Author + information
- Received July 13, 2017
- Revision received November 28, 2017
- Accepted December 5, 2017
- Published online May 21, 2018.
- Mahmood K. Razavi, MDa,∗ (, )
- Dennis Donohoe, MDb,
- Ralph B. D’Agostino Jr., PhDc,
- Michael R. Jaff, DOd,
- George Adams, MDe,
- on behalf of the DANCE Investigators
- aSt. Joseph Heart and Vascular Institute, Orange, California
- bIndependent consultant, Durham, North Carolina
- cWake Forest University School of Medicine, Winston-Salem, North Carolina
- dHarvard Medical School, Boston, Massachusetts
- eUniversity of North Carolina, Rex Hospital, Raleigh, North Carolina
- ↵∗Address for correspondence:
Dr. Mahmood K. Razavi, St. Joseph Heart and Vascular Institute, 1100 West Stewart Avenue, Orange, California 92868.
Objectives This study was designed to evaluate outcomes of adventitial dexamethasone delivery adjunctive to standard endovascular revascularization in femoropopliteal peripheral artery disease.
Background Drug-coated balloons and drug-eluting stents improve patency of endovascular interventions with passive diffusion of antiproliferative drugs. Adventitial dexamethasone delivery targets the initial triggers of the inflammatory reaction to injury, thus potentially providing a potent antirestenotic strategy.
Methods The single-arm DANCE (Dexamethasone to the Adventitia to Enhance Clinical Efficacy After Femoropopliteal Revascularization) trial enrolled 262 subjects (283 limbs) with symptomatic peripheral artery disease (Rutherford category 2 to 4) receiving percutaneous transluminal angioplasty (PTA) (n = 124) or atherectomy (ATX) (n = 159) in femoropopliteal lesions ≤15 cm in length. A mixture of dexamethasone/contrast medium (80%/20%) was delivered to the adventitia and perivascular tissues surrounding target lesions in all subjects. Thirty-day assessments included major adverse limb events (MALE) and post-operative death. Twelve-month assessments included primary patency, freedom from clinically driven target lesion revascularization (CD-TLR), Rutherford scoring, and walking impairment questionnaire.
Results At 12 months, primary patency rates in DANCE-ATX and -PTA per-protocol populations were 78.4% (74.8% intent-to-treat [ITT]) and 75.5% (74.3% ITT), respectively. Rates of CD-TLR in DANCE-ATX and -PTA subjects were 10.0% (13.1% ITT) and 11.0% (13.7% ITT), respectively. There were no 30-day MALE + post-operative death events nor 12-month device- or drug-related deaths or MALE.
Conclusions Direct adventitial delivery of dexamethasone appears to be an effective and safe therapy to prevent restenosis. Randomized studies are needed to further test this possibility. (Dexamethasone to the Adventitia to Enhance Clinical Efficacy After Femoropopliteal Revascularization [DANCE]; NCT01983449)
Relatively poor patency has been a limitation of percutaneous transluminal angioplasty (PTA), atherectomy (ATX) and bare-metal stents in patients with peripheral artery disease. To that end, reducing mechanically induced inflammatory injury to the vessel wall and suppression of neointimal hyperplasia with local drug delivery have been the foci of research to improve outcomes (1–3). The cytotoxic antiproliferative agent paclitaxel has been delivered via drug-coated balloons (DCB) and drug-eluting stents (DES) with superior patency rates when compared with uncoated PTA alone (4–6). The addition of new drug delivery tools and techniques to the peripheral interventional armamentarium appears necessary to further reduce restenosis rates.
Vascular inflammation has been strongly linked to the development of atherosclerosis (7,8) and restenosis (9), particularly in peripheral arteries. The adventitia and perivascular tissues surrounding the artery play a major role in the regulation of inflammation, cell recruitment, and cell proliferation after vascular injury (10,11), thus these tissues have become a target for managing inflammation and preventing restenosis after endovascular intervention. The corticosteroid dexamethasone has a potent anti-inflammatory action and reduces the signal cascade that leads to hyperproliferative responses. Thus, it is considered for its potential ability to reduce the signals that lead to restenosis without the cytotoxicity associated with paclitaxel.
