 18.3 (59/322) 24.8 (80/322) 87.0 (280/322) 87.9 (283/322) 19.9 (64/322) 13.7 (44/322) 24.5 (79/322) 56.2 (181/322) 18.3 (59/322) 30.1 (97/322)

58.78, 69.52b 8.33, 15.68b 12.86, 21.31b 68.10, 78.05b 25.75, 36.10b 14.25, 22.99b 20.22, 29.94b 82.78, 90.44b 83.82, 91.24b 15.66, 24.66b 10.11, 17.91b 19.93, 29.61b 50.60, 61.71b 14.25, 22.99b 25.16, 35.46b

38.5 (124/322) 1.9 (6/322) 30.1 (97/322) 17.4 (56/322) 17.1 (55/322) 8.1 (26/322) 2.8 (9/322) 21.1 (68/322) 0.0 (0/322) 8.4 (27/322)

33.17, 44.07b 0.69, 4.01b 25.16, 35.46b 13.41, 21.98b 13.13, 21.65b 5.34, 11.61b 1.29, 5.24b 16.79, 25.99b 0.00, 1.14b 5.60, 11.97b

Variable

CABG, Coronary artery bypass grafting; CI, confidence interval; COPD, chronic obstructive pulmonary disease; SD, standard deviation. Four patients’ symptomatic status could not be determined due to insufficient data. a By normal approximation method. b By Clopper-Pearson exact method.

JOURNAL OF VASCULAR SURGERY April 2012

976.e5 Matsumura et al

Event Free Survival

100%

90%

80%

70%

60%

• Censored

50% 0

100

200

300

400

500

600

700

800

900

1000

Days Post Index Procedure Days Post Index Procedure Number at Risk

0 322

Event Free (%) 100%

(0, 30]

(30, 180]

(180, 365] (365, 730]

(730, 1095]

322

319

300

293

238

99.7%

95.9%

94.0%

88.9%

82.4%

Appendix Fig (online only). Kaplan-Meier curve is shown for all subjects with available data for freedom from death between 31 days and 3 years. The event-free rate is 94.0% at 1 year, 88.9% at 2 years, and 82.4% at 3 years.