Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Meta-analysis of outcomes from drug-eluting stent implantation in femoropopliteal arteries

  • Mingxuan Li ,

    Roles Conceptualization, Data curation, Methodology, Resources, Software, Supervision, Writing – original draft

    limingxuan1011@163.com

    Affiliation Beijing Fengtai You’anmen Hospital, Beijing, China

    ⨯
  • Haixia Tu,

    Roles Resources, Validation, Writing – review & editing

    Affiliation Beijing Fengtai You’anmen Hospital, Beijing, China

    ⨯
  • Yu Yan,

    Roles Resources, Validation

    Affiliation Beijing Fengtai You’anmen Hospital, Beijing, China

    ⨯
  • Zhen Guo,

    Roles Methodology

    Affiliation Beijing Fengtai You’anmen Hospital, Beijing, China

    ⨯
  • Haitao Zhu,

    Roles Data curation

    Affiliation Beijing Fengtai You’anmen Hospital, Beijing, China

    ⨯
  • Jiangliang Niu,

    Roles Methodology

    Affiliation Beijing Fengtai You’anmen Hospital, Beijing, China

    ⨯
  • Mengchen Yin

    Roles Methodology

    Affiliation Beijing Fengtai You’anmen Hospital, Beijing, China

    ⨯

Abstract

Objective

In recent years, studies of drug-eluting stent (DES) for femoropopliteal artery diseases (FPADs) have been gradually published. To explore whether this type of stent is superior to the traditional bare metal stent (BMS), we performed this study.

Methods

A systematic search for randomized controlled trials (RCTs) in Excerpta Medica Database (Embase), PubMed, Web of Science (WOS), and Cochrane Library was performed on November 29, 2022. We innovatively adopted the hazard ratio (HR), the most appropriate indicator, as a measure of the outcomes that fall under the category of time-to-event data. The HRs was extracted directly or indirectly. Then, the meta-analyses using random effects model were performed. The bias risks of included papers were assessed by the Cochrane Risk of Bias 2.0 tool. This study was registered on the PROSPER platform (CRD42023391944) and not funded.

Results

Seven RCTs involving 1,889 participants were found. After pooled analyses, we obtained results without propensity on each of the following 3 outcomes of interest: in-stent restenosis (ISR) -free survival, primary patency (PP) survival, and target lesion revascularization (TLR) -free survival (P >0.05, respectively). Because the results of pooled analyses of the other two outcomes of interest (all-cause death free survival and clinical benefit survival) had high heterogeneity both, they were not accepted by us.

Conclusion

For FPADs, the DES has not yet demonstrated superiority or inferiority to BMS, in the ability to maintain PP, avoid ISR and TLR.

Citation: Li M, Tu H, Yan Y, Guo Z, Zhu H, Niu J, et al. (2023) Meta-analysis of outcomes from drug-eluting stent implantation in femoropopliteal arteries. PLoS ONE 18(9): e0291466. https://doi.org/10.1371/journal.pone.0291466

Editor: Redoy Ranjan, BSMMU: Bangabandhu Sheikh Mujib Medical University, BANGLADESH

Received: July 10, 2023; Accepted: August 25, 2023; Published: September 21, 2023

Copyright: © 2023 Li et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the paper and its Supporting Information files.

Funding: The authors received no specific funding for this work.

Competing interests: The authors have declared that no competing interests exist.

1. Introduction

Femoropopliteal artery disease (FPAD) is the most common peripheral artery disease (PAD) [1]. Nowadays, the endovascular-first strategy is now recommended for the majority of symptomatic FPAD patients [2–4]. The complex anatomical and dynamic challenges (such as compression, shortening, torsion) increase the limitations of endovascular treatment strategy [4–6]. Among numerous endovascular treatment modalities, bare metal stent implantation (BMSI) is one of the most proven for its superiority, especially in comparison with the percutaneous transluminal angioplasty (PTA) i.e., plain old balloon angioplasty (POBA) alone [7–11]. However, the cumulative rate of in-stent restenosis (ISR) at 1 year after BMSI is still ≈30% [7, 9, 12], and that of longer lesions is even ≈50% [12, 13].

After success in the field of coronary artery diseases, drug-eluting stents (DESs) have been gradually applied to femoropopliteal (FP) artery lesions. In the BATTLE trial, Gouëffic et al. reported no statistically significant difference in DES implantation (DESI) regarding the ability to avoid postoperative ISR compared to BMSI (Cox model, P = 0.64) [14]. Whereas in another randomized controlled trial (RCT) published in the same year, a hazard ratio (HR) with a value of 3.57 [95% confidence interval (CI), 1.19–10.00] suggesting the superiority on avoiding ISR of DES was reported by Falkowski et al [15]. A similar condition also occurred between the other two RCTs. Dake et al. reported in 2016 that DES had better primary patency (PP) postoperatively compared with BMS (Log-Rank test, P = 0.03) [16]; however, Mori et al. reported in 2017 that there was no significant difference between them (P = 0.52) [17]. At present, there are few meta-analyses on relevant RCTs. In particular, we have not found any meta-analysis using HR, the most appropriable indicator, as the outcome measure rather than the odds ratio (OR) or the relative risk (RR), for a variety of the postoperative outcomes of FPADs that belong to time-to-event data by type [18].

2. Materials and methods

2.1. Study protocol

According to the PRISMA framework, this study was registered on the PROSPERO platform (CRD42023391944). The data for analyses were from published RCTs and no new participants were included. Therefore, no ethical approvals and consent forms of the participants were required. The PRISMA 2020 checklist is shown in S1 Table.

2.2. Search strategy

A systematic search in Excerpta Medica Database (Embase), PubMed, Web of Science (WOS), and Cochrane Library was performed on December 29, 2022. We searched for all relevant articles without date limit using “eluting”, “stent”, “limb” and all possible synonyms of them. The literature search was independently performed by ML. All entry terms and search commands can be found in the S1 Fig.

2.3. Study selection

We defined the HRs (DESI vs. BMSI) of the following five outcomes of FPADs as the primary outcome measures of interest: ISR-free survival, PP survival, TLR-free survival, all-cause death (ACD) -free survival, and clinical benefit (CB) -free survival. Studies that simultaneously met the following criteria were included by us: 1) the study design was RCT; 2) the language of publication was English; 3) the operation site was human FP artery suffered from FPAD; 4) the number of participants in each arm was not less than 10; 5) there are two independent arms, DESI and BMSI; 6) at least one of the aforementioned HRs was directly reported, or at least one of the Kaplan-Meier (K-M) survival curves for relevant outcomes was given. The FPAD was defined as a symptomatic disease caused by intraluminal atherosclerotic stenosis or occlusion of the FP artery between inguinal ligament and tibial plateau. We did not limit the definition of the above five outcomes. Studies that did not meet the above criteria or that only included complete duplicates of the outcome data available for extraction were excluded by us.