Adventitial and perivascular drug delivery through microneedle infusion catheters is proposed as an alternative to DCB and drug-eluting stents. The principal benefit of this approach is that the therapy may be tailored to the patient and the pathogenesis of the disease, because microneedle injection of liquid therapeutic agents is not limited to agents that can be coated onto balloons or stents.
Evidence of clinical benefit by targeting vascular inflammation with adventitial and perivascular drug delivery in a robust clinical evaluation has not previously been available. A single-center pilot study of adventitial dexamethasone delivery using the Bullfrog Micro-Infusion Device (Mercator MedSystems, Emeryville, California) in 20 subjects reported an 81% primary patency rate at 1-year follow-up (12). The DANCE (Dexamethasone to the Adventitia to Enhance Clinical Efficacy After Femoropopliteal Revascularization) trial was subsequently initiated to assess the safety and efficacy of the adventitial delivery of dexamethasone in a larger group of patients with symptomatic femoropopliteal artery disease.
The DANCE trial is a prospective, multicenter, single-arm, open-label study investigating the safety and effectiveness of adventitial drug delivery of dexamethasone (ADD-DEX) to reduce restenosis after femoropopliteal peripheral artery interventions. The drug was delivered using the Bullfrog Micro-Infusion Device (Mercator MedSystems) (Figure 1). The study consisted of 2 pre-specified concurrently enrolling cohorts: DANCE-ATX, in which target lesions were primarily revascularized with any commercially available directional, rotational, orbital or laser atherectomy or debulking device (ATX), and DANCE-PTA, in which target lesions were treated with PTA. Both cohorts allowed for stent placement at the investigators’ discretion. There were no differences in the eligibility criteria between the 2 groups. The primary efficacy endpoint, 12-month primary patency, was compared to contemporary published PTA and DCB treatments from pivotal trials of DCB (4,5).
The DANCE trial protocol was approved by institutional review boards at each site, and all patients provided written informed consent before enrollment. The trial was registered on www.clinicaltrials.gov (NCT01983449), and conducted in accordance with the Declaration of Helsinki, good clinical practice guidelines, and applicable U.S. laws and regulations.
Independent core laboratories analyzed all duplex ultrasonography (VasCore, Boston, Massachusetts) and angiography (Cardiovascular Research Foundation, New York, New York) images. The clinical events committee, consisting of the national principal investigators and an independent medical monitor, reviewed all major adverse events.
Definitions and study endpoints
Infusion success was defined by the visual confirmation of dexamethasone/contrast distribution around the lesion after injection and was graded by the angiographic core laboratory. Revascularization technical failure, defined as >30% residual lumen stenosis, did not disqualify patients from receiving the drug therapy, but was subsequently reviewed by the angiographic core laboratory to determine eligibility for the per protocol population. The per-protocol (PP) analysis excluded: 1) subjects with ≥35% residual stenosis at the end of the case as determined by the angiographic core laboratory, which was 5% higher than the technical failure limit to allow for investigator versus core laboratory margin of error; 2) subjects with target lesions extending into the tibial arteries; and 3) subjects concurrently treated with other drug-eluting products in the same limb. Intent-to-treat (ITT) was defined as any subject who signed a consent form and received the ADD-DEX procedure.
Reported vessel measurements, including lesion length, percent diameter stenosis, rate of occlusions, dissection severity, and calcification scores, were determined by the angiographic core laboratory. Severe calcification was defined by the core laboratory as apparent radiopacity on both sides of the lesion before the revascularization.
