MATERIALS AND METHODS:

Transcription

1 Clinical Studies Randomized, Multicenter Study Comparing Expanded Polytetrafluoroethylene covered Endoprosthesis Placement with Percutaneous Transluminal Angioplasty in the Treatment of Superficial Femoral Artery Occlusive Disease Richard R. Saxon, MD, Michael D. Dake, MD, Robert L. Volgelzang, MD, Barry T. Katzen, MD, and Gary J. Becker, MD PURPOSE: To compare the safety and effectiveness of the Viabahn endoprosthesis with that of percutaneous transluminal angioplasty (PTA) alone in the treatment of symptomatic peripheral arterial disease (PAD) affecting the superficial femoral artery (SFA). MATERIALS AND METHODS: From 1998 to 1999, patients with symptomatic SFA PAD were enrolled in a prospective, multicenter randomized study and underwent either PTA alone n ( 100) or PTA followed by stent-graft placement (expanded polytetrafluoroethylene/nitinol self-expanding stent-graft) n ( 97) for stenoses or occlusions of the SFA that were 13 cm long or shorter. At baseline, there were no significant differences between the PTA an stent-graft treatment groups, including chronic limb ischemia status and treated lesion length. RESULTS: The stent-graft group had a significantly higher technical success rate (95% vs P 66%, <.0001) and 1-year primary vessel patency rate at duplex ultrasonography (65% vs 40%, P.0003). A patency benefit was seen for lesions at least 3 cm long. At 12 months, chronic limb ischemia status was 15% further improved for the stent-graft P group.003). There were no significant differences between treatment groups with regard to the occurrence of early or late major adverse events. CONCLUSIONS: In this multicenter study, the patency, technical success, and clinical status results obtained with stent-grafts were superior to those obtained with PTA alone. J Vasc Interv Radiol 2008; 19: Abbreviations: ABI ankle-brachial index, FPA femoropopliteal artery, PAD peripheral arterial disease, PTA percutaneous transluminal angioplasty, SFA superficial femoral artery, TASC I Transatlantic Inter-Society Consensus I PERIPHERAL arterial disease (PAD) to increase as the population ages andabetes mellitus increases. Most cases is a worldwide problem that is likelythe prevalence of hypertension and di-of PAD involve the lower limbs, especially the femoropopliteal artery (FPA) segment, in which atherosclerotic stenoses or occlusions develop early in the disease process. Although patients From the Department of Diagnostic Imaging and Interventional Radiology, San Diego Cardiac and Vascu- Gore & Associates. may initially have no symptoms, over lar Institute, Tri-City Medical Center, 4002 Vista Way, Written on behalf of the other Study Principal Investigators listed by order of enrollment: Gerald mild claudication to tissue loss neces- time symptoms can progress from Oceanside, CA (R.R.S.), University of Virginia, Grubbs, Allen Meglin, James Swischuk, Gordon Charlottesville, Virginia, (M.D.D.), Northwestern Memorial Hospital, Chicago, Illinois (R.L.V.), Baptist Carsitating amputation (1). The most common treatments for PAD include life- McLean, Zvonimir Krajcer, Thomas Naslund, Katherine Krol, Andrew Blum, Rodney Raabe, Michael Stecker, Karen Ehrman, Michael Meyerovitz, style modifications (diet and exercise), diac and Vascular Institute, Miami, Florida (B.T.K.), and University of Arizona College of Medicine, Tucson, Arizona (G.J.B.). Received July 22, 2007; final re- medical therapy, bypass surgery using Hal Folander, Michael Hallisey, John Laird, Scott Schultz, Charles McGlade, Ted Chambers, Williams vision received February 2, 2008; accepted February 4, Moore, Joseph Galichia, and James Lutz. an autologous saphenous vein or a Address correspondence to R.R.S.; prosthetic graft, and endovascular rsaxon5@cox.net SIR, 2008 therapy. Although endovascular treatment of PAD in the superficial R.R.S., M.D.D., and B.T.K. are consultants for W.L. DOI: /j.jvir femoral 823

2 824 PTFE-covered Endoprosthesis in Treatment of SFA Occlusive Disease June 2008 JVIR artery (SFA) is often the invasive treatment of choice due to decreased morbidity relative to surgery (2), there are little data establishing that any of the various newer options (eg, atherectomy, cryoplasty, subintimal angioplasty, laser therapy, stent-grafts) leads to any improvement in patency over PTA alone. Because of a myriad of choices and little comparative level 1 data, it is hard to make a case for any definitive or correct technique to use for any given lesion. PTA has been used as the endovascular standard therapy to treat FPA lesions in patients with chronic limb ischemia, although results have been variable and somewhat disappointing. One-year restenosis rates have ranged from 25% to 80% (3 7). PTA treatment failure has been associated with hypertension, diabetes, poor run-off, renal insufficiency, multiple segmental lesions, long lesions, more distal lesions, residual stenosis, and Transatlantic Inter-Society Consensus I (TASC I) grade B to D lesions (8 11). The results of endovascular treatment of the FPAs supplementing PTA with implantation of metallic stents have also been variable. Results of metaanalyses and randomized studies indicate that stent placement provides somewhat higher long-term patency rates; however, there are major issues with in-stent restenosis and loss of patency over time (5,12,13). Metallic stents covered with polyester or expanded polytetrafluoroethylene were developed with the aim of improving long-term patency and reducing complications. Initial results with early stent-grafts in the SFA were poor with respect to both patency and complications, especially for devices covered with polyester (4,14,15). However, several subsequent nonrandomized studies showed that substantially better results could be achieved in the SFA or FPA with self-expanding nitinol stents covered with expanded polytetrafluoroethylene (Viabahn [formerly Hemobahn] Endoprosthesis; W.L. Gore, Flagstaff, Arizona) (16 23). One-year primary patency rates in these investigations ranged from 58% to 93%. Herein, we report the 1-year results from the multicenter, randomized premarket approval trial that led to approval of the Viabahn endoprosthesis for use in the SFA by the U.S. Food and Drug Administration in June This study compared PTA alone and with use of stent-grafts in the treatment of SFA disease. Single-center results (28 patients) from this trial, showing a 2-year primary patency rate of 