All the retrieved literature information was imported into the Endnote X9 software, followed by duplicate removal and abstract review. Then, the full texts of all available articles that passed the preliminary screening were downloaded and read to identify those that could be finally included into the study. At the stage of assessing the full texts, the bibliographies and citations were also screened for other potential articles. Two authors (HT and YY) independently performed the study selection. Any discrepancies were resolved by consensus.

2.4. Data extraction

After identifying the included studies, we extracted the basic characteristics of the ones and their participants, as well as the primary outcome measures. Directly reported HRs derived from Cox proportional hazard models were preferentially adopted and extracted. If no HRs were found but K-M survival curves were given, we took advantage of the Engauge Digitizer 11.3, an open-source software that can convert graphics to numbers, to transform the information in curves and calculate the HRs [19–21]. Tierney et al. had comprehensively summarized relevant statistical theories and given an HR calculation spreadsheet (Excel format) with preset formula [19]. We used this spreadsheet to calculate HRs, replacing the manual calculation process.

Data extraction was performed by a pair of independent authors (ML and HZ). Any queries and discrepancies were resolved through further discussion to reach a consensus.

2.5. Risk of bias assessment

We assessed the bias risks of included RCTs using the Cochrane Risk of Bias 2.0 (RoB 2.0) tool [22]. It evaluates 5 domains: randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, and selection of the reported results. The risk for each of the 5 domains and overall is described as low, some concerns, or high. The highest risk level among all domains was adopted as the overall assessment result.

Risk of bias assessment was performed by a pair of independent authors (ML and ZG). When the two authors had different opinions on a certain assessment result, the worse one was taken.

2.6. Statistical analysis

Stata (Stata Corp., College Station, TX, United States) version 16.0 was used for all statistical analyses. In each meta-analysis, we took the natural logarithms of the extracted HR value and the maximum and minimum values of its 95% CI respectively for each study, and then included the three obtained variables into the "metan" command. To reduce the error, only a random effects model rather than a fixed effects model was taken, regardless of the degree of heterogeneity among studies [23]. And we corrected the degree of freedom by the restricted maximum likelihood estimation [24]. For any hypothesis test, only the results with a P value less than 0.05 could be considered statistically significant. All pooled analyses were performed independently by ML.

The formula of the Cox proportional hazard model (1) and those for meta-analysis based on extracted or transformed data (2, 3) are shown as follows [19, 20, 25]: (1) (2) (3)

2.7. Heterogeneity assessment and sensitivity analysis

The heterogeneity across the studies was assessed and reported as a percentage using the I2 index value and as a P value using the Cochrane Q test of χ2 [26, 27]. I2 <50% suggests a low heterogeneity, and I2 ≥50% suggests high. P <0.1 for Q test suggests a low heterogeneity, and P ≥0.1 suggests high. Only the models with low heterogeneity suggested by both tests were adopted.

After obtaining a model with low heterogeneity that was included at least 3 studies, the checking calculation was performed by omitting the included studies one by one, to analyze the sensitivity of the model. When a new pooled effect size (ES), i.e., ln (HR) value, which was far away from the ES obtained previously or even beyond the 95% CI range, was obtained from the meta-analysis after omitting one study, this study was considered to bring about instability. In initial plan, once a meta-analysis model had a high heterogeneity and ≥4 studies had been included, we would try to find out the variables that cause heterogeneity by meta regression or empirically, and conduct subgroup meta-analyses according to them.

The above works were performed by a pair of independent authors (HZ and ZG). When the two authors had different opinions on a certain assessment result, the worse one was taken.

2.8. Publication bias assessment

The following methods of publication bias assessment were performed only if no fewer than 3 studies were included in each meta-analysis. One side, the Egger’s test was used [28]. P <0.05 means a high publication bias. And a funnel plot also be drawn [29]. An apparently asymmetric plot with the ES value as axis reflects a high publication bias. The assessment was performed by a pair of independent authors (JN and MY). When the two authors had different opinions on a certain assessment result, the worse one was taken.

2.9. Evidence quality grade assessment

After finishing all meta-analyses, we used the GRADE (Grading of Recommendations Assessment, Development and Evaluation) system to evaluate the qualities of evidence and make recommendations [30]. The evidence derived from meta-analyses is all initially set to high, and respectively graded as high, moderate, low, or very low, finally. Its rating will be lowered by a corresponding number of levels once it appears suspect in terms of overall bias, publication bias, inconsistency, imprecision, or indirectness.

The assessment was performed by a pair of independent authors (ML and HT). When an assessment result was discordant and consensus could not be reached, the one with lower grade was adopted.

3. Results

3.1. Studies and data

We initially identified 1,234 articles by searching 4 academic databases, of which 507 were evaluated after removing duplicates. And 23 articles were retained after the title abstract sieve. After reviewing the full text, 7 RCTs were finally included in this study [14–17, 31–33]. Among them, one [31] reported the pooled, longer-term follow-up results from 2 periods [34, 35] of the SIROCCO trial. The PRISMA flowchart of study selection is shown in Fig 1.

thumbnail
Download:
Fig 1. PRISMA flowchart.

https://doi.org/10.1371/journal.pone.0291466.g001

The publication years of the 7 included RCTs ranged from 2006 to 2022 [31, 33]. There was only one paper was from a single center rather than multiple centers [15]. Three papers only included de novo FP lesions [14, 17, 32], two also included restenosis lesions [16, 31], and two did not mention this aspect of pathological characteristics [15, 33]. The arms in each study were independent with no patient conversions, and the stents implantations were all primary, that is, regardless of whether PTAs had been performed previously or their effects, except for 1 paper [16]. Unlike others, the control arm in the original trial design of this study (Zilver PTX) was PTA with or without BMSI. Among the patients in this arm, those who underwent acute PTA failure were further divided into two subgroups for comparison: DESI and BMSI [16]. This kind of stent implantation was considered as secondary and provisional. The investigators reported data available on outcomes of this subset. To avoid the inclusion of patients who underwent PTA alone, we adopted only the above patients in this study (S2 Fig).

