Author + information
- Received January 7, 2013
- Revision received May 3, 2013
- Accepted May 7, 2013
- Published online October 8, 2013.
- Johannes Lammer, MD∗∗ (, )
- Thomas Zeller, MD†,
- Klaus A. Hausegger, MD‡,
- Philipp J. Schaefer, MD§,
- Manfred Gschwendtner, MD‖,
- Stefan Mueller-Huelsbeck, MD, PhD¶,
- Thomas Rand, MD#,
- Martin Funovics, MD∗,
- Florian Wolf, MD∗,
- Aljoscha Rastan, MD†,
- Michael Gschwandtner, MD∗∗,
- Stefan Puchner, MD∗,
- Robin Ristl, PhD†† and
- Maria Schoder, MD∗
- ∗Departments of Cardiovascular and Interventional Radiology, Medical University Vienna, Vienna, Austria
- †Department of Angiology, Universitaets-Herzzentrum Freiburg-Bad Krozingen, Bad Krozingen, Germany
- ‡Department of Diagnostic and Interventional Radiology, Klinikum Klagenfurt, Klagenfurt, Austria
- §Department of Radiology, University Clinics Schleswig-Holstein, Kiel, Germany
- ‖Department of Diagnostic and Interventional Radiology, Elisabethinen Hospital, Linz, Austria
- ¶Department of Diagnostic and Interventional Radiology, Diakonissen Hospital, Flensburg, Germany
- #Department of Radiology, Hietzing Hospital, Vienna, Austria
- ∗∗Department of Angiology, Medical University Vienna, Vienna, Austria
- ††Center for Medical Statistics, Informatics and Intelligent Systems, Medical University Vienna, Austria
- ↵∗Reprint requests and correspondence:
Dr. Johannes Lammer, Department of Cardiovascular and Interventional Radiology, Vienna General Hospital, Medical University Vienna, Waehringer Guertel 18-20, Vienna A-1090, Austria.
Objectives The hypothesis that endovascular treatment with covered stents has equal risks but higher efficacy than bare-metal stents (BMS) in long femoropopliteal artery disease was tested.
Background Although endovascular treatment of short superficial femoral artery lesions revealed excellent results, efficacy in long lesions remains unsatisfactory.
Methods In a prospective, randomized, single-blind, multicenter study, 141 patients with symptomatic peripheral arterial disease were assigned to treatment with heparin-bonded, covered stents (Viabahn 72 patients) or BMS (69 patients). Clinical outcomes and patency rates were assessed at 1, 6, and 12 months.
Results Mean ± SD lesion length was 19.0 ± 6.3 cm in the Viabahn group and 17.3 ± 6.6 cm in the BMS group. Major complications within 30 days were observed in 1.4%. The 12-month primary patency rates in the Viabahn and BMS groups were: intention-to-treat (ITT) 70.9% (95% confidence interval [CI]: 0.58 to 0.80) and 55.1% (95% CI: 0.41 to 0.67) (log-rank test p = 0.11); treatment per-protocol (TPP) 78.1% (95% CI: 0.65 to 0.86) and 53.5% (95% CI: 0.39 to 0.65) (hazard ratio: 2.23 [95% CI: 1.14 to 4.34) (log-rank test p = 0.009). In lesions ≥20 cm, (TransAtlantic Inter-Society Consensus class D), the 12-month patency rate was significantly longer in VIA patients in the ITT analysis (VIA 71.3% vs. BMS 36.8%; p = 0.01) and the TPP analysis (VIA 73.3% vs. BMS 33.3%; p = 0.004). Freedom from target lesion revascularization was 84.6% for Viabahn (95% CI: 0.72 to 0.91) versus 77.0% for BMS (95% CI: 0.63 to 0.85; p = 0.37). The ankle-brachial index in the Viabahn group significantly increased to 0.94 ± 0.23 compared with the BMS group (0.85 ± 0.23; p < 0.05) at 12 months.