87% in the stent-graft group (n 15) and 25% in the PTA group (n 13, P.002), were previously published (24,25). The purpose of this study was to compare the safety and effectiveness of the Viabahn stent-graft with that of PTA alone in patients with chronic lower limb ischemia or chronic lifestyle-altering claudication due to SFA atherosclerotic disease. MATERIALS AND METHODS A total of 244 limbs in 241 patients were treated at 25 investigational sites in the United States between June 1998 and December The study was conducted in accordance with the principles outlined in the Declaration of Helsinki and with approval from the institutional review boards of the participating centers. All enrolled patients provided informed consent for their participation. Patients Patients considered for enrollment had documented de novo or restenotic atherosclerotic or occlusive lesion(s) of the SFA up to 13 cm in length (vessel diameter ranged from 4.8 to 12 mm; maximum diameter, 7.8 mm) that caused either chronic lifestyle-altering claudication or chronic lower limb ischemia. Exclusion criteria included previous bypass surgery involving the treatment site, planned or concurrent lower limb bypass surgery, intolerance to oral antiplatelet therapy or heparin, bleeding disorders, renal failure, bacteremia, a lesion within 0.5 cm of the profunda femoris artery origin, previous stent implantation in the target lesion, and fewer than one continuously patent run-off infrapopliteal artery with stenosis of less than or equal to 50% diameter. Before enrollment, resting and exercise ankle-brachial indexes (ABIs) were obtained and a walking assessment was performed. After a baseline arteriogram and preliminary PTA with a residual stenosis of less than or equal to 50% luminal diameter narrowing, patients were enrolled in the trial. A total of 244 limbs in 241 patients were treated at 25 investigational sites between June 1998 and December For the purposes of this study, the term cases refers to treated limbs. Each site was permitted up to two training cases. A total of 47 training cases were studied; 197 cases were randomized, with 100 assigned to PTA and 97 to stent-graft placement. None of the investigators had any experience with the Viabahn stent-graft before this trial (because this was the first time it was available in the United States). Each investigative site implanted two training cases to become comfortable with the device before randomizing patients. As summarized in Tables 1 and, 2 with the exception of sex and preprocedure resting ABI, the demographic, risk factor, and PAD characteristics were similar for both PTA and stent-graft cases. There were significantly more men than women in the study (P.05) but no significant baseline differences between the PTA and stent-graft treatment groups with regard to cardiovascular risk factors, chronic limb ischemia status, treated lesion length (mean standard deviation, 7 cm 4 for both groups), and percentage stenosis (mean, 81% 17 and 77% 18, respectively); resting ABIs were and , respectively (P.005). The distribution of Rutherford- Becker chronic limb ischemia scores (26) was comparable between groups. Eighty-eight percent (88/100) of patients in the PTA group and 91% (87/ 96) of patients in the stent-graft group had claudication (ischemia scores 3). Twelve percent (12/100) of patients in the PTA group and 9% (9/96) of patients in the stent-graft group had chronic limb ischemia scores of 4 or 5 (rest pain or minor tissue loss). No patients had major tissue loss. Lesion lengths were similar between groups, and there were a similar number of occlusions (results according to lesion length are shown in Tables 3 and ). 4 TASC I and II had not yet been written at the time of this study, and TASC II was not yet available when this manuscript was submitted. References to TASC I categories are included, but lesions were not classified according to TASC I category because this sys-

3 Volume 19 Number 6 Saxon et al 825 Table 1 Summary of Demographic Characteristics and Cardiovascular Risk Factors Characteristic PTA Group (n 100) Stent-Graft Group (n 97) P Value Mean age (y)* (40 84) (46 88) NS M:F ratio 70/30 80/17.05 Smokers 51 (51) 45 (46) NS Coronary artery disease 46 (46) 48 (49) NS Previous myocardial infarction 30 (30) 23 (24) NS Congestive heart failure 8 (8) 14 (14) NS Arrhythmia 12 (12) 15 (15) NS Angina 20 (20) 24 (25) NS Stroke 7 (7) 10 (10) NS Hypertension 68 (68) 65 (65) NS Diabetes mellitus 34 (34) 36 (37) NS Renal disease 10 (10) 6 (6) NS Hyperlipidemia 59 (59) 58 (60) NS Note. Unless otherwise indicated, data are given as numbers of patients and numbers in parentheses are percentages. NS not significant. * Numbers in parentheses are the range. Table 2 Characteristics of PAD in the Two Treatment Groups Characteristic PTA Group (n 100) Stent-Graft Group (n 97) P Value Patent run-off vessels to ankle 1 30 (30) 25 (27) NS 2 40 (40) 45 (46) NS 3 30 (30) 26 (27) NS 1 lesion at treatment site 32 (32) 42 (43) NS Mean lesion length (cm)* NS (0.4 13) (0.5 13) Mean stenosis at narrowest site (%)* NS (0 100) Mean preprocedure resting ABI* ( ) (21 100) ( ) Note. Unless otherwise indicated, numbers in parentheses are percentages. NS not significant. * Numbers in parentheses are ranges. Table 3 Technical Success Rates according to Length of Treated Lesion Lesion Length (cm) PTA Group Stent-Graft Group P Value 3 91% (19/21) 95% (18/19) NS % (17/28) 95% (18/19) % (15/21) 93% (27/29) % (11/24) 95% (20/21) % (4/6) 100% (9/9).014 Note. Technical success was defined as completion of treatment with no recovery procedure to the target lesion required the day of the procedure and less than 30% stenosis at posttreatment angiography. Numbers in parentheses are the number of technically successful cases/the total number of lesions in the length range. NS not significant..005 tem was not available at the time of enrollment. Ninety-nine of the 100 treated lesions in the PTA group and 89 of the 97 in the stent-graft group were de novo lesions (P.020). One patient in the PTA group and eight patients in the stent-graft group were treated for restenosis at a previous PTA site. Twenty-nine lesions in the PTA group and 20 in the stent-graft group were occlusions (P.190). Procedure Patients enrolled in the study were prospectively randomized to treatment with stent-graft placement or PTA. Patients randomized to PTA were eligible for additional angioplasty within the lesion to obtain