A total of 939 patients underwent DESI and 950 underwent BMSI were included, respectively. Their mean age exceeded at least 65y (from reported data), and their majority was male. More specific main characteristics of these studies and baseline data of patients are shown in Table 1. Except for those not explicitly mentioned, the definitions of outcomes of interest in the included studies were close. The specific definitions are shown in Table 2.

thumbnail
Download:
Table 1. Main characteristics and baselines of the included studies.

https://doi.org/10.1371/journal.pone.0291466.t001

thumbnail
Download:
Table 2. Definitions of the outcomes during follow up extracted from the included studies.

https://doi.org/10.1371/journal.pone.0291466.t002

3.2. Risks of bias

After assessment, the overall risk of bias in 2 included studies was "high" [16, 32], in four it was "some concerns" [14, 15, 17, 31], and in one it was "low" [33]. The detailed final assessment results are shown in Table 3.

thumbnail
Download:
Table 3. Risk bias assessment results of included studies.

https://doi.org/10.1371/journal.pone.0291466.t003

3.3. ISR-free survival

Two studies [14, 15] reported the HRs directly and another [32] gave the K-M survival curve. The result of one [15] was particularly in favor of DESI (HR = 3.57, P = 0.023), whereas those of 2 others [14, 32] were not (P >0.05). The obtained model had a low heterogeneity (I2 = 34.6%, P = 0.217 for Q test), and suggested that there was no statistical difference between the two arms in the risk of postoperative ISR [HR = 1.51 (95% CI, 0.99–2.31), P = 0.058] (Fig 2A). The stability of the model was satisfactory (Fig 2B). The P derived by Egger’s test which value was 0.222 and the visual symmetrical funnel plot, both suggested a low publication bias of the model (Fig 2C).

thumbnail
Download:
Fig 2.

Forest plot of ISR-free survival (random effects model, P = 0.058) (A); Sensitivity analysis of the model assuming that each study is omitted separately [ln (HR)] (B); Funnel plot with pseudo 95% confidence limits (C). Abbreviations: HR, hazard ratio; CI, confidence interval; REML, restricted maximum likelihood; BMSI, bare metal stent implantation; DESI, drug-eluting stent implantation; ISR, in-stent restenosis.

https://doi.org/10.1371/journal.pone.0291466.g002

3.4. PP survival

Three studies [16, 17, 33] gave the K-M survival curves and another [14] reported the HR directly. The result of one [16] of them was particularly in favor of DESI (P = 0.003), whereas those of the other three were not (P >0.05). After pooling those of these 4 studies, we obtained a result without propensity (P = 0.941). However, the high I2 value (66.3%) and the low P value (0.031) for Q test jointly suggested that the model was highly heterogeneous, so it was not accepted by us (Fig 3A). After omitting 3 studies separately, we obtained only one model which had lowest heterogeneity (I2 = 45.0%, P = 0.163 for Q test). The result remained nonsignificant in favor of either DESI or BMSI, that is, HR = 1.28 (95% CI, 0.79–2.06) and P = 0.312 (Fig 3B). We could not perform sensitivity analysis and assessment of publication bias due to too few included studies. The stability of the model was satisfactory (Fig 3C). The Egger’s test yielded a P value of 0.334, and the funnel plot also appeared symmetrical, both suggesting a low publication bias of the obtained model (Fig 3D).

thumbnail
Download:
Fig 3.

Forest plot of PP survival (random effects model with high heterogeneity) (A); Forest plot of PP survival (random effects model with low heterogeneity, P = 0.312) (B); Sensitivity analysis of the model assuming that each study is omitted separately [ln(HR)] (C); Funnel plot with pseudo 95% confidence limits (D). Abbreviations: HR, hazard ratio; CI, confidence interval; REML, restricted maximum likelihood; BMSI, bare metal stent implantation; DESI, drug-eluting stent implantation; PP, primary patency.

https://doi.org/10.1371/journal.pone.0291466.g003

3.5. TLR-free survival

One study that directly reported the HR [14] and other two that gave the K-M survival curves [16, 31] were included in a meta-analysis. All the results of these 3 studies did not support a statistical difference in postoperative TLR risk between BMSI and DESI (P >0.05). So without surprise, the pooled analysis derived a same result [HR = 1.66 (95% CI, 0.93–2.96), P = 0.089] (Fig 4A). An I2 value of 0.0% and a P value for Q test of 0.608 suggested low heterogeneity of the derived model (Fig 4A). Sensitivity analysis suggested high stability of this model (Fig 4B). The model had P = 0.799 for Egger’s test and its funnel plot was roughly symmetrical, which suggested no high publication bias (Fig 4C).

thumbnail
Download:
Fig 4.

Forest plot of TLR-free survival (random effects model, P = 0.089) (A); Sensitivity analysis of the model assuming that each study is omitted separately [ln (HR)] (B); Funnel plot with pseudo 95% confidence limits (C). Abbreviations: HR, hazard ratio; CI, confidence interval; REML, restricted maximum likelihood; BMSI, bare metal stent implantation; DESI, drug-eluting stent implantation; TLR, target lesion revascularization.

https://doi.org/10.1371/journal.pone.0291466.g004

3.6. Survival of ACD-free and CB

There were 2 studies with available data in terms of ACD-free survival. One [14] of them directly reported an HR without propensity (7.3, 95% CI 0.9–59.3), and the other [15] gave a relevant K-M curve with same suggestion (P = 0.576). We pooled the results of the two studies but obtained a highly heterogeneous model (I2 = 75.0%, P = 0.045 for Q test), so we did not accept it (Fig 5A). There were similarly only 2 studies with available data on CB survival: one [14] directly reported an HR without propensity (0.9, 95% CI 0.5–1.6) and the other [16] gave a K-M curve in favor of DESI (P = 0.02). The result of pooled analysis was similarly highly heterogeneous (I2 = 65.7%, P = 0.088 for Q test) and not accepted by us (Fig 5B).

thumbnail
Download:
Fig 5.