The primary safety endpoint was a composite of 30-day major adverse limb events (MALE), defined as major amputation and/or major reintervention requiring surgical bypass of the index limb or thrombolysis/thrombectomy, and post-operative death (POD), defined as any death within 30 days of the index procedure.
The primary efficacy endpoint was 12-month primary patency, defined as a composite of freedom from binary restenosis and clinically driven target lesion revascularization (CD-TLR). Restenosis was determined by: 1) peak systolic velocity ratio of ≥2.4 at the index lesion on duplex ultrasonography or other correlating factors as determined by the core lab; or 2) angiographic evidence of >50% recurrent stenosis per core lab. A TLR was determined to be CD-TLR if a patient experienced worsening of symptoms, deterioration in Rutherford class, or reduction in ankle-brachial or toe-brachial index (ABI/TBI) ≥20% or >0.15 compared with post-procedure ABI/TBI.
Secondary endpoints included ABI and Rutherford-Becker class at each follow-up time visit, walking impairment questionnaire at 6 and 12 months, and success of drug infusion.
Subjects were eligible if they were diagnosed with Rutherford category 2 to 4 peripheral artery disease (moderate-to-severe claudication or ischemic rest pain), had 1 or more angiographic stenoses >70% and lesion lengths ≤15 cm (each as visually assessed by the investigator) in the superficial femoral and popliteal arteries with a reference vessel diameter of 3 to 8 mm, and met all other eligibility criteria. Subjects with more than 1 lesion per limb meeting these criteria had 1 lesion identified at the investigator’s discretion as the target lesion.
At the discretion of the investigator, subjects underwent either ATX with/without PTA or PTA, in either case with or without stent placement, and were enrolled in their respective cohort of the trial. Antithrombotic and antiplatelet medical therapy was administered according to the standard of care at each participating site. ADD-DEX therapy was administered after ATX or PTA as described later in the text. Stent placement, when determined necessary by the investigator, was performed after ADD-DEX and was not considered a technical failure nor TLR.
ADD-DEX: dosage and administration
The study drug consisted of generic dexamethasone sodium phosphate injection, USP, 4 mg/ml solution, which is indicated for local delivery in a variety of conditions to reduce acute inflammation. Adventitial/perivascular administration of the drug was made with the Bullfrog device (Figure 1). Two Bullfrog models (3- to 6-mm and 4- to 8-mm treatment diameters) were used in the study. The drug was combined in a ratio of 80% dexamethasone (4 mg/ml) to 20% contrast, yielding 3.2 mg/ml dexamethasone. The dosage of dexamethasone was lesion-length dependent, with an intended ratio of 1.6-mg drug (0.5 ml) per 1-cm lesion length. The protocol allowed for ±20% variability in the volume administered, based upon anatomic variation. The injection volume (Figure 2) was determined within a pilot study (12) to achieve distribution along and around the target lesion. Multiple infusions (0.2 to 4.0 ml each) with a single device were allowed along the lesion as assessed by the distribution pattern of the admixed contrast (Figure 3). Diffusion of the drug was graded by the angiographic core laboratory on a scale of A through F (Figure 4), as previously described (12). Successful infusions are determined by grade A through C, because drug has been seen to distribute beyond the extent of contrast distribution in animal studies (unpublished data, Juan Granada and Kirk Seward, April 2007).
Follow-up duplex ultrasonography, hemodynamic testing, and assessment of adverse events were performed at 4 weeks, 6 months, and 12 months.