an optimal result if needed. Patients with at least 30% residual stenosis or a significant dissection after additional PTA could have uncovered stents placed as a bail out procedure at the discretion of the operator, but crossover to the endoprosthesis arm was not allowed. In general, PTA was performed via 6-F sheaths. In the stent-graft treatment group, stent-grafts were oversized 5% 20% relative to the native vessel diameter measured angiographically. Devices were placed preferably from a retrograde over-the-bifurcation approach but also from an ategrage approach (at operator discretion). Patients in both the PTA and stent-graft groups empirically received anticoagulation with 5,000 U of heparin during the procedure. Eight-French sheaths were required for stent-graft placement at the time of this study; closure devices were not yet available and were not used. After deployment, an angioplasty balloon with a diameter equal to that of the stent-graft was inflated throughout the entire length of the device. An appropriate oral antiplatelet therapy was administered after the procedure at the discretion of the operator. This included aspirin and occasional ticlopidine. Clopidogrel was not yet available. Patients were not given warfarin sodium unless they had already been taking it before the procedure for other indications. Technical improvements in stent-graft technique that have subsequently been described (24,25) were developed during and after this study and were not known or used consistently by the investigators. For example, avoiding balloon injury outside the device and not overdilating the native vessel to avoid pain after the procedure were concepts that

4 826 PTFE-covered Endoprosthesis in Treatment of SFA Occlusive Disease June 2008 JVIR Table 4 Target Vessel Patency Rates according to Length of Treated Lesion Lesion Length (cm) PTA Group Stent-Graft Group P Value 3 66% (14/21) 68% (13/19) NS (.985) % (11/28) 63% (12/19) % (6/21) 66% (19/29) % (9/24) 67% (14/21) % (1/6) 56% (5/9).031 Note. Numbers in parentheses are the number of patent cases/the total number of lesions in that size range. NS not significant. did not exist when this trial was initiated. Original Study Design This study was originally designed to enroll a maximum of 415 patients and was statistically powered to show a 15% increase in patency in the stentgraft treatment group. The primary efficacy outcome of the study was primary patency at 12 months, which was defined as technical success without interrupted blood flow and no procedures performed within the original treated segment. Additional efficacy outcomes included treatment success, technical success, and clinical success at 12 months. Treatment success was defined as completion of the randomization treatment without an additional recovery procedure or major adverse event, angiographic results demonstrating less than 50% residual stenosis, absence of a dissection longer than the treatment length, and color flow Doppler ultrasonography (US) demonstrating patency of the treated vessel. Technical success was defined as initial treatment success with no major adverse events within 30 days of the procedure and improvement in segmental limb pressure indexes by 0.15 or more relative to that at baseline (before treatment) in the treated limb 30 days after the procedure. Because the definition of primary patency required technical success at 30 days, any major adverse event within 30 days (eg, myocardial infarction) led to a loss of primary patency. Clinical success at 12 months was defined as improvement of at least one category by using the Rutherford-Becker Clinical Status Scale (26). Patients with tissue loss were required to improve by at least two categories and reach the level of claudication to be considered as showing clinical improvement. Vessel occlusions, resting ABIs, chronic limb ischemia scores, and clinical status scores were also recorded. Complications were categorized as mild (easily tolerated sign or symptom), moderate (enough injury or discomfort to cause intolerance with normal activity), severe (incapacitating or requiring substantial intervention), or life-threatening. The study was stopped early because of two issues. First, there was a need to move the manufacturing site of the device and to make delivery system modifications. The original delivery system had an olive on the front end that was occasionally difficult to remove after deployment. Second, it was recognized that the initial study design was flawed, primarily due to the endpoint definition of primary patency. The study was stopped with only 50% of the originally planned patients enrolled. It was recognized that requiring ABI improvement by 0.15 over baseline to maintain primary patency was a study definition that could not be met. Many patients did not comply or did not have exercise ABIs obtained at follow-up. Despite high resting ABIs, they were not more than 0.15 greater than the preprocedure value obtained at rest. Because many patients with patent treatment sites at duplex US and clinical status improvement lost patency with this original study definition, it was believed that this original primary patency definition was flawed. Revised Study Endpoints and Definitions In an effort to address this shortcoming of study endpoint definitions, technical success and target vessel patency at 1 year were redefined to more accurately reflect current medical practice (27). Technical success was redefined as successful completion of the planned or randomized treatment with no rescue procedure to the target lesion required on the day of the procedure and posttreatment angiographic results demonstrating less than a 30% residual diameter stenosis. Target vessel patency at 1 year was defined as no target vessel revascularization procedure and no evidence of restenosis or occlusion within the originally treated vessel on the basis of centrally read color flow Doppler US videotapes (occlusion and restenosis are defined as no color flow or at least a doubling of focal peak systolic velocity, respectively). On the basis of the study data collected, it was not possible to identify the precise location of the original target lesion site in all PTA cases; thus, a more conservative assessment of vessel patency (from SFA origin to the popliteal artery above the knee) was applied. Thus, patency was lost not only for restenosis in the target lesion but also for loss of patency throughout the target vessel. Follow-up Patients were assessed at hospital discharge and 1, 6, and 12 months after the revascularization procedure. Color flow Doppler US and measurement of ABI were performed at baseline and at each follow-up visit. The Rutherford-Becker category and clinical status change were determined at 1-, 6-, and 12-month follow-up. Change in clinical status was designated as improved (Rutherford- Becker score, 3 to 1), no change (Rutherford-Becker score, 0), or worse (Rutherford-Becker score, 1 to 3). Adverse events were recorded throughout the year of follow-up. Statistical Analysis Statistical analyses included the Fisher exact test for nonordinal categorical variables, the Wilcoxon rank-sum test for continuous and ordinal variables, and the Kaplan-Meier method with the log-rank test. All analyses were performed on an intention-totreat basis with use of software (SAS version 8.2; SAS, Cary, North Caro-