Conclusions This randomized trial in symptomatic patients with peripheral arterial disease who underwent endovascular treatment for long femoropopliteal lesions demonstrated significant clinical and patency benefits for heparin-bonded covered stents compared with BMS in lesions ≥20 cm and for all lesions in the TPP analysis. In the ITT analysis for all lesions, which was flawed by major protocol deviations in 8.5% of the patients, the difference was not significant. (GORE VIABAHN® endoprosthesis with bioactive propaten surface versus bare nitinol stent in the treatment of TASC B, C and D lesions in superficial femoral artery occlusive disease; ISRCTN48164244)
Epidemiological data demonstrate a prevalence as high as 19% in the general population and 29% in patients age >70 years of lower extremity peripheral arterial disease (PAD) in the United States and Europe (1,2). Superficial femoral artery (SFA) stenoses and occlusions are the most common cause of symptomatic PAD (3). Endovascular treatment of short lesions has been recommended by the TransAtlantic Inter-Society Consensus (TASC) and by the American Heart Association (4,5). Randomized controlled trials (RCTs) comparing percutaneous transluminal balloon angioplasty (PTA) and primary stenting in short lesions of the SFA have demonstrated the superiority of primary stent placement (6,7). More recently, drug-eluting balloons and drug-eluting stents have been evaluated in RCTs, although only in short lesions (8–11). With advancing technologies, endovascular recanalization of long lesions (TASC II classes C and D) is technically feasible in a large percentage of patients (12), but the durability of PTA of long SFA lesions remains poor (13,14). Primary stent placement in long SFA lesions has shown promising results in a single-arm study (15). However, bare-metal stents (BMS) do not prevent neointimal hyperplasia, and the risk for in-stent restenosis and stent fracture grows with lesion length (16). An expanded polytetrafluoroethylene (ePTFE) covering of stents prevents in-growth of neointimal tissue regardless of lesion length. Therefore, the use of ePTFE-covered stents in long femoropopliteal artery lesions may increase the patency rate. It has been demonstrated in animal studies that immobilized heparin on ePTFE vascular grafts may reduce platelet deposition and neointimal hyperplasia (17,18).
To evaluate the hypothesis that heparin-bonded ePTFE-covered stents are superior to BMS in patients with symptomatic PAD and long SFA lesions, including the proximal popliteal artery, the VIASTAR (Viabahn Endoprosthesis With PROPATEN Bioactive Surface [VIA] Versus Bare Nitinol Stent in the Treatment of Long Lesions in Superficial Femoral Artery Occlusive Disease) trial was initiated.
The VIASTAR trial is a prospective, randomized, single-blind, multicenter study that was physician initiated. The protocol was developed and conducted in accordance with the International Conference on Harmonisation/Good Clinical Practice Guideline and the Declaration of Helsinki (ISO 14155-1 and ISO 14155-2). It was approved by the local ethics committees and registered at the ISRCTN Register. All patients had to give written informed consent. Patients were recruited in 7 European centers. Randomization was done by using numbered envelopes, and the correct sequence was monitored centrally. Data were recorded on an electronic case report form; monitoring was provided by the coordination center for clinical studies (Koordinierungszentrum fuer Klinische Studien) of the Medical University Vienna, and 12-month results of color-coded Doppler ultrasound (CDUS) examinations were anonymized and blinded before review by CoreLab Bad Krozingen (Bad Krozingen, Germany).
The study hypothesis was that the endovascular treatment of symptomatic long SFA lesions with a heparin-bonded Viabahn endoprosthesis (VIA) or a BMS carries equal risks for adverse events within 30 days after the index procedure, and the 12-month patency rate of the heparin-bonded Viabahn endoprosthesis would be superior by 25%. The ePTFE covering of the Viabahn endoprosthesis may avoid in-growth of neointimal tissue, and the heparin bonding may contribute to the reduction of thrombosis and neointimal hyperplasia.
The major inclusion criteria were symptomatic PAD in the Rutherford-Becker clinical stage 2 to 5 (moderate to severe claudication and critical limb ischemia), de novo arteriosclerotic stenosis or occlusion of the SFA and proximal popliteal artery 10 to 35 cm in length (TASC II classes B, C, and D) (4), patent or successfully treated iliac artery inflow, and outflow of at least 1 tibial artery. The major exclusion criteria were untreated inflow lesions, any previous stenting or surgery in the target artery, serum creatinine level >2.5 mg/dl, septicemia, and known intolerance to heparin, antithrombotic study medications, or contrast agents.
The primary safety endpoint was a composite of serious procedural adverse events, including death, myocardial infarction, study limb amputation, and access site and treatment site complications requiring surgery, blood transfusion, or prolonged hospital stay within 30 days of the index procedure. The primary efficacy endpoint was the primary patency rate measured at 1 year post-procedure. Primary patency was defined as no evidence of restenosis ≥50% or occlusion within the study lesion based on CDUS with a peak systolic velocity ratio ≤2.5 and no target lesion revascularization (TLR) within 12 months (19). Secondary endpoints were clinical improvement according to the Rutherford category at 1, 6, and 12 months, walking distance according to the walking impairment questionnaire (WIQ), change of the ankle-brachial Doppler index (ABI) at discharge and 1, 6, and 12 months, secondary patency, and freedom from TLR.