Forest plot of ACD survival (random effects model with high heterogeneity) (A); Forest plot of CB survival (random effects model with high heterogeneity) (B). Abbreviations: HR, hazard ratio; CI, confidence interval; REML, restricted maximum likelihood; BMSI, bare metal stent implantation; DESI, drug-eluting stent implantation; ACD, all-cause death; CB, clinical benefit.

https://doi.org/10.1371/journal.pone.0291466.g005

3.7. Evidence quality grade

By pooling the results of RCTs, we obtained 3 corollaries that could be recommended. They revealed that the DESI did not demonstrate a statistically advantage or disadvantage compared to BMSI in terms of ISR-free survival, PP survival, and TLR-free survival for FPADs. After assessment according to the GRADE criteria, their recommendation ratings were all lowered to "low" due to the shortcomings both of "study limits" and "evidence indirectness". Details are detailed in Table 4.

thumbnail
Download:
Table 4. Evidence quality grade assessment of pooled outcomes of interest.

https://doi.org/10.1371/journal.pone.0291466.t004

4. Discussion

This study systematically reviewed and analyzed multiple follow-up outcomes of DESI performed in FP arteries by including 7 RCTs. The results show that, compared to the traditional BMSI, there were no clear superiority or inferiority of DESI in terms of ISR-free survival, PP survival, and TLR-free survival.

For FP lesions, the recognized disadvantage of PTA alone is a high rate of restenosis and concomitant need for TLR [4]. Unlike the coronary, PAD involves longer segments, often at multiple levels with decreased flow rates leading to restenosis even when immediate angiographic results are excellent [11]. To address this limitation, BMSs began to be applied in FPADs in the 1980s [4]. The BMS, represented by nitinol stent, can provide radial support and compression resistance, resulting in extended patency of target lesions [7]. However, stents left in the vascular lumen have brought new problems. The core one is ISR, which reduces the PP and relapses clinical manifestations of ischemia, thus requiring the TLR. Even many studies have reported that the cumulative ISR rate at the first postoperative year after BMSI for FPADs had exceeded 30% [7, 9, 36, 37].

Proliferation of vascular smooth muscle cells (SMCs) is a significant cause of neointimal hyperplasia, which ultimately causes the ISR [38, 39]. The anti-proliferation mechanism of some drugs such as paclitaxel, sirolimus, etc. is theoretically beneficial to antagonize the proliferation of SMCs and reduce the ISR rate. The superiority of stents eluted with the above drugs for coronary arteries has been extensively demonstrated in previous studies [40–43], even for infrapopliteal arteries with similar diameters [44]. It is no surprise that such promising stents have gradually been used in FP lesions. In a retrospective study published in 2016 [45], Soga et al. reported that there were no statistically significant differences (P >0.05) all between DES and BMS on the cumulative incidence of 1-year TLR, major amputation, and other outcomes for FPADs. However, Meng et al. reported in a retrospective study in 2018 that DES had a lower 2-year cumulative ISR rate (0% vs. 23%, P = 0.009) [46]. In recent years, although there have been some meta-analyses on outcomes of our interest, they all used OR as the outcome measure rather than the more appropriate HR [47, 48]. In order to include latest studies and obtain more convincing results through more scientific analysis methods, we have performed this systematic review and meta-analysis of the published RCTs.

There is no doubt that the outcomes of interest in the presented study are time-to-event data all, that is, time is a significant variable that affects the occurrence of outcomes [18]. Because there is no way for the pathophysiological process of atherosclerosis to be completely terminated, restenosis will necessarily occur. So over time, the rate of outcomes such as ISR will gradually increase. For example, for a cohort of FPAD patients, the cumulative ISR rate at the first postoperative month will not be the same as that at the 10th postoperative year. A more straightforward explanation is that a patient will surely die but not necessarily within 1 year postoperatively. Among the numerous indicators including more common OR and RR, HR is the most appropriate measure for the time-to-event outcomes [18, 19]. Another advantage of this is that it is not necessary to consider the time variable, so when selecting which studies would be included in a pooled analysis on a certain outcome, it is no longer limited by the follow-up period, which is helpful to increase the number of included studies. Therefore, we decided to adopt the HR as outcome measure in this study. But a followed practical problem was that RCTs reporting the HRs directly might be few. Fortunately, most of the retrieved RCTs gave the survival curves after analyses of some outcomes, which provided the possibility to extract data. We used the software to extract the survival probabilities at different follow-up moments, and then the HR value with its 95% CI of a certain outcome was derived according to formulas. This method of deriving HR value has high accuracy and stability, and has been widely recognized [19–21]. We think that this may be the first meta-analysis to do so in relevant area.

Pooled analysis suggested no significant difference between the two treatment modalities in their ability to maintain postoperative ISR-free survival for FPADs (P = 0.058). Unlike the results without propensity of the other 2 studies, that of 1 study favored DESI strongly [HR = 3.57 (95% CI, 1.19–10.00)] [15], which led to a result with significant difference almost in the final analysis. The range of 95% CI of HR obtained in this study was wide (1.19–10.00), which led to the reduction of the accuracy of this result, thus making its weight in the pooled analysis small (16%). We note that the mean length of treated target lesions in the BMSI arm in this study was 128 mm, which was subjectively significantly larger than those in the other 2 studies (96 mm and 76 mm). There is indeed evidence that the restenosis rate after BMSI is significantly higher, or even more than 50% at 1st postoperative year, for long FP lesions rather than short ones [12, 13]. However, in a prospective non-randomized trial published in 2017 specifically for long (>15cm) FP lesions, there was also no significant difference between the two treatment modalities (P = 0.62 for Log-Rank test) [49]. Since most of the definitions of PP in the included studies are not the opposite of ISR simply, and even some [14] had not given exact explanation, we regarded the PP survival as an independent outcome. After the three inconsistent but low heterogeneity results were pooled, the derived result suggested that there was no difference between the two stents in the ability to maintain postoperative PP survival. In this analysis, unlike the other two trials, the Zilver PTX trial [16] reported a result favoring DES (P = 0.003). The initial grouping of this study at the design stage was primary DESI versus PTA (with or without different types of stent implantation); however, various subgroups were also analyzed in the analysis stage. The comparison result we could extract were only from the "secondary DESI versus secondary BMSI, after acute PTA failure". This suboptimal stent implantation was the most noticeable difference between this study and the others. Theoretically, DES seems to be more useful for the immediate arterial dissection after PTA than BMS. But unfortunately, we have not found any other RCTs or even other controlled studies that reported similar comparisons. The result of the pooled analysis of TLR survival were not surprising, because the results of the 3 studies included were all without propensity (P >0.05). Notably, the Zilver PTX trail [16] reported a PP survival outcome in favor of DES; but this scenario was not seen with respect to TLR-free survival. This seems to indicate that the TLR is not inevitable after the occurrence of ISR. Many studies have confirmed the benefits of persistent physical exercise on improving quality of life, especially on extending walking distance, so as to avoid TLR [50–52]. We believe that this may be related to its ability to enhance blood circulation compensation rather than just maintain the patency of the target lesion. Of course, this may need to be supported by more evidence. Regrettably, in terms of ACD-free survival and CB survival, due to the high heterogeneity of the studies included in the analyses, the results obtained were not accepted by us. We look forward to more studies.