5 Volume 19 Number 6 Saxon et al 827 chronic limb ischemia status was significantly better in the stent-graft group. The clinical success rate was significantly higher for stent-graft cases than for PTA cases (84% (81/97) vs 69% (69/100), respectively; P.025). Adverse Events Figure. lina). A P value of less than.05 was considered to be indicative of a statistically significant difference. RESULTS Primary target vessel patency during follow-up by Kaplan-Meier analysis. The intention-to-treat procedure was executed successfully in 96 of the 97 stent-graft cases (99%) and 89 of the 100 PTA-only cases (89%); this difference was statistically significant (P.0005). The single procedure failure in the stent-graft group involved a device deployment failure that resulted in a surgical bypass. The leading olive on the delivery system became trapped, and the operator elected to convert to an open repair. The Viabahn deployment system was subsequently modified substantially in an effort to improve safety. This deployment system is no longer used with the commercially available device. In the PTA group, procedural failures were attributed to eight flow-limiting dissections and three acute occlusions necessitating additional treatment. These were treated with placement of an uncovered stent. Crossover to the stent-graft arm of the trial was not allowed. Technical Success As detailed in Table 3, the rate of technical success was significantly higher in the stent-graft group than in the PTA group (95% vs 66%, P.0001). In the subgroup analysis, the technical success of stent-graft cases failed to reach statistical superiority over PTA-only cases when lesions shorter than 3 cm were treated. Target Vessel Patency During the 1-year follow-up period, 29 patients in the PTA group and 15 in the stent-graft group were censored from the study because of patientor investigator-requested withdrawal, noncompliance with the study protocol, loss to follow-up because of moving away, an adverse event, or death. The numbers of patients examined at hospital discharge and 1, 6, and 12 months after treatment were 99, 96, 82, and 69, respectively, in the PTA group (original, n 100) and 95, 92, 76, and 78, respectively, in the stent-graft group (original, n 97). The primary patency rate of the target vessel at 1 year was significantly higher in the stent-graft group than in the PTA group (65% vs 40%, P.0003). Results of Kaplan-Meier analysis of target vessel patency during follow-up is shown in the Figure. Subgroup analysis of patency according to lesion length showed no difference in treatment groups for lesions shorter than 3 cm but increasing benefit of the stent-graft with increasing lesion length (Table 4). There were six occlusions in the PTA group and eight in the stent-graft group during the year after the revascularization procedure (P.593). Clinical Status The chronic limb ischemia categories before the revascularization procedure and 6 and 12 months after are shown in Table 5. There were no differences between the PTA or stentgraft groups before or 1 month after therapy. At 6 and 12 months, however, Table 6 shows the major adverse events that occurred early ( 30 days) and late ( 30 days to 1 year) during follow-up. There were 21 major adverse events in the PTA group and 20 in the stent-graft group. Three patients died during the study, but none of the deaths were related to the randomly assigned revascularization treatment. None of the differences between the two groups with regard to the number of serious adverse events were significant. The most common minor adverse events were early hematoma and pain; these were observed primarily in the stent-graft group, but the difference did not reach statistical significance (13 cases of hematoma in the stent-graft group vs seven in the PTA group, P.161); these were minor hematomas not requiring transfusion or surgical intervention. Thigh pain occurred in 10 cases in the stent-graft group and three in the PTA group (P.047). Pain was transient and, in general, resolved in less than 2 months. This is thought by the authors to be secondary to overexpansion of the SFA because balloons and devices as much as 20% larger than the target vessel diameter were sometimes used. Additional Outcomes Resting ABIs increased in both study groups after revascularization. Mean values for the PTA and stentgraft groups were 0.94 and 0.97, respectively, at discharge, 0.85 and 0.95 at 6 months, and 0.89 and 0.94 at 1 year. The values at 6 months and 1 year were significantly higher in the stent-graft group than in the PTA group (P.0008 and P.0253, respectively); however, the mean preprocedural ABIs in the stent-graft group were not significantly higher than those in the PTA group (P.0815). Although planned to be part of routine follow-up in the study design, exercise ABIs were rarely obtained after the intervention due to lack of pa-