The intervention was performed percutaneously with an antegrade or crossover femoral artery access under local anesthesia. Successful wire traversal of the target lesion was an entry criterion for the study. Angiography of the target lesion was performed in 2 planes before and after intervention. VIA and BMS were inserted through a 6- or 7-F sheath. In the VIA group, the Viabahn endoprosthesis (WL Gore, Flagstaff, Arizona) with the contoured proximal edge and heparin-bonded surface (PROPATEN Bioactive Surface) was used. The BMS used were the Life-Stent (BARD Peripheral Vascular, Inc., Tempe, Arizona), the Protégé EverFlex Stent (ev3 Inc., Plymouth, Minnesota), and the SMART-Control Stent (Cordis Corporation, Johnson & Johnson, Warren, Massachusetts). These stents were selected because of reported patency rates ≥80% at 1 year (7,20,21). The BMS choice was at the operators' discretion.
As study medication, all patients received 5,000 IU of heparin during the procedure. After treatment, patients were to take 100 mg of aspirin daily for life and 75 mg of clopidogrel (Plavix, Sanofi, Hoechst, Germany) daily for at least 6 months. Most patients started taking clopidogrel 1 or 2 days before the index procedure; for those who did not, a loading dose of 300 mg was given before or during the intervention.
The follow-up patient visits were scheduled at discharge and 1, 6, and 12 months; if patients were symptomatic, these visits included physical examination, WIQ, ABI, and CDUS. In addition, peripheral computed tomography angiography was performed at 12 months in a study subcohort.
It was calculated that 140 patients would need to be enrolled for the study to have a statistical power of 80% to detect an absolute difference of patency of 25% at 12 months. A 2-sided p value of 0.05 was considered to indicate statistical significance. Analysis of the data for the primary and secondary patency rate, and the freedom from TLR, was performed according to the intention-to-treat (ITT) group and the treatment per-protocol (TPP) group. Primary patency, secondary patency, and freedom from TLR rates were calculated by using Kaplan-Meier product-limit estimation, and the log-rank test was used to test for differences between groups. Hazard ratios for primary patency were calculated from a Cox proportional hazards regression model accounting for treatment group and stented length. Metric variables were described by mean ± SD, and the equality between the treatment groups was analyzed by using Student t tests. Categorial variables were described by absolute and relative frequencies, and the homogeneity of distributions between the 2 treatment groups was analyzed by using chi-square tests or, if the test assumption was violated, by the Fisher exact test. Descriptive statistics were calculated for the ITT data. Calculations were performed by using SAS version 9.3 (SAS Institute, Inc., Cary, North Carolina).
From March 2009 to March 2011, a total of 141 patients (100 men, 41 women; mean age 69 ± 9 years; age range 47 to 90 years) were included. Demographic and clinical characteristics were similar, with no significant differences between groups, except in length of the stented segment (Tables 1 and 2). The mean lesion length was 19.0 ± 6.3 cm in the VIA group and 17.3 ± 6.6 cm in the BMS group (p = 0.13).
Acute technical success was achieved in all but 1 patient who had a malposition of a BMS that needed surgical removal followed by bypass. At discharge, ABI had significantly improved in the VIA and BMS groups, from 0.58 ± 0.17 and 0.58 ± 0.16 (p = 0.94) to 0.93 ± 0.18 and 0.96 ± 0.14 (p = 0.24), respectively.
Complications within 30 days after the index procedure were recorded in 20 patients, 11 (15%) in the VIA group and 9 (13%) in the BMS group (Table 2). Severe adverse events within 30 days were observed once in each group: a pseudoaneurysm and hematoma (VIA group), which needed surgery, and a malposition of the stent (BMS group), which was removed surgically, followed by bypass (2 of 141 [1.4% for both groups]). There were no deaths, myocardial infarctions, or study limb amputations. Thus, the hypothesis of the primary safety endpoint was met.
Six patients in each group had to be excluded from ITT analysis because of major protocol violations: 6 due to screening failure (patients with lesions <10 cm and >35 cm, and/or no tibial artery run-off), 4 due to incorrect treatment (PTA only without any device placement, “spot stenting” with <50% covering of the lesion after PTA), 1 who did not take any of the antithrombotic study medication, and 1 patient withdrew the informed consent after discharge (Fig. 1). Of the 141 randomized patients (ITT), 129 patients (VIA 66; BMS 63) were available for the TPP analysis. Causes for exclusion from ITT analysis are shown in Figure 1. At 12 months, 63 of 72 VIA patients and 54 of 69 BMS patients were available for follow-up and 57 of 66 VIA patients and 52 of 63 BMS patients for TPP analysis.
At 1 year, the primary patency rate of all patients, including those with major protocol violations (ITT), was 70.9% (95% confidence interval [CI]: 0.58 to 0.80) versus 55.1% (95% CI: 0.41 to 0.67; log-rank test p = 0.11) in the VIA versus BMS group, respectively. In patients treated according to the study protocol (TPP), ≥50% restenoses were observed in 9 versus 22 patients and occlusions in 6 versus 4 patients in the VIA versus BMS group. Four of 6 VIA occlusions and 2 of 4 BMS occlusions occurred within the first 6 months despite double antiplatelet therapy. The remaining occlusions occurred between 6 and 12 months with aspirin medication only. The VIA occlusions were treated in 3 patients by fibrinolysis and in the other 3 patients by bypass surgery. One patient with an acutely thrombosed VIA presented with acute limb ischemia. The 4 BMS occlusions were treated in 2 patients (1 bypass, 1 fibrinolysis) and remained untreated in the other 2 patients. In the 9 relevant (≥50%) VIA restenoses, edge stenosis at the proximal or distal end was observed. In the 22 relevant (≥50%) restenoses of BMS, diffuse instent restenosis was seen most commonly, but edge stenoses were observed as well. Therefore, at 1 year in patients treated according to the study protocol, the primary patency rate (TPP) was 78.1% (95% CI: 0.65 to 0.86) versus 53.5% (95% CI: 0.39 to 0.65; log-rank test p = 0.009) in the VIA versus BMS group, respectively (Table 3, Figs. 2A and 2B). Thus, within the first year after the index procedure, the risk of binary restenosis, occlusion, and TLR was 2.23 (95% CI: 1.15 to 4.35) times more likely when using a BMS. The hazard ratio of BMS versus VIA calculated from a multiple Cox proportional hazards model accounting for treatment group and length of treated segment was 2.71 (95% CI: 1.36 to 5.39). Length of stented segment was included in the model because significant differences in this variable were observed between treatment groups. Freedom from clinically driven TLR was 84.6% (95% CI: 0.72 to 0.91) in the VIA group versus 77.0% (95% CI: 0.63 to 0.85) in the BMS group at 1-year follow-up (log-rank test p = 0.37). The secondary patency rate (TPP) after TLR and bypass surgery was 89.9% (95% CI: 0.78 to 0.95) versus 75.2% (95% CI: 0.61 to 0.84) in the VIA versus BMS group (log-rank test p = 0.058).
A stratified analysis was performed to evaluate the effect of TASC classification and lesion length on primary patency. At 12 months, patients with long lesions (≥20 cm, TASC class D) had a significantly higher primary patency rate when treated with Viabahn endoprosthesis at ITT analysis (VIA 71.3% [95% CI: 0.53 to 0.83]; BMS 36.8% [95% CI: 0.17 to 0.57]; log-rank test p = 0.01) and at TPP analysis (VIA 73.3% [95% CI: 0.54 to 0.85]; BMS 33.3% [95% CI: 0.13 to 0.54]; log-rank test p = 0.004) (Figs. 3A and 3B). In the group with lesion length <20 cm, a trend toward higher patency rate after VIA was seen (VIA 86.4%; BMS 63.0%), but the difference was not significant (log-rank test p = 0.106). In a Cox proportional hazards regression model, sex, clinical stage according to the Rutherford category, diabetes, renal failure, calcification severity, and stenosis versus occlusion had no significant influence on outcomes.
The ABI (at discharge 0.93 ± 0.18 vs. 0.97 ± 0.14) of the VIA group versus BMS group was 0.96 ± 0.19 vs. 0.98 ± 0.14 [p = NS], 0.94 ± 0.24 vs. 0.87 ± 0.24, and 0.94 ± 0.23 vs. 0.85 ± 0.23 [p < 0.05] at 1, 6, and 12 months, respectively (Table 3). An improvement by ≥1 Rutherford category was reported in 84.0% in both groups at 12 months (Fig. 4A). The mean walking distance according to the WIQ was 136.3 m versus 115.1 m (p = 0.13) at baseline and 785.8 m versus 565.9 m (p = 0.17) in the VIA versus BMS cohort at 12 months, respectively (Fig. 4B). These data include patients after TLR and bypass.