The longer lesions have been identified as an independent predictor of restenosis [53]. In the previously mentioned prospective study specifically for the long FP lesions [49], not only the ISR-free survival, but also all the other reported outcomes have suggested no statistical differences between the two treatment modalities. Other relevant controlled studies were not retrieved. Several other studies reported a cumulative ISR rate of 30–40% and a cumulative TLR rate of 19–21% at 1st postoperative year of DESI for above lesions [54, 55]. This is clearly no better than the data in one of our included studies, in which the mean length of the included lesions in the BMSI arm was 128 mm and the above cumulative rates at 3rd postoperative year were 35.4% and 31.5%, respectively [15]. In conclusion, the available evidence does not support a superiority of DES compared with BMS for long FP lesions. Another noteworthy special type of FP lesion is the ISR. BMSI is not the accepted default treatment modality for this disease [56, 57]. And we have not found any study specifically comparing DES and BMS in this area. In Zilver PTX single-arm trail, the treatment of 119 FP-ISR lesions with DES had an estimated PP rate of 78.8% at 1st postoperative year [58]. And Tomoi et al. reported that the estimated freedom from recurrent ISR (ReISR) at 2nd DESI postoperative year was 79% in a study published in 2016 [59]. These results are even very close to those of BMSI for de novo FP lesions reported in our included studies [14, 32]. Therefore, we believe that the postoperative patency of DES is better than that of BMS, for FP-ISR lesions.

In recent years, the fluoropolymer-based paclitaxel-eluting stent, Eluvia™, described as "the new generation of DES", has been more and more widely used in clinic. Theoretically, the polymer on this type of stent guarantees longer drug elution period comparing to the polymer-free DES (represented by Zilver® PTX®, that accounted for the majority of the DESs in the included studies), and thus enables the drug to function longer [60]. Moreover, the good biocompatibility of this polymer may make the stent safer than the other early polymer-coated DESs [31, 61–63]. A prospective one-arm trail, CAPSICUM [64], reported a 13% cumulative ISR rate at 1st postoperative year of Eluvia implantation for FP lesions, which seems to be significantly better than the 28% reported by DEBATE trail [32]. A subgroup study from the IMPERIAL RCT [65] reported a 1-year cumulative PP rate of 87%, which was similar to that reported by the Kanagawa PTA trail (84%) [17]. But it is important to note that the former was specifically for long lesions (>140 mm), and the result appears to be better than that of the aforementioned study on the Zilver PTX DES (52%) [39]. Another prospective one-arm trail, MAJESTIC [66], reported a 1-year estimated PP rate of 84%, which also seems to be significantly better than the 75% reported by the BATTLE trail [14]. The results of studies in which two types of DES, Eluvia and Zilver PTX, were directly compared were also similar. Eluvia performed better in both PP survival and TLR-free survival in the IMPERIAL trial (1-year follow-up period, P <0.02) [60]. Soga et al. reported in 2019 that the cumulative mean late lumen loss and ISR rate in the Eluvia arm were lower at 1st postoperative year (P <0.02), but the superiority on cumulative TLR rate was not yet significant (0% vs. 8%, P = 0.08) [67]. And Gray et al. reported in 2022 that, Eluvia was more effective and less costly than Zilver PTX [68]. In 2022, the 1-year results of the EMINENT trial directly comparing DES and BMS were published [33]. Except for that of PP (83% vs 74%, P <0.01), there was no intergroup difference in the 1-year cumulative rates of all outcomes (P >0.05); Moreover, after data conversion and extraction, we found no difference in PP survival between groups (P = 0.312). This has beat to our confidence in Eluvia DES. We look forward to more RCTs about this novel DES.

This presented study has some limitations. First, the number of studies included in each analysis was all limited, which reduced the persuasiveness of the study. If the numbers of included studies (i.e., the sample size) increased, some results in favor of DES might be obtained. Second, although the image information extraction technology was used to increase the sample size and the data conversion was automatically finished with software, the obtained data had been reprocessed after all, which may reduce the statistical validity to a certain extent. Third, due to some objective reasons, the corollaries derived from analyses were all "low", which was the penultimate level after GRADE grading, reducing the credibility of recommendations.

5. Conclusions

The results of this study show that, for FPADs, the DES has not yet demonstrated superiority or inferiority to BMS, in the ability to maintain PP, avoid ISR and TLR.

Supporting information

S1 Table. PRISMA checklist.

https://doi.org/10.1371/journal.pone.0291466.s001

(PDF)

S1 Fig. Search terms used in the literature search on different platforms.

https://doi.org/10.1371/journal.pone.0291466.s002

(TIF)

S2 Fig. Patient selection during data extraction from the Zilver PTX trial [16].

https://doi.org/10.1371/journal.pone.0291466.s003

(TIF)