6 828 PTFE-covered Endoprosthesis in Treatment of SFA Occlusive Disease June 2008 JVIR Table 5 Chronic Limb Ischemia Status during Follow-up Chronic Limb Ischemia Category* PTA Group Stent-Graft Group P Value Before procedure 0 1/100 (1) 1/96 (1) NS 1 7/100 (7) 11/96 (12) NS 2 33/100 (33) 26/96 (27) NS 3 47/100 (47) 49/96 (51) NS 4 5/100 (5) 4/96 (4) NS 5 7/100 (7) 5/96 (5) NS 6 mo after procedure 0 29/78 (37) 39/70 (56) /78 (19) 14/70 (20) /78 (26) 11/70 (16) /78 (17) 5/70 (7) /78 (0) 0/70 (0) NS 5 0/78 (0) 1/70 (1) NS 6 1/78 (1) 0/70 (0) NS 1 y after procedure 0 25/67 (37) 37/72 (51) /67 (19) 17/72 (24) /67 (28) 15/72 (21) /67 (15) 3/72 (4) (0) 0 (0) NS 5 0 (0) 0 (0) NS 6 0 (0) 0 (0) NS Note. Numbers in parentheses are percentages. NS not significant. * Ischemia was graded with the Rutherford-Becker classification (26). tient and investigator compliance. Therefore, only resting ABIs were available before and after intervention for comparison. In the PTA arm, 21 patients had less than 0.15 improvement in resting ABI, although in nine patients the resting ABI was greater than 0.90 at follow-up. In the stentgraft arm, 37 patients had less than 0.15 improvement in resting ABI, although in 25 patients the resting ABI was more than 0.90 at follow-up. DISCUSSION To our knowledge, this is the first prospective, multicenter, randomized pre-market approval trial in the United States to show significant improvement in 1-year primary patency with an implanted device (stent or stent-graft) compared to PTA in the SFA (Figure, P.003). Technical success and clinical outcomes were also significantly better in the stent-graft group than in the PTA group. Throughout the follow-up period, there were no significant differences between the PTA and stent-graft groups with regard to the number of serious adverse events. Definitions of clinical status and clinical success were not changed from the original study protocol. Placement of stentgrafts was associated with a 9% improvement in clinical status and a 15% improvement in chronic limb ischemia score relative to PTA at 1 year. Therefore, the results of this multicenter trial confirm both a clinical utility and patency benefit to the Viabahn endoprosthesis over PTA in SFA occlusive disease. Enrollment in this study was terminated in The decision to terminate the study was not due to concerns about device safety or efficacy but rather due to problems with the original study outcome definitions and study hypothesis. In addition, the study was stopped to allow modifications to be made to the delivery system that markedly decreased the size of the olive on the front end of the delivery catheter. Occasionally, with the original delivery system, the olive would get caught on the leading end of the stent-graft after deployment, making it difficult to remove the delivery system from the patient. The currently available device deploys in an entirely different fashion, and this deployment problem has been fixed. One hundred ninety-seven patients were enrolled randomly in the study, approximately half the intended number. Patients who had been enrolled before study termination completed 1-year follow-up. The original definitions of both primary patency and technical success had shortcomings and were not consistent with current standards. Eventually, it was determined that the trial might yield significant information if an analysis was done with use of more uniformly accepted definitions of technical success and primary patency (based on duplex US follow-up). The final data analysis was completed in late 2004 as part of a pre-market approval submission to the Food and Drug Administration by using both the original trial definitions and the refined definitions of primary vessel patency and technical success. The original primary study endpoint was a complicated composite variable including treatment success, technical success, and patency of the vessel at the lesion site. According to the study definitions, patency was lost with the occurrence of adverse events and failure to increase ABI by at least 0.15 over that at baseline. Many of the enrolled patients had claudication and high baseline resting ABIs ( 0.85). These patients had a significant decrease in their ABIs with exercise on a treadmill, which allowed enrollment in the study. However, most patients were doing well after treatment and refused further exercise studies or these studies were not performed for other reasons during follow-up. Therefore, resting ABIs had to be compared at follow-up to the preprocedure resting ABI. This situation, by definition, led to patency being lost due to a failure to increase ABI by at least 0.15 above the preprocedure level, even though the vessel was widely patent at 1 year and symptoms were resolved. In fact, nine patients in the PTA group and 25 in the stentgraft group had a resting ABI of at least 0.9 at 1 year with patent vessels at duplex US but lost patency with the original trial definition. Moreover, adverse events up to 30 days after the procedure unrelated to patency (ie, myocardial infarction, access site com-

7 Volume 19 Number 6 Saxon et al 829 Table 6 Early and Late Major Adverse Events Adverse Event plications) occurred that also resulted in loss of patency with the original trial definition regardless of patient s clinical status. To correct this problem, the redefined primary patency was based solely on duplex US and angiographic follow-up data. This study reports primary target vessel patency as opposed to binary in-stent restenosis based patency or target lesion based patency. Because the precise location of the treated PTA lesions was not recorded in a manner that was anatomically defined to allow accurate reproducibility, target lesion patency could not be determined with duplex US. This necessitated a comparative statistical analysis of primary target vessel patency. Stenoses within the target vessel but outside the treated lesion were, by definition, considered a loss of patency. Due to the diffuse nature of SFA occlusive disease, we believe target vessel patency rates are as much as 10% 20% lower than target lesion patency rates in the PTA Group (n 100) Stent-Graft Group (n 97) Early ( 30 d after procedure) Death* 0 1 Amputation 1 0 Arrhythmia 0 1 Other cardiac event 3 0 Distal embolization 1 4 Hematoma 0 1 Infection 0 1 Pulmonary effusion 0 1 Stroke 0 1 False aneurysm 0 1 Late ( 30 d to 1 y after procedure) Death* 2 0 Amputation 1 1 Myocardial infarction 4 1 Arrhythmia 2 0 Congestive heart failure 0 1 Other cardiac event 2 2 Distal embolism 1 0 Infection 1 0 Pain 0 1 Pulmonary embolism 1 0 Renal failure 0 2 Stroke 1 1 Other 1 0 Note. There were no statistically significant differences between the two groups. * The causes of death were sepsis (one early event) and myocardial infarction (two late events). Other cardiac event an event other than arrhythmia, congestive heart failure, or myocardial infarction. SFA (28,29). The 1-year primary target vessel patency rate in both treatment groups (65% and 40% for the stentgraft and PTA groups, respectively) is lower than that which