The primary patency at 12 months post-procedure was tested for each cohort against historical (for superiority) and contemporary (for noninferiority) performance goals. The historical and contemporary performance goals were determined by the weighted average of 12-month patency rates of PTA or DCB, respectively, from comparator published, pivotal DCB studies (4,5). On the basis of these studies, a combined primary patency rate of 52.5% was seen with PTA and 72.3% was seen with DCB. The test against the historical performance goal used an exact binomial test for superiority with an alpha level of 0.025. The test against the contemporary performance goal used an exact binomial test for noninferiority with a margin of 10% and alpha level of 0.025 for significance. The noninferiority margin was determined clinically significant based on the observed treatment effect in historical studies (4,5). Subjects with missing 12-month and subsequent data had 6-month data (whether patent or not patent) imputed forward if it was available, whereas subsequent patency determination without TLR was used to adjudicate any missing or alternative data. Sensitivity analysis was performed to determine the effect of missing data, with no changes in the findings. Sample size was chosen to provide sufficient information to assess safety endpoints compared to the published results with current standard of care, and with the intention that >100 enrolled limbs in each group would demonstrate statistical significance with an observed patency rate difference of 16%.
Number and percentage of MALE+POD are summarized. Other parameters are summarized at each time point by descriptive statistics, including number of observations, means, SDs, and 95% confidence intervals by Wilson score method for continuous parameters. For comparison of population statistics and efficacy endpoints between groups, chi-square tests were used for binary endpoints, and 2-sample Student’s t-tests for continuous endpoints. Primary patency and CD-TLR endpoints were also examined using Kaplan-Meier (K-M) time-to-event methods. Unless otherwise indicated, data are presented as mean ± SD and statistical significance was determined by p < 0.05. Statistical analyses were performed using SAS version 9.2 (SAS Institute, Cary, North Carolina).
Baseline and procedural characteristics
Between November 2013 and December 2015, 262 patients for a total of 283 limbs (159 ATX and 124 PTA) were enrolled at 27 U.S. sites. Twenty-one patients received treatments in both limbs during separate procedures (9 bilateral ATX, 11 ATX/PTA, and 1 bilateral PTA). Patient and lesion characteristics are listed in Tables 1 and 2⇓⇓, respectively, and include a comparison to the combined demographic information from the comparator trials (4,5). Limbs excluded from the PP analysis included 19 from the ATX group (2 concurrently treated with a DES, 16 with residual stenosis ≥35% per core lab, and 1 with a target lesion extending into the tibial arteries) and 17 from the PTA group (16 with residual stenosis ≥35% per core lab and 1 with a target lesion extending into the tibial arteries). A small percentage of infusions (4.4%, 7 of 158 ATX and 0.8%, 1 of 124 PTA) were not graded for drug distribution based on low quality or incomplete angiography. Successful infusions (grade A to C) occurred in 98.0% (148 of 151) of ATX and 98.4% (121 of 123) of PTA cases when grading was available, with the balance of infusions graded as moderate success (grade D). No grade F infusions were confirmed.
At 12 months, 123 of 159 limbs remained in the ATX cohort, and 110 of 124 limbs remained in the PTA cohort. Reasons for discontinuation in each cohort (ATX/PTA) included any adverse event (n = 3/1), death (n = 8/2), loss to follow-up (n = 12/4), physician decision (n = 1/1), protocol deviation (n = 3/0), site termination by sponsor (n= 1/0), and withdrawal of consent by the subject (n = 8/6). It is noted that 1 subject death was in a patient with both limbs enrolled, 1 in ATX and 1 in PTA. Thus, the total number of subject deaths was 9 of 262 enrolled.
The overall 12-month primary patency in DANCE was 74.6% (179 of 240 [95% confidence interval (CI): 68.7% to 79.7%], PP: 77.0%, 161 of 209 [95% CI: 70.9% to 82.2%]). The 12-month primary patency in the ATX cohort of DANCE was 74.8% (95 of 127 [95% CI: 66.6% to 81.5%], PP: 78.4%, 87 of 111, [95% CI: 69.8% to 85.0%]), and in the PTA cohort was 74.3% (84 of 113 [95% CI: 65.6% to 81.5%], PP: 75.5%, 74 of 98 [95% CI: 66.1% to 83.0%]). No statistically significant patency differences were observed between PTA and ATX groups. The rate of CD-TLR at 12 months by K-M analysis in ATX subjects was 13.1% (PP: 10.0%), and in PTA subjects was 13.7% (PP: 11.0%). Because the 12-month follow-up endpoint included a window of ±1 month, a 13-month K-M survival analysis was performed to account for all in-window follow-up. Based on the K-M analysis, the point estimate of primary patency at 12 months in the PP group was 84.2% (ATX) and 79.3% (PTA), which declined to 79.6% (ATX) and 76.2% (PTA) at 13 months, whereas CD-TLR rates were 10.0% (ATX) and 11.0% (PTA) at both 12- and 13-month time points, each displayed in Figure 5.