References

  1. 1. Higgins JP, Altman DG, Gøtzsche PC, Jüni P, Moher D, Oxman AD, et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ. 2011; 343:d5928. pmid:22008217
  2. 2. Norgren L, Hiatt WR, Dormandy JA, Nehler MR, Harris KA, Fowkes FG, et al. Inter-Society Consensus for the Management of Peripheral Arterial Disease (TASC II). J Vasc Surg. 2007; 45(Suppl S):S5–67.
  3. 3. Rooke TW, Hirsch AT, Misra S, Sidawy AN, Beckman JA, Findeiss LK, et al. 2011 ACCF/AHA Focused Update of the Guideline for the Management of Patients With Peripheral Artery Disease (updating the 2005 guideline): a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. J Vasc Surg. 2011; 54(5):e32–58.
  4. 4. Burket MW. Drug-Eluting Stents Are the Default Strategy for Superficial Femoral Artery Intervention Now. Circulation. 2016; 133(3):320–9. pmid:26783275
  5. 5. Wensing PJ, Scholten FG, Buijs PC, Hartkamp MJ, Mali WP, Hillen B. Arterial tortuosity in the femoropopliteal region during knee flexion: a magnetic resonance angiographic study. J Anat. 1995; 187(Pt 1):133–9. pmid:7591974
  6. 6. Cheng CP, Wilson NM, Hallett RL, Herfkens RJ, Taylor CA. In vivo MR angiographic quantification of axial and twisting deformations of the superficial femoral artery resulting from maximum hip and knee flexion. J Vasc Interv Radiol. 2006; 17(6):979–87. pmid:16778231
  7. 7. Schillinger M, Sabeti S, Loewe C, Dick P, Amighi J, Mlekusch W, et al. Balloon angioplasty versus implantation of nitinol stents in the superficial femoral artery. N Engl J Med. 2006; 354(18):1879–88. pmid:16672699
  8. 8. Laird JR, Katzen BT, Scheinert D, Lammer J, Carpenter J, Buchbinder M, et al. Nitinol stent implantation vs. balloon angioplasty for lesions in the superficial femoral and proximal popliteal arteries of patients with claudication: three-year follow-up from the RESILIENT randomized trial. J Endovasc Ther. 2012; 19(1):1–9. pmid:22313193
  9. 9. Dick P, Wallner H, Sabeti S, Loewe C, Mlekusch W, Lammer J, et al. Balloon angioplasty versus stenting with nitinol stents in intermediate length superficial femoral artery lesions. Catheter Cardiovasc Interv. 2009; 74(7):1090–5. pmid:19859954
  10. 10. Schillinger M, Sabeti S, Dick P, Amighi J, Mlekusch W, Schlager O, et al. Sustained benefit at 2 years of primary femoropopliteal stenting compared with balloon angioplasty with optional stenting. Circulation. 2007; 115(21):2745–9. pmid:17502568
  11. 11. Marmagkiolis K, Hakeem A, Choksi N, Al-Hawwas M, Edupuganti MM, Leesar MA, et al. 12-month primary patency rates of contemporary endovascular device therapy for femoro-popliteal occlusive disease in 6,024 patients: beyond balloon angioplasty. Catheter Cardiovasc Interv. 2014; 84(4):555–64. pmid:24740749
  12. 12. Armstrong EJ, Saeed H, Alvandi B, Singh S, Singh GD, Yeo KK, et al. Nitinol self-expanding stents vs. balloon angioplasty for very long femoropopliteal lesions. J Endovasc Ther. 2014; 21(1):34–43. pmid:24502482
  13. 13. Rocha-Singh KJ, Bosiers M, Schultz G, Jaff MR, Mehta M, Matsumura JS, et al. A single stent strategy in patients with lifestyle limiting claudication: 3-year results from the Durability II trial. Catheter Cardiovasc Interv. 2015; 86(1):164–70. pmid:25676568
  14. 14. Gouëffic Y, Sauguet A, Desgranges P, Feugier P, Rosset E, Ducasse E, et al. A Polymer-Free Paclitaxel-Eluting Stent Versus a Bare-Metal Stent for De Novo Femoropopliteal Lesions: The BATTLE Trial. JACC Cardiovasc Interv. 2020; 13(4):447–57.
  15. 15. Falkowski A, Bogacki H, Szemitko M. Assessment of Mortality and Factors Affecting Outcome of Use of Paclitaxel-Coated Stents and Bare Metal Stents in Femoropopliteal PAD. J Clin Med. 2020; 9(7):2221–31.
  16. 16. Dake MD, Ansel GM, Jaff MR, Ohki T, Saxon RR, Smouse HB, et al. Durable Clinical Effectiveness With Paclitaxel-Eluting Stents in the Femoropopliteal Artery: 5-Year Results of the Zilver PTX Randomized Trial. Circulation. 2016; 133(15):1472–83. pmid:26969758
  17. 17. Mori S, Hirano K, Yamauchi Y, Hayashi E, Doijiri T, Takamura T, et al. Penetration rate of the placement of a drug-eluting stent for the treatment of superficial femoral artery lesions in Japan. Heart Vessels. 2017; 32(9):1093–8. pmid:28451834
  18. 18. Tudur C, Williamson PR, Khan S, Best LY. The value of the aggregate data approach in meta-analysis with time-to-event outcomes. Journal of the Royal Statistical Society A. 2001; 164:357–70.
  19. 19. Tierney JF, Stewart LA, Ghersi D, Burdett S, Sydes MR. Practical methods for incorporating summary time-to-event data into meta-analysis. Trials. 2007; 8:16–31. pmid:17555582
  20. 20. Parmar MKB, Torri V, Stewart L. Extracting summary statistics to perform meta-analyses of the published literature for survival endpoints. Statistics in Medicine. 1998; 17(24):2815–34. pmid:9921604
  21. 21. Williamson PR, Smith CT, Hutton JL, Marson AG. Aggregate data meta-analysis with time-to-event outcomes. Stat Med. 2002; 21(22):3337–51. pmid:12407676
  22. 22. Sterne JAC, Savović J, Page MJ, Elbers RG, Blencowe NS, Boutron I, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ. 2019; 366:l4898. pmid:31462531
  23. 23. DerSimonian R, Kacker R. Random-effects model for meta-analysis of clinical trials: an update. Contemp Clin Trials. 2007; 28(2):105–14. pmid:16807131
  24. 24. Lindstrom MJ, Bates DM. Newton-Raphson and EM Algorithms for Linear Mixed-Effects Models for Repeated-Measures Data. J Am Stat Assoc. 1988; 83(404):1014–22.
  25. 25. Cox DR. Regression models and life tables. Journal of the Royal Statistical Society (B). 1972; 34:187.
  26. 26. Thakkinstian A, McElduff P, D’Este C, Duffy D, Attia J. A method for meta-analysis of molecular association studies. Stat Med. 2005; 24(9):1291–306. pmid:15568190
  27. 27. Whitehead A, Whitehead J. A general parametric approach to the meta-analysis of randomized clinical trials. Stat Med. 1991; 10(11):1665–77. pmid:1792461
  28. 28. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997; 315(7109):629–34. pmid:9310563
  29. 29. Biljana M, Jelena M, Branislav J, Milorad R. Bias in meta-analysis and funnel plot asymmetry. Stud Health Technol Inform. 1999; 68:323–8. pmid:10724898
  30. 30. Guyatt GH, Oxman AD, Vist GE, Kunz R, Falck-Ytter Y, Alonso-Coello P, et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ. 2008; 336(7650):924–6. pmid:18436948
  31. 31. Duda SH, Bosiers M, Lammer J, Scheinert D, Zeller T, Oliva V, et al. Drug-eluting and bare nitinol stents for the treatment of atherosclerotic lesions in the superficial femoral artery: long-term results from the SIROCCO trial. J Endovasc Ther. 2006; 13(6):701–10. pmid:17154704
  32. 32. Miura T, Miyashita Y, Soga Y, Hozawa K, Doijiri T, Ikeda U, et al. Drug-Eluting Versus Bare-Metal Stent Implantation With or Without Cilostazol in the Treatment of the Superficial Femoral Artery. Circ Cardiovasc Interv. 2018; 11(8):e006564. pmid:30354784
  33. 33. Gouëffic Y, Torsello G, Zeller T, Esposito G, Vermassen F, Hausegger KA, et al. Efficacy of a Drug-Eluting Stent Versus Bare Metal Stents for Symptomatic Femoropopliteal Peripheral Artery Disease: Primary Results of the EMINENT Randomized Trial. Circulation. 2022; 146(21):1564–76. pmid:36254728
  34. 34. Duda SH, Pusich B, Richter G, Landwehr P, Oliva VL, Tielbeek A, et al. Sirolimus-eluting stents for the treatment of obstructive superficial femoral artery disease: six-month results. Circulation. 2002; 106(12):1505–9. pmid:12234956
  35. 35. Duda SH, Bosiers M, Lammer J, Scheinert D, Zeller T, Tielbeek A, et al. Sirolimus-eluting versus bare nitinol stent for obstructive superficial femoral artery disease: the SIROCCO II trial. J Vasc Interv Radiol. 2005; 16(3):331–8. pmid:15758128
  36. 36. Krankenberg H, Schlüter M, Steinkamp HJ, Bürgelin K, Scheinert D, Schulte KL, et al. Nitinol stent implantation versus percutaneous transluminal angioplasty in superficial femoral artery lesions up to 10 cm in length: the femoral artery stenting trial (FAST). Circulation. 2007; 116(3):285–92. pmid:17592075
  37. 37. Dussaillant GR, Mintz GS, Pichard AD, Kent KM, Satler LF, Popma JJ, et al. Small stent size and intimal hyperplasia contribute to restenosis: a volumetric intravascular ultrasound analysis. J Am Coll Cardiol. 1995; 26(3):720–4. pmid:7642865
  38. 38. Matsumura JS, Yamanouchi D, Goldstein JA, Pollock CW, Bosiers M, Schultz GA, et al. The United States StuDy for EvalUating EndovasculaR TreAtments of Lesions in the Superficial Femoral Artery and Proximal Popliteal By usIng the Protégé EverfLex NitInol STent SYstem II (DURABILITY II). J Vasc Surg. 2013; 58(1):73–83.
  39. 39. Kearney M, Pieczek A, Haley L, Losordo DW, Andres V, Schainfeld R, et al. Histopathology of in-stent restenosis in patients with peripheral artery disease. Circulation. 1997; 95(8):1998–2002. pmid:9133506
  40. 40. Morice MC, Serruys PW, Sousa JE, Fajadet J, Ban Hayashi E, Perin M, et al. A randomized comparison of a sirolimus-eluting stent with a standard stent for coronary revascularization. N Engl J Med. 2002; 346(3):1773–80.
  41. 41. Grube E, Silber S, Hauptmann KE, Mueller R, Buellesfeld L, Gerckens U, et al. TAXUS I: six- and twelve-month results from a randomized, double-blind trial on a slow-release paclitaxel-eluting stent for de novo coronary lesions. Circulation. 2003; 107(1):38–42. pmid:12515740
  42. 42. Otsuka F, Vorpahl M, Nakano M, Foerst J, Newell JB, Sakakura K, et al. Pathology of second-generation everolimus-eluting stents versus first-generation sirolimus- and paclitaxel-eluting stents in humans. Circulation. 2004; 129(2):211–23.
  43. 43. Neumann FJ, Sousa-Uva M, Ahlsson A, Alfonso F, Banning AP, Benedetto U, et al. 2018 ESC/EACTS Guidelines on myocardial revascularization. Eur Heart J. 2019; 40(2):87–165. pmid:30165437
  44. 44. Zhang J, Xu X, Kong J, Xu R, Fan X, Chen J, et al. Systematic Review and Meta-Analysis of Drug-Eluting Balloon and Stent for Infrapopliteal Artery Revascularization. Vasc Endovascular Surg. 2017; 51(2):72–83. pmid:28103754
  45. 45. Soga Y, Takahara M, Iida O, Nakano M, Yamauchi Y, Zen K, et al. Propensity Score Analysis Comparing Clinical Outcomes of Drug-Eluting vs Bare Nitinol Stents in Femoropopliteal Lesions. J Endovasc Ther. 2016; 23(1):33–9. pmid:26763256
  46. 46. Meng FC, Chen PL, Lee CY, Shih CC, Chen IM. Real-World Comparison of Drug-Eluting and Bare-Metal Stents in Superficial Femoral Artery Occlusive Disease with Trans-Atlantic Intersociety Consensus B Lesions: A 2-Year, Single-Institute Study. Acta Cardiol Sin. 2018; 34(2):130–6. pmid:29643698
  47. 47. Ding Y, Zhou M, Wang Y, Cai L, Shi Z. Comparison of Drug-Eluting Stent with Bare-Metal Stent Implantation in Femoropopliteal Artery Disease: A Systematic Review and Meta-Analysis. Ann Vasc Surg. 2018; 50:96–105. pmid:29514049
  48. 48. Varetto G, Gibello L, Boero M, Frola E, Peretti T, Spalla F, et al. Angioplasty or bare metal stent versus drug-eluting endovascular treatment in femoropopliteal artery disease: a systematic review and meta-analysis. J Cardiovasc Surg (Torino). 2019; 60(5):546–56. pmid:31527577