would have been calculated if target lesion binary restenosis rates could have been assessed. Consequently, the primary target vessel patency rates in this study cannot be directly compared with those of other trials that do not use the same benchmark. However, we believe this analysis is perhaps more clinically relevant than the use of target lesion patency. Target vessel patency is what matters to the patient s long-term clinical outcome. Historically, endovascular techniques have had better primary patency when treating short, stenotic lesions compared with longer, more diffuse lesions or occlusions. The target vessel patency of the PTA group in this study follows this trend. Because endovascular techniques tend to have poor patency in longer lesions, the TASC I working group recommended that complex lesions longer than 5 cm not be treated with an endovascular approach (5). Due to many technical developments since TASC I, the new TASC II guidelines have expanded the role of endovascular therapy but still recommend surgical bypass for complex lesions and those longer than 15 cm (30). Because PAD in the FPA is often quite extensive at presentation, following TASC II recommendations means a large number of patients with FPA disease will be treated with a surgical bypass. Although many might disagree with this conservative approach to the use of endovascular treatment, it is based on the lack of level 1 data confirming a role for endovascular techniques in the treatment of longer FPA lesions. In this study, the patency outcome for PTA and stent-graft placement was the same for lesions shorter than 3 cm; for lesions longer than 3 cm, there was a statistically significant increase in target vessel patency in the stent-graft group. Moreover, target vessel patency of lesions in the stent-graft group was independent of lesion length (Table 4). More recent singlecenter experience has consistently reported the same finding, with longterm data out to 4 years showing primary patency rates of approximately 60% for patients with FPA stenoses longer than 10 cm (31 33). One recent study of FPA stent-grafts specifically analyzed patency according to lesion length and found no statistically significant difference (34). In fact, 1-year primary patency was more than 90% for lesions longer than 12 cm and was superior to the patency of lesions shorter than 12 cm; however, this difference in patency was not maintained at 4 years. This finding contradicts the conventional wisdom that endovascular techniques are limited by lesion length and calls into question the TASC II guidelines. In fact, stent-grafts might be most useful in patients with long-segment SFA disease. Although more long-term multicenter and comparative data would be helpful, we believe one can now treat FPA lesions longer than 10 cm with the Viabahn stent-graft with a reasonable expectation of durable patency that rivals previously reported data for above-knee synthetic FPA bypass (provided there is adequate vessel

8 830 PTFE-covered Endoprosthesis in Treatment of SFA Occlusive Disease June 2008 JVIR size, inflow, and outflow) (35 38). A randomized, single-center trial by Kedora et al (32) comparing surgical bypass and Viabahn stent-graft placement for long lesions (average length 25 cm in the stent-graft group) showed no difference in patency at 1 year. Moreover, the morbidity and mortality with endovascular treatment is significantly lower than that with surgical bypass (32 39). Therefore, the use of this device in lieu of surgery in selected patients with severe clinical symptoms and TASC II B to D lesions may now be warranted. Although this study addresses a very relevant issue (PTA vs stent-graft placement), the SFA treatment landscape today almost 10 years after initiation of this trial has changed substantially. Surgical bypass, nitinol stents and stent-grafts (with varying mechanical characteristics), and PTA are now augmented with subintimal angioplasty, cutting balloon angioplasty, directional atherectomy, laser atherectomy, and cryoplasty. Drugeluting and biodegradable stents for the SFA are also under development. Arguably, most of these treatment modalities may have a place in SFA treatment; the question is what place do they have and how do they compare? Completed studies indicate that nitinol stents provide superior results compared with stainless steel stents; however, in short lesions ( 3 cm) PTA and stents perform about the same. Moreover, nitinol stents have improved patency in cases of failed PTA and for moderate-length lesions of 3 10 cm but are limited by decreased patency in longer lesions and by stent fracture with overlapping devices (40 45). Device integrity and flexibility have become a major issue in the FPA. Although no radiographic follow-up was performed in this study, no stent fractures were identified and there was no loss of patency due to stent fracture. The flexibility of the Viabahn device enables substantial axial loading and compression. It appears well suited to the mechanical forces of the FPA. Recently, more flexible spiral nitinol stent designs have been developed. One large randomized study using such a design demonstrated a benefit to nitinol stents relative to angioplasty alone for long lesion lengths of approximately 13 cm (41). Even though this trial confirmed a patency benefit over PTA, primary stent patency was still limited, with a binary in-stent restenosis rate of 33% and 46 % at 1 year and 2 years, respectively (42). Primary target vessel patency could not be calculated from the published data. No studies comparing the Viabahn stent-graft to nitinol stents have yet to be completed. The VIBRANT U.S. multicenter trial comparing stentgrafts to nitinol stents for lesions longer than 8 cm has completed enrollment and will, it is hoped, help determine the relative role of these two promising techniques. In short to moderate-length lesions, newer nitinol stents may well be preferable to stentgrafts due to smaller sheath sizes, potentially lower rates of adverse events, and ease of use. However, the data are not yet available to make this comparison. Although the Viabahn stent-graft and newer, more flexible nitinol stent designs have both demonstrated improved patency relative to PTA in objective, comparative trials, there are many new techniques (eg, atherectomy and cryoplasty) that may have benefits over permanent implantable devices. However, to our knowledge, these newer techniques