Comparison with historical performance goals
In the primary analysis, both the ATX and PTA DANCE groups (ITT) were superior (p < 0.001) to the 52.5% historical performance goal. In the secondary analysis, both the ATX and PTA DANCE groups were noninferior to the 72.3% contemporary performance goal, whether examining the PP (p < 0.001 and p < 0.004, respectively) or ITT (p = 0.002 and p = 0.005, respectively) population.
There was a statistically significant difference with more stents used in the DANCE trial subjects than in the comparator trials. Post hoc sensitivity analysis of stent use was performed to determine whether the real-world usage of stents in the DANCE trial affected the outcomes. Primary patency rates did not vary significantly (p = NS) in limbs with stents or without, and both stented and nonstented subgroups of each ITT cohort maintained statistical significance versus the historical performance goal (Figure 6).
There were no 30-day device- nor drug-related severe adverse events. All-cause mortality was reported in 8 of 159 (5.0%) ATX and 2 of 124 (1.6%) PTA enrollments, which included 1 death in a subject enrolled in both groups. Causes of death in ATX were 4 cardiac (2.5%), 1 cancer (0.6%), 1 aspiration leading to respiratory failure (0.6%), and 2 unknown (1.3%). Causes of death in PTA were both cardiac-related. MALE were reported in 2 subjects (both bypass). There was no reported stent use related to injury resulting from perivascular drug delivery or use of the Bullfrog catheter, nor were there adverse event reports of vessel dissection related to the procedural use of the Bullfrog catheter. Although the use of dexamethasone is known to increase blood glucose, there was only 1 reported case of post-operative hyperglycemia. This was in a diabetic subject and was controlled with subcutaneous insulin.
Clinical, functional, and hemodynamic outcomes
In the ITT population, 85.5% of subjects had sustained or improved Rutherford scores without target lesion reintervention, and 75.5% of subjects had improvement in Rutherford scores without reintervention from baseline to 12 months. Walking impairment questionnaire scores improved from baseline (46 ± 26% distance, 40 ± 28% speed, 45 ± 29% stairs) to 12 months (69 ± 30% distance, 61 ± 33% speed, 64 ± 32% stairs) (p < 0.001 for each). ABI or TBI improved from baseline (0.77 ± 0.24) to 12 months (0.94 ± 0.25) (p < 0.001).
The results of this prospective, multicenter trial demonstrate the safety and efficacy of direct adventitial delivery of dexamethasone following angioplasty or atherectomy in femoropopliteal lesions, presenting high mid-term primary patency and freedom from CD-TLR rates. In the DANCE trial, the fraction of subjects with severe calcification, popliteal artery involvement, complex lesions (TASCII [Inter-Society Consensus for the Management of Peripheral Arterial Disease] B through D), or advanced disease (Rutherford category 4) were each statistically significantly greater than subjects enrolled into the trials used to generate performance goals (Tables 1 and 2). Each of these risk categories confers a higher risk of restenosis based on published studies (13,14). The use of stents in the DANCE trial was not discouraged and left to the operators’ discretion, which together with the inclusion of more complex disease may explain the higher stent use versus the comparator trials. Despite the inclusion of a population at a higher risk of restenosis, the 12-month primary patency rate of each of the ITT and PP populations within each of the ATX and PTA groups were noninferior to the performance goal derived from results of DCB groups in the comparator trials and statistically superior to the performance goal derived from results of PTA control arms in the comparator trials.