The purpose of the current study was to evaluate whether a heparin-bonded ePTFE-covered endoprosthesis such as the VIA is superior to bare-metal nitinol stents (BMS) in patients with symptomatic PAD and long lesions of the SFA and proximal popliteal artery. The ITT analysis did not show a significant difference but a trend in favor of the heparin-bonded ePTFE-covered endoprosthesis (VIA 71%; BMS 55%; log-rank test p = 0.11). However, it was flawed by major protocol deviations in 8.5% of the patients. In the TPP analysis, the Viabahn endoprosthesis showed a primary patency rate of 78.1% versus 53.5% (p = 0.009) for BMS at 12 months. Thus, within the first year after the index procedure, the risk of a binary restenosis, occlusion, and TLR was 2.23 times more likely when using a BMS. This confirmed the original study hypothesis. The fact that the TPP was significantly at odds with the ITT analysis testifies that the rate of protocol violations was high and diluted the statistical power. In lesions ≥20 cm (TASC class D), the 12-month patency rate was significantly higher in VIA patients in the ITT analysis (VIA 71.3% vs. BMS 36.8%; p = 0.01) and the TPP analysis (VIA 73.3% vs. BMS 33.3%; p = 0.004). Compared with baseline, ABI was improved in both study cohorts with a higher ABI in the Viabahn cohort of 0.94 ± 0.23 versus a BMS ABI of 0.85 ± 0.23 (p < 0.05) at 12 months. Thus, objective parameters showed a significant benefit. However, this improved primary patency and ABI for the Viabahn endoprosthesis group did not translate into a significant reduction of clinically driven TLR at 12 months (84.6% after Viabahn endoprosthesis versus 77.0% after BMS; log-rank test p = 0.37). The relatively low clinically driven TLR rate in the BMS group in relation to the high rate of instent restenosis is difficult to explain. One could argue that in-stent restenosis in BMS still enables collateral flow through side branches, which may mitigate symptoms. However, 70% of the lesions in the BMS group were total occlusions without side branches. The lower TLR rate may be due to the slow progressive narrowing of the stent lumen due to diffuse intimal hyperplasia in BMS. An additional post-interventional exercise effect may have caused less obvious recurrence of claudication. In the VIA group, the recurrence was in one-third of the patients due to a total thrombotic occlusion that caused a rapid deterioration of the walking distance. In 1 of 6 patients, the thrombotic occlusion caused an acute limb ischemia. The number of occlusions in the VIA group was higher (6 vs. 4); this finding, however, was not significant. The VIA occlusions were treated in 3 patients by fibrinolysis and in the other 3 patients by bypass. The observed mean walking distance according to the WIQ was longer in the VIA cohort at 12 months but did not reach statistical significance (785.8 m vs. 565.9 m; p = 0.17). The clinical improvement of ≥1 according to the Rutherford classification was 84.0% in both groups. However, the reported ABI, walking distance, and clinical improvement are parameters that correspond to the secondary patency rate because they also include patients after TLR and bypass.
The endovascular treatment of PAD has the advantages of low morbidity and mortality rates, and early recovery, and it can be done as an outpatient procedure. BMS, drug-eluting stents, and drug-eluting balloons have demonstrated their superiority over plain PTA in short lesions only (6–11). However, a prospective RCT for long lesions focusing on various endovascular techniques has not yet been published. In a prospective single-arm study with BMS in long SFA lesions, 1-year freedom from TLR was reported in 68.2% (15). This finding was even lower than the freedom of BMS from TLR in the current study (77%). The freedom from TLR in the Viabahn arm of this study was 85%.
Bypass surgery with venous conduits must still be considered as the gold standard when comparing primary patency rates. However, considering the high secondary patency rate of 89.9% in the Viabahn arm of this study, it was demonstrated that similar results compared with surgery could be achieved at a low procedural risk. In the current study, the rate of severe complications requiring surgery or prolonged hospital stay was 1.4%, and no deaths or study limb amputations were observed. Thus, the overall risk compared with surgery seems to be less (3,22). In the event that no vein is available or that the vein should be preserved for aorto-coronary bypass surgery, the 1-year patency rate after synthetic graft femoropopliteal above-knee bypass drops to 76%, as has been shown in a recent RCT (23).