have no objective data at this time that can help confirm a benefit over PTA, nitinol stents, or stent-grafts. More prospective, randomized studies comparing these newer interventional treatments also need to be performed. This study has limitations. It was stopped with a smaller patient population than was originally planned. This decision may have limited the power of the study and led to a failure to demonstrate significant differences that may have been present. For example, there were four distal embolizations in the stent-graft arm (4.1%) and only one distal embolization in the PTA arm (1%). Although no significant statistical difference was calculated, we believe the difference may be real. Placement of any endovascular device may have a higher embolization rate than treatment with PTA alone. Changes have been made to the Viabahn delivery system in an attempt to improve ease of use and safety. The new delivery system is far smoother on its outer surface than the system used for this trial. It is important to note that most of the cases complicated by embolization occurred in patients who were enrolled in the trial after thrombolyisis was performed for acute or subacute SFA occlusions. Distal vessel occlusions after embolization either resolved with further lytic therapy or were clinically insignificant. Nonetheless, caution may be warranted when using stent-grafts in such cases. Because some residual thrombus may be present after thrombolysis, we would consider placing a smooth sheath across the lesion before device placement or the use of distal protection devices in such circumstances. If residual thrombus is suspected, further thrombolytic therapy before device placement may also be warranted. Although the difference did not reach statistical significance, the frequency of minor puncture site hematomas not requiring therapy was also higher in the stent-graft group. This may have been due to the use of larger 8-F sheaths and the lack of closure device availability at the time of this study. This is less of an issue currently, because the new Viabahn delivery system can often be placed via 7-F sheaths and closure devices are available. This study was also limited by the lack of an uncovered stent arm that precludes direct comparisons with nitinol stents. Moreover, no cost analysis was performed. Finally, there were a relatively small number of female patients (25%) included in this trial. This is reflective of the distribution of PAD in the population as a whole. Women did not demonstrate as pronounced an advantage as men with respect to treatment success, clinical success, and redefined patency. However, the differences in treatment effects between sexes were not statistically significant, as reflected by the treatment-by-sex interaction. Advances in medical care during the past decade may have a substantial effect on endograft outcome in the SFA. For example, clopidogrel was not available at the time of this trial and even ticlopidine was inconsistently used. Periprocedural antiplatelet therapy is far different at this point in time and could positively affect the outcomes of the trial if it were repeated today. Although this is more likely to disproportionately affect the stentgraft arm of the trial, it is likely to have

9 Volume 19 Number 6 Saxon et al 831 improved patency results in both arms. There was very little experience with SFA endograft use at the beginning of this trial, and lessons have been learned since the trial was completed. We now believe that extending stent-grafts into a relatively diseasefree portion of the FPA will help avoid disease progression and failure outside the device. Moreover, careful attention to not ballooning outside the device may limit vessel injury and the development of edge stenosis (the most common form of restenosis and failure seen after endograft placement). These issues were not clearly understood at the beginning of this trial and may have had an adverse effect on outcome. In conclusion, the results of this multicenter, prospective randomized study with 1 year of follow-up showed that SFA stent-graft placement with the Viabahn endoprosthesis was safe and provided significantly better results than PTA alone with respect to technical success, target vessel patency, and clinical success in the treatment of SFA occlusive disease. The results achieved with stent-grafts in SFA lesions at least 3 cm long were particularly notable, as patency was independent of lesion length. The fact that patency was independent of lesion length, which has been confirmed in other single center trials (32,34), suggests that this device should be considered for use in longer FPA lesions. Studies with longer follow-up are needed to define the durability of stent-graft revascularization in patients with chronic limb ischemia more clearly. In addition, further randomized clinical trials comparing stent-grafts and new bioactive stentgrafts with other endovascular techniques or with bypass surgery are warranted. References 1. Rutherford RB, Baker JD, Ernst C, et al. Recommended standards for reports dealing with lower extremity ischemia: revised version. J Vasc Surg 1997; 26: BASIL trial participants. Bypass versus angioplasty in severe ischaemia of the leg (BASIL): multicenter, randomised controlled trial. Lancet 2005; 366: Becker GJ, Katzen BT, Dake MD. Noncoronary angioplasty. Radiology 1989; 170: Lammer J. Femoropopliteal artery obstructions: from the balloon to the stent-graft. Cardiovasc Intervent Radiol 2001; 24: TASC. Management of peripheral arterial disease (PAD): TransAtlantic Inter-Society Consensus (TASC). J Vasc Surg 2000; 31:S1 S Currie IC, Wakeley CJ, Cole SE, et al. Femoropopliteal angioplasty for severe limb ischemia. Br J Surg 1994; 81: Löfberg AM, Karacagil S, Ljungman C, et al. Percutaneous transluminal angioplasty of the femoropopliteal arteries in limbs with chronic critical lower limb ischemia. J Vasc Surg 2001; 34: Black JH III, LaMuraglia GM, Kwolek CJ, Brewster DC, Watkins MT, Cambria RP. Contemporary results of angioplasty-based infrainguinal percutaneous interventions. J Vasc Surg 2005; 42: Kudo T, Chandra FA, Ahn SS. The effectiveness of percutaneous transluminal angioplasty for the treatment of critical limb ischemia: a 10-year experience. J Vasc Surg 2005; 41: Capek P, McLean GK, Berkowitz HD. Femoropopliteal angioplasty: factors influencing long-term success. Circulation 1991; 83:I Jämsén TS, Manninen HI, Jaakkola