Glucocorticoids have long been considered for the treatment of inflammation and prevention of restenosis after vascular intervention. However, studies of corticosteroids for the reduction of restenosis have produced mixed signals and had limited success. For example, methylprednisolone has been studied with systemic dosing (15), and dexamethasone-eluting stents have been studied (16) in attempts to reduce restenosis. Systemic delivery of 1,000-mg methylprednisolone did not lead to efficacy, likely due to limited local uptake. In the early era of drug-eluting coronary stents, dexamethasone elution of approximately 50 μg/cm of lesion length from a phosphorylcholine-coated stent yielded positive results in comparison to controls (17,18), but the delivery platform caused significant underlying restenosis (16), and dexamethasone was abandoned in search of more potent agents to elute from the limited surface area of stents.
With the focused microinfusion technique described here, dosages in excess of 1.6 mg/cm of lesion length are possible (more than 30 times the dosage concentration delivered from prior dexamethasone coronary stents). This higher dosing ability leads to a potential to produce greater anti-inflammatory concentrations of the drug in the local tissues, thereby interrupting the signal cascade from carrying forward into a proliferative phase.
The favorable outcome observed in this study supports the hypothesis that the control of inflammation in the arterial wall may play an important role in reducing post-intervention restenosis. Dexamethasone is known to down-regulate the production of molecules including monocyte chemoattractive protein (MCP)-1 (19), tumor necrosis factor (TNF)-α (20), interleukin (IL)-10, matrix metalloproteinase (MMP)-9, and nuclear factor-kappa-light-chain-enhancer of activated B-cells (NF-kB) (21,22), which have been linked to inflammatory restenosis in cell culture (23,24).
The use of DCB and DES as intra-arterial drug delivery platforms has the disadvantages of short exposure time and limited contact surface area, respectively. They also encounter challenges with drug crossing calcified atherosclerotic plaques. This has thus far restricted the scope of therapeutic agents to antiproliferative drugs such as paclitaxel and sirolimus analogues. Delivery of agents into the adventitia and perivascular tissues not only avoids the limitations of drug delivery from solid platforms such as balloons and stents, but also targets the tissues that are most involved in the restenosis cascade (7,10) avoiding the uncertainty of whether the therapeutic dose penetrated the atherosclerotic vessel wall containing plaque and calcium. With the ADD-DEX approach, the anti-inflammatory hypothesis to control restenosis with upstream targeting was tested without reformulating dexamethasone from its commercially available formulation.
The limitations of the DANCE trial include the lack of a concurrent, randomized control group to analyze data in a head-to-head fashion. Because comparative trials of atherectomy versus angioplasty have not been adequately powered or blinded (25), the authors relied on meta-analysis (26) describing the lack of difference between angioplasty and atherectomy to justify comparison of the DANCE-ATX data to a performance goal derived from nonatherectomy treatments. The enrollment of a similar group of patients but with greater proportion of subjects at a higher risk for loss of patency after vascular intervention may be weighed against the lack of the randomized control group. The approach of comparing treatment in peripheral interventions to objective performance goals has been previously proposed (27). As with any peripheral interventional therapy, there is a desire for longer-term follow-up to determine whether patency rates and clinical benefits are sustained over time, which could be answered with a robust comparative trial. Lesion grade (>70% stenosis), length, and technical failure (>30% residual stenosis) were visually estimated rather than assessed by quantitative vascular angiography, thus leading to some enrollment outside of the study eligibility criteria based on strict core laboratory examination. Finally, the rate of stent placement was higher in DANCE trial subjects as compared with those in DCB trials for reasons mentioned earlier in the text. Although this might lead to the assumption that stent use was a confounding variable, the sensitivity analysis of the nonstented group of DANCE trial subjects showed consistent improvement over historical performance goal and no statistical difference from the stented group. Thus, the real-world approach to stent placement used in the DANCE trial was not interpreted as affecting the outcome of the study.
This study demonstrates that adventitial delivery of anti-inflammatory drugs may be an effective strategy for maintaining patency after percutaneous revascularization. Despite the inclusion of a high proportion of subjects with complex disease, our results of adventitial dexamethasone injection appear comparable to those of DCB in the femoropopliteal segment. Randomized clinical trials of adventitial drug delivery in this setting are warranted.
WHAT IS KNOWN? Endovascular intervention in the femoropopliteal segment is limited by high restenosis rates mediated by inflammation, cellular migration, and proliferation. Improved patency has been demonstrated in this segment with the addition of localized drug delivery using drug-coated balloons and drug-eluting stents, but there is room for improvements.
WHAT IS NEW? This study showed that direct delivery of therapeutic agents into the vascular wall is feasible and safe, and that targeting vascular wall inflammation by injection of dexamethasone appears to be effective, producing favorable rates of restenosis and target lesion revascularization.
WHAT IS NEXT? Confirmation of these findings through randomized trials is necessary and is currently underway. Furthermore, the hypothesis that multidrug therapy to target various stages of the restenosis cascade further improves outcome may be tested with the drug delivery methodology described here.
The authors thank each of the DANCE site investigators and study staff for their individual and team contributions; Kirk Seward, PhD, for editorial assistance; and Target Health, Inc., for statistical analysis.
The study was funded by Mercator MedSystems, Inc. Dr. Razavi has been a consultant for Abbott Vascular, Biomet/Zimmer, Mercator MedSystems, Spectranetics, Medtronic, Microvention/Terumo, Veniti, and Boston Scientific. Dr. Donohoe has been a clinical and regulatory consultant to Mercator MedSystems; and is a clinical consultant for C.R. Bard. Dr. D’Agostino has been a consultant for Mercator MedSystems, Edwards Lifesciences, Sanofi, Coherus, Puma Pharmaceuticals, and RedHill Biopharma; and has been on the data safety monitoring boards for Medtronic. Dr. Jaff has been an unpaid advisor to Abbott Vascular, Boston Scientific, Cordis, and Medtronic; has been a consultant to Philips Volcano and Venarum; and has equity ownership in PQ Bypass, Vascular Therapies, and EFemoral. Dr. Adams has been a clinical consultant for Mercator MedSystems, Cook Medical, Spectranetics, C.R. Bard, Boston Scientific, Cardiovascular Systems, Inc., and Medtronic.
- Abbreviations and Acronyms
- ankle-brachial index
- adventitial drug delivery
- clinically driven target lesion revascularization
- drug-coated balloon(s)
- drug-eluting stent(s)
- major adverse limb event(s)
- post-operative death
- per protocol
- percutaneous transluminal angioplasty
- toe-brachial index
- Received July 13, 2017.
- Revision received November 28, 2017.
- Accepted December 5, 2017.
- 2018 American College of Cardiology Foundation
- Kokkalis E.,
- Aristokleous N.,
- Houston J.G.
- Tepe G.,
- Laird J.,
- Schneider P.,
- et al.
- Dake M.D.,
- Ansel G.M.,
- Jaff M.R.,
- et al.
- Libby P.
- Michel J.B.,
- Thaunat O.,
- Houard X.,
- Meilhac O.,
- Caligiuri G.,
- Nicoletti A.
- Miller F.J. Jr..
- Owens C.D.,
- Gasper W.J.,
- Walker J.P.,
- Alley H.F.,
- Conte M.S.,
- Grenon S.M.
- Ribichini F.,
- Tomai F.,
- Paloscia L.,
- et al.
- Poon M.,
- Liu B.,
- Taubman M.B.
- Ambler G.K.,
- Radwan R.,
- Hayes P.D.,
- Twine C.P.