Covered stents have the advantage of no in-growth of neointimal tissue, which is the drawback of BMS. However, graft thrombosis as disadvantage has been observed. In this study, the Viabahn endoprosthesis with the heparin-bonded surface (PROPATEN bioactive surface) was used (24,25). The stably bound, endpoint immobilized heparin at the inner surface of the ePTFE graft has been shown in the Scandinavian Propaten trial to significantly prolong the primary patency rate to 86.4% for heparin-bonded ePTFE grafts versus 79.9% for plain ePTFE grafts (p = 0.043) at 1 year (26). All patients were on antiplatelet medication with aspirin 100 mg daily for life and clopidogrel 75 mg daily for at least 6 months, which may be necessary to achieve a low thrombosis rate.
Twelve (8.5%) of the 141 patients had a major protocol violation. There were 6 violations in each study arm, and they were randomly distributed.
A strength of the study is that it is the first published RCT in symptomatic PAD patients and endovascular treatment of long femoropopliteal artery lesions. Clinical and imaging endpoints were both considered, and both study endpoints were met.
The clinical implication of the current study is that it demonstrates, when treating PAD in patients with long diffuse femoropopliteal artery disease, heparin-bonded ePTFE-covered stents yield clinical and patency benefits compared with BMS.
The authors thank Mrs. Johanna Moyses, research coordinator at the division of Cardiovascular and Interventional Radiology, Medical University Vienna, for the data management, WL Gore for the financial support for the CoreLab review of the 12-month CDUS studies, and the Medical University Vienna for monitoring by the Clinical Coordination Center Koordinierungszentrum fuer Klinische Studien.
Dr. Lammer is a member of the scientific advisory board of Abbott Vascular, Boston Scientific, and WL Gore; and has received lecture fees from Medtronic, WL Gore, and Terumo. Dr. Zeller is a member of the scientific advisory board of Boston Scientific, Medtronic-Invatec, WL Gore, Angioslide, Medtronic-Ardian, and Covidien-ev3; and has received lecture fees from sanofi-aventis, CR Bard, J&J Cordis, Covidien-ev3, Boston Scientific, Straub Medical, Invatec, Biotronik, and Pathway Medical. Dr. Mueller-Huelsbeck has served as a consultant for Boston Scientific; workshop organizer and lecturer for Termuo; and presented educational cases for Abbott Vascular. Dr. Funovics has received lecture fees from Bolten Medical. Dr. Schoder has received lecture fees from J&J Cordis and Biotronik. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- ankle-brachial index
- bare-metal stent(s)
- confidence interval
- color-coded Doppler ultrasound
- expanded polytetrafluoroethylene
- peripheral arterial disease
- percutaneous transluminal balloon angioplasty
- randomized controlled trial
- superficial femoral artery
- TransAtlantic Inter-Society Consensus
- target lesion revascularization
- treatment per-protocol
- Viabahn endoprosthesis
- walking impairment questionnaire
- Received January 7, 2013.
- Revision received May 3, 2013.
- Accepted May 7, 2013.
- American College of Cardiology Foundation
- Criqui M.H.,
- Fronek A.,
- Barrett-Connor E.,
- Klauber M.R.,
- Gabriel S.,
- Goodman D.
- Hirsch A.T.,
- Haskal Z.J.,
- Hertzer N.R.,
- et al.
- Laird J.R.,
- Katzen B.T.,
- Scheinert D.,
- et al.,
- RESILIENT Investigators
- Werk M.,
- Langner S.,
- Reinkensmeier B.,
- et al.
- Dake M.D.,
- Ansel G.M.,
- Jaff M.R.,
- et al.,
- Zilver PTX Investigators
- Stabile E.,
- Virga V.,
- Salemme L.,
- et al.
- Scheinert D.,
- Scheinert S.,
- Sax J.,
- et al.
- Mohler E.R. 3rd.,
- Gornik H.L.,
- Gerhard-Herman M.,
- et al.
- Bosiers M.,
- Torsello G.,
- Gissler H.M.,
- et al.
- McQuade K.,
- Gable D.,
- Pearl G.,
- Theune B.,
- Black S.
- Lindholt J.S.,
- Gottschalksen B.,
- Johannesen N.,
- et al.