  49. 49. Vent PA, Kaladji A, Davaine JM, Guyomarch B, Chaillou P, Costargent A, et al. Bare Metal Versus Paclitaxel-Eluting Stents for Long Femoropopliteal Lesions: Prospective Cohorts Comparison Using a Propensity Score-Matched Analysis. Ann Vasc Surg. 2017; 43:166–75. pmid:28300676
  50. 50. Fakhry F, van de Luijtgaarden KM, Bax L, den Hoed PT, Hunink MG, Rouwet EV, et al. Supervised walking therapy in patients with intermittent claudication. J Vasc Surg. 2012; 56(4):1132–42. pmid:23026425
  51. 51. Mazari FA, Khan JA, Carradice D, Samuel N, Abdul Rahman MN, Gulati S, et al. Randomized clinical trial of percutaneous transluminal angioplasty, supervised exercise and combined treatment for intermittent claudication due to femoropopliteal arterial disease. Br J Surg. 2012; 99(1):39–48. pmid:22021102
  52. 52. Murphy TP, Cutlip DE, Regensteiner JG, Mohler ER, Cohen DJ, Reynolds MR, et al. Supervised exercise versus primary stenting for claudication resulting from aortoiliac peripheral artery disease: six-month outcomes from the claudication: exercise versus endoluminal revascularization (CLEVER) study. Circulation. 2012; 125(1):130–9. pmid:22090168
  53. 53. Iida O, Takahara M, Soga Y, Nakano M, Yamauchi Y, Zen K, et al. 1-Year Results of the ZEPHYR Registry (Zilver PTX for the Femoral Artery and Proximal Popliteal Artery): Predictors of Restenosis. JACC Cardiovasc Interv. 2015; 8(1):1105–12. pmid:26117463
  54. 54. Zeller T, Rastan A, Macharzina R, Tepe G, Kaspar M, Chavarria J, et al. Drug-coated balloons vs. drug-eluting stents for treatment of long femoropopliteal lesions. J Endovasc Ther. 2014; 21(3):359–68. pmid:24915582
  55. 55. Phillips JA, Falls A, Kolluri R, Whipp A, Collins C, Mohir-Sadaai S, et al. Full Drug-Eluting Stent Jacket: Two-Year Results of a Single-Center Experience With Zilver PTX Stenting for Long Lesions in the Femoropopliteal Arteries. J Endovasc Ther. 2018; 25(3):295–301. pmid:29544372
  56. 56. Ho KJ, Owens CD. Diagnosis, classification, and treatment of femoropopliteal artery in-stent restenosis. J Vasc Surg. 2017; 65(2):545–57. pmid:28126181
  57. 57. Kim W, Choi D. Treatment of Femoropopliteal Artery In-stent Restenosis. Korean Circ J. 2018; 48(3):191–7. pmid:29557105
  58. 58. Zeller T, Dake MD, Tepe G, Brechtel K, Noory E, Beschorner U, et al. Treatment of femoropopliteal in-stent restenosis with paclitaxel-eluting stents. JACC Cardiovasc Interv. 2013; 6(3):274–81. pmid:23517839
  59. 59. Tomoi Y, Soga Y, Iida O, Shiraki T, Kobayashi Y, Hiramori S, et al. Impact of Drug-Eluting Stent Implantation for Femoropopliteal In-Stent Occlusion. J Endovasc Ther. 2016; 23(3):461–7. pmid:27004496
  60. 60. Gray WA, Keirse K, Soga Y, Benko A, Babaev A, Yokoi Y, et al. A polymer-coated, paclitaxel-eluting stent (Eluvia) versus a polymer-free, paclitaxel-coated stent (Zilver PTX) for endovascular femoropopliteal intervention (IMPERIAL): a randomised, non-inferiority trial. Lancet. 2018; 392(10157):1541–51. pmid:30262332
  61. 61. Lammer J, Bosiers M, Zeller T, Schillinger M, Boone E, Zaugg MJ, et al. First clinical trial of nitinol self-expanding everolimus-eluting stent implantation for peripheral arterial occlusive disease. J Vasc Surg. 2011; 54(2):394–401. pmid:21658885
  62. 62. Stone GW, Teirstein PS, Meredith IT, Farah B, Dubois CL, Feldman RL, et al. A prospective, randomized evaluation of a novel everolimus-eluting coronary stent: the PLATINUM (a Prospective, Randomized, Multicenter Trial to Assess an Everolimus-Eluting Coronary Stent System [PROMUS Element] for the Treatment of Up to Two de Novo Coronary Artery Lesions) trial. J Am Coll Cardiol. 2011; 57(16):1700–8. pmid:21470815
  63. 63. Kelly CR, Teirstein PS, Meredith IT, Farah B, Dubois CL, Feldman RL, et al. Long-Term Safety and Efficacy of Platinum Chromium Everolimus-Eluting Stents in Coronary Artery Disease: 5-Year Results From the PLATINUM Trial. JACC Cardiovasc Interv. 2017; 10(23):2392–400.
  64. 64. Iida O, Takahara M, Soga Y, Yamaoka T, Fujihara M, Kawasaki D, et al. 1-Year Outcomes of Fluoropolymer-Based Drug-Eluting Stent in Femoropopliteal Practice: Predictors of Restenosis and Aneurysmal Degeneration. JACC Cardiovasc Interv. 2022; 15(6):630–8. pmid:35331454
  65. 65. Golzar J, Soga Y, Babaev A, Iida O, Kawasaki D, Bachinsky W, et al. Effectiveness and Safety of a Paclitaxel-Eluting Stent for Superficial Femoral Artery Lesions up to 190 mm: One-Year Outcomes of the Single-Arm IMPERIAL Long Lesion Substudy of the Eluvia Drug-Eluting Stent. J Endovasc Ther. 2020; 27(2):296–303. pmid:31989856
  66. 66. Müller-Hülsbeck S, Keirse K, Zeller T, Schroë H, Diaz-Cartelle J. Long-Term Results from the MAJESTIC Trial of the Eluvia Paclitaxel-Eluting Stent for Femoropopliteal Treatment: 3-Year Follow-up. Cardiovasc Intervent Radiol. 2017; 40(12):1832–8. pmid:28948322
  67. 67. Soga Y, Fujihara M, Tomoi Y, Iida O, Ishihara T, Kawasaki D, et al. One-Year Late Lumen Loss between A Polymer-Coated Paclitaxel-Eluting Stent (Eluvia) and a Polymer-Free Paclitaxel-Coated Stent (Zilver PTX) for Femoropopliteal Disease. J Atheroscler Thromb. 2020; 27(2):164–71. pmid:31257301
  68. 68. Gray WA, Griffiths RI, Elroy PWM, Amorosi SL, McGovern AM, Jaff MR, et al. Cost-effectiveness of a paclitaxel-eluting stent (Eluvia) compared to Zilver PTX for endovascular femoropopliteal intervention. J Med Econ. 2022; 25(1):880–7. pmid:35703041