PA, Matsi PJ. Long-term outcome of patients with claudication after balloon angioplasty of the femoropopliteal arteries. Radiology 2002; 225: Muradin GS, Bosch JC, Stijnen T, Hunink MG. Balloon dilation and stent implantation for treatment of femoropopliteal arterial disease: metaanalysis. Radiology 2001; 221: Cejna M, Thurnher S, Illiasch H, et al. PTA versus Palmaz stent placement in femoropopliteal artery obstructions: a multicenter prospective randomized study. J Vasc Interv Radiol 2001; 12: Ahmadi R, Schillinger M, Maca T, Minar E. Femoropopliteal arteries: immediate and long-term results with a Dacron-covered stent-graft. Radiology 2002; 223: Henry M, Amor M, Cragg A, et al. Occlusive and aneurysmal peripheral arterial disease: assessment of a stentgraft system. Radiology 1996; 201: Lammer J, Dake MD, Bleyn J, et al. Peripheral arterial obstruction: prospective study of treatment with a transluminally placed self-expanding stent-graft. International Trial Study Group. Radiology 2000; 217: Cragg AH, Dake MD. First report of percutaneous fem/pop endo graft. Radiology 1993; 187: Bauermeister G. Endovascular stentgrafting in the treatment of superficial femoral artery occlusive disease. J Endovasc Ther 2001; 8: Hartung O, Otero A, Dubuc M, et al. Efficacy of Hemobahn in the treatment of superficial femoral artery lesions in patients with acute or critical ischemia: a comparative study with claudicants. Eur J Vasc Endovasc Surg 2005; 30: Railo M, Roth WD, Edgren J, et al. Preliminary results with endoluminal femoropopliteal thrupass. Ann Chir Gynaecol 2001; 90: Jahnke T, Andresen R, Müller-Hülsbeck S, et al. Hemobahn stent-grafts for treatment of femoropopliteal arterial obstructions: midterm results of a prospective trial. J Vasc Interv Radiol 2003; 14: Bray PJ, Robson WJ, Bray AE. Percutaneous treatment of long superficial femoral artery occlusive disease: efficacy of the Hemobahn stent-graft. J Endovasc Ther 2003; 10: Daenens K, Maleux G, Fourneau I, Nevelsteen A. Hemobahn stentgrafts in the treatment of femoropopliteal occlusive disease. J Cardiovasc Surg (Torino) 2005; 46: Saxon RR, Coffman JM, Gooding JM, Natuzzi E, Ponec DJ. Long-term results of eptfe stent-graft versus angioplasty in the femoropopliteal artery: single center experience from a prospective, randomized trial. J Vasc Interv Radiol 2003; 14: Saxon RR, Coffman JM, Gooding JM, Ponec DJ. Endograft use in the femoral and popliteal arteries. Tech Vasc Interv Radiol 2004; 7: Rutherford RB, Becker GJ. Standards for evaluating and reporting the results of surgical and percutaneous therapy for peripheral arterial disease. J Vasc Intervent Radiol 1991; 2: Sacks D, Marinelli DL, Martin LG, Spies JB. Society of Interventional Radiology Technology Assessment Committee. Reporting standards for clinical evaluation of new peripheral arterial revascularization devices. Radiology 2003; 14:S Vroegindeweij D, Tielbeek AV, Buth J, van Kints MJ, Landman GH, Mali WP. Recanalization of femoropopliteal occlusive lesions: a comparison of longterm clinical, color duplex US, and arteriographic follow-up. J Vasc Interv Radiol 1995; 6: Machan L. Drug eluting stents in the infrainguinal circulation. Techn Vasc Interv Radiol 2004; 7:28 32.

10 832 PTFE-covered Endoprosthesis in Treatment of SFA Occlusive Disease June 2008 JVIR 30. Inter-Society Consensus for the Management of Peripheral Arterial Disease (TASC II). J Vasc Surg 2007; 45(suppl A):S5A S67A. 31. Bleyn J. Ten-year follow-up of Hemobahn in SFA occlusions. Presented at Internation Congress XX on Endovascular Interventions, Scottsdale, Arizona, February 9 15, J Endovasc Therapy, 1007 Feb, 14 Suppl 1, I1 I35, Abstracts from the International Congress XX on Endovascular Interventions. 32. Kedora J, Hohmann S, Garrett W, Munschaur C, Theune B, Gable D. Randomized comparison of percutaneous Viabahn Stent-Grafts versus prosthetic femoral-popliteal bypass in the treatment of superficial femoral arterial occlusive disease. J Vasc Surg 2007; 45(1): Fischer M, Schwabe C, Schulte KL. Value of the Hemobahn/Viabahn endoprosthesis in the treatment of long chronic lesions of the superficial femoral artery: 6 years of experience. J Endovasc Ther 2006; 13: Saxon R, Coffman J, Gooding J, Ponce DJ. Long-term patency and clinical outcome of the Viabahn stent-graft for femoropopliteal artery obstructions. J Vasc Interv Radiol 2007; 18: Veith FJ, Gupta SK, Ascer E, et al. Six-year prospective multicenter randomized comparison of autologous saphenous vein and expanded polytetrafluoroethylene grafts in infrainguinal arterial reconstructions. J Vasc Surg 1986; 3: Johnson WC, Lee KK. A comparative evaluation of polytetrafluoroethylene, umbilical vein, and saphenous vein bypass grafts for femoral-popliteal aboveknee revascularization: a prospective randomized Department of Veterans Affairs cooperative study. J Vasc Surg 2000; 32: Burger DH, Kappetein AP, van Bockel JH, Breslau PJ. A prospective randomized trial comparing vein with polytetrafluoroethylene in above-knee femoropopliteal bypass grafting. J Vasc Surg 2000; 32: Green RM, Abbott WM, Matsumoto T, et al. Prosthetic above-knee femoropopliteal bypass grafting: five-year results of a randomized trial. J Vasc Surg 2000; 31: Conrad MF, Cambria RP, Stone HD, et al. Intermediate results of percutaneous endovascular therapy of femoropopliteal occlusive disease: a contemporary series. J Vasc Surg 2006; 44: Duda SH, Bosiers M, Lammer J, et al. Sirolimus-eluting versus bare nitinol stent for obstructive superficial femoral artery disease: the SIROCCO II trial. J Vasc Interv Radiol 2005; 16: Schillinger M, Sabeti S, Loewe C, et al. Balloon angioplasty versus implantation of nitinol stents in the superficial femoral artery. N Engl J Med 2006; 354: Schillinger M, Sabeti S, Dick P, et al. Sustained benefit at 2 years of primary femoropopliteal stenting compared with ballon angioplasty with optional stenting. Circulation 2007; 115: Gray BH, Sullivan TM, Childs MB, et al. High incidence of restenosis/reocclusion of stents in the percutaneous treatment of long-sement superficial femoral artery disease after suboptimal angioplasty. J Vasc Surg 1997; 25: Scheinert D, Scheinert S, Sax J, et al. Prevalence and clinical impact of stent fractures after femoropopliteal stenting. J Am Coll Cardiol 2005; 45: Schlager O, Dick P, Sabeti S, et al. Long-segment SFA stenting: the dark sides in-stent restenosis, clinical deterioration, and stent fractures. J Endovasc Ther 2005; 12: