Acute Real-World Outcomes From the Phoenix Post-Approval Registry

Objectives. An evaluation of the 30-day safety and performance outcomes of the Phoenix atherectomy system (Philips Volcano Corporation) was performed in real-world patients with peripheral artery disease (PAD). Methods. The Phoenix Post-Approval Registry is an all-comer study that enrolled patients with infrainguinal PAD. Patients treated with the Phoenix atherectomy system were followed for 30 days to observe device-related complications. Outcomes evaluated include procedural (final target lesion(s) residual stenosis of ≤30% after treatment with Phoenix and any other adjunctive therapy) and technical success (defined as achieving a post-Phoenix [prior to any adjunctive therapy] residual diameter stenosis of ≤50%), target-vessel revascularization (TVR), target-lesion revascularization (TLR), target-limb amputation, ankle brachial index, Rutherford clinical category, and wound, ischemia, foot infection (WIfI) classification. Results. Of the 500 patients enrolled, 259 had CLI, including 26.3% with Rutherford class 6. Procedural success rates were 97.3% for non-CLI patients and 98.2% for CLI patients. Technical success rates were 71.5% for non-CLI patients and 77.9% for CLI patients. Complication rates post Phoenix atherectomy were <1%. Through the 30-day follow-up, there were 6 patients (1.3%; 2 claudicants, 4 CLIs) who underwent TLR and 8 patients who underwent TVR. There were no major amputations in the non-CLI and CLI cohorts. In the CLI cohort, 16/235 (6.8%) underwent minor amputations. Higher stages of Rutherford class and WIfI classification were associated with amputations at 30 days. Conclusion. The Phoenix atherectomy system is a safe and effective treatment option in the acute setting for patients with PAD, including those with advanced Rutherford class. Randomized controlled trials are needed to confirm these results.

J INVASIVE CARDIOL 2022;34(1):E1-E7. Epub 2021 December 19.

Key words: atherectomy, peripheral artery disease

Endovascular treatment of symptomatic peripheral arterial disease (PAD) has widely been accepted as a first-line treatment for lower-limb revascularization. While there are a multitude of treatment options within this domain, there is no single application or algorithm that has shown durability for the varied clinical and anatomic manifestations of lower-limb PAD. Atherectomy, an increasingly popular form of endovascular treatment,¹ has shown promise as a therapeutic tool for PAD. Atherectomy has been shown to be an effective option either alone or as an adjunctive treatment.^2-5 There are 4 categories of atherectomy devices on the market: directional; orbital; atheroablative; and rotational.⁶ Each of the aforementioned device types has its respective advantages and limitations. Further research has yet to elucidate the safest and most effective type of atherectomy.

The Phoenix atherectomy system (Philips Volcano Corporation) was developed as a novel over-the-wire device that leverages and supplements the current functions of existing atherectomy devices while reducing the risks of vessel injury and distal embolization through the design of the Phoenix cutter head. The single insertion device has a distal tip with a front cutting metal element to directly access and cut the lesion. The various device components, such as the blades and Archimedes screw, allow for continuous cutting, capturing, and clearing of atherosclerotic lesions. The device has been commercially available since 2014.

The primary objective of the Phoenix Post-Approval Registry was to evaluate the early clinical outcomes of all patients treated with the Phoenix atherectomy system for PAD under real-world conditions following approval by the United States Food and Drug Administration (FDA). This publication specifically focuses on device performance through 30 days for all patients enrolled in the registry.

Study design. The single-arm Phoenix Post-Approval Registry was a prospective, multicenter, observational study comprising patients with PAD who were treated with the Phoenix atherectomy system in a postmarket setting (ClinicalTrials.gov NCT02475200). Institutional review board (IRB) approval of the protocol and informed consent were obtained by the investigator. The study was conducted in accordance with applicable sections of the Code of Federal Regulations (21 CFR 50, 54, 56), ISO 14155:2011, and the International Conference on Harmonization Good Clinical Practice Guidelines. All patients were assessed prior to index procedure (baseline), during index procedure, and 1 month post index procedure. Patients with Rutherford clinical categories of 4, 5, or 6 at baseline (the CLI cohort) were followed for 12 months.

Study population and eligibility criteria. There were minimal inclusion/exclusion criteria. Patients were eligible for inclusion if they were over the age of 18 years, provided written informed consent, and met the requirements of the Phoenix atherectomy system catheter instructions for use (IFU) criteria. According to the IFU, the system is only intended for use in the peripheral vasculature, namely, the femoral, popliteal, or infrapopliteal arteries. Patients were excluded from the trial if they could not comply with the protocol, had in-stent restenosis, or were participating in other investigational trials or devices.

Device description. The Phoenix atherectomy system, comprised of the Phoenix atherectomy catheter and the Phoenix atherectomy handle, is an over-the-wire atherectomy device used to cut, continuously capture, and passively clear intraluminal disease (such as atheromatous plaque or calcium) with a single insertion. The Phoenix atherectomy catheter, a flexible double-lumen catheter, contains a distal tip with a front cutter that can clear a broad range of intraluminal material types, including soft plaques and those with calcific lesions, above (ATK) and below the knee (BTK). When activated, the catheter’s torque shaft rotates and the attached cutter excises the plaque from within the arteries. The Archimedes screw then clears the debulked tissue. The catheter’s distal end is radiopaque and enables fluoroscopic visualization. The proximal end contains 2 ports (a side port for debris transfer and a main port for the guidewire). The proximal end of the catheter is also where the Phoenix atherectomy handle, a battery-powered motor, connects. The device is available in the following diameters: 1.8 mm, 2.2 mm, and 2.4 mm. The largest diameter device has an additional deflecting mechanism. The Phoenix wire support clip can be used to facilitate guidewire management because it supports and secures the guidewire.

Endpoints. All endpoints were assessed similarly in patients with claudication and CLI. The primary safety endpoint was freedom from device-related complications, whereas the primary efficacy endpoint was procedural success, defined as final target-lesion residual stenosis ≤ 30% in vessels adhering to the prespecified sizing targets following treatment using quantitative angiography, intravascular ultrasound (IVUS), or a visual estimate. Technical success refers to the achievement of acute debulking, defined as a post-Phoenix, preadjunctive therapy, residual diameter stenosis ≤50% in vessels that meet the prespecified sizing targets assessed by the investigator using various imaging modalities (quantitative angiography, IVUS, or visual estimate) at the time of treatment. Regardless of the imaging modality used, all procedural and technical success calculations were derived from angiographic data. Technical success was evaluated based upon a device type to vessel size ratio. A 1.8 mm Phoenix atherectomy system corresponded to a 3.5 mm reference vessel diameter (RVD). Similarly, a 2.2 mm device could be used on a vessel with a 4.5 mm RVD.

The secondary endpoints focused on the rates of target-vessel revascularization (TVR), target-lesion revascularization (TLR), and target-limb amputation (major and minor), as well as changes in ankle brachial index (ABI), Rutherford clinical category, and wound, ischemia, foot infection (WIfI) classifications. Major amputations were defined as any amputation performed above the level of the ankle and minor amputations were defined as any amputation below the level of the ankle.

Statistical analyses. All patients enrolled in the registry were accounted for in the analysis, as part of the intention-to-treat population. The data were summarized using descriptive statistics. Continuous variables are presented as mean ± standard deviation and categorical variables are presented as fractions (percentage). The comparisons between CLI and non-CLI patients utilized t-tests for approximately normally distributed continuous outcomes, Wilcoxon rank-sum test for ordinal outcomes, and Fisher’s exact test or Chi-square test for categorical outcomes. The statistical significance of patient change from baseline to follow-up evaluations was assessed using either the paired t-test for approximately normal continuous outcomes or Wilcoxon signed-rank test for ordinal outcomes. Lastly, the Chi-square test was used to compare amputation rates for diabetics vs non-diabetics and the Jonckheere-Terpstra test was used to compare amputation rates across Rutherford and WIfI scores.

Five hundred patients (284 men; mean age, 70.1 ± 10.1 years) were enrolled in the registry (241 non-CLI, 259 CLI). The baseline demographics between the CLI and non-CLI cohorts showed significant differences in male gender (61.8% non-CLI vs 52.1% CLI; P=.03), current use of antiplatelet therapy (83.8% non-CLI vs 74.1% CLI; P<.01), and history of smoking (81.7% non-CLI vs 59.1% CLI; P<.001) (Table 1).

Baseline lesion characteristics differed significantly between the 2 groups (Table 2). A greater percentage of CLI patients had BTK lesions, chronic total occlusions, and more severe calcification than the non-CLI patients. Conversely, non-CLI patients had higher occurrences of restenotic lesions and ATK lesions in comparison with CLI patients. The average baseline diameter stenosis was significantly more severe for the CLI group compared with the non-CLI group (90.0 ± 12.0% vs 87.2 ± 13.9%, respectively; P<.01); however, the average lesion length in both groups was statistically similar (103.3 ± 116.3 mm in the non-CLI group vs 103.0 ± 100.8 mm in the CLI group) (Table 2). Most patients (97.0%) were treated with 1 Phoenix device, while the remaining 3.0% were treated with 2 Phoenix devices.

Following treatment with the Phoenix atherectomy system, plain old balloon angioplasty (POBA) was the most common adjunctive therapy and was used significantly more often in non-CLI patients than in CLI patients (Table 3). Drug-coated balloons (DCBs) and stenting were the second- and third-most common adjunctive procedures, respectively. There was no bail-out stenting performed.

The rates of procedural success were comparable in CLI patients (98.2%) and non-CLI patients (97.3%). Furthermore, technical success rates were 67.9% for CLI patients and 71.5% in non-CLI patients.

Immediately post atherectomy, there were no instances of pseudoaneurysms, thrombotic events, or distal embolizations; however, 1 perforation and 2 dissections occurred. After procedure completion, which included Phoenix atherectomy and any additional adjunctive therapies, there was 1 pseudoaneurysm in a posterior tibial artery, 2 instances of thrombus, 1 distal embolization, and 1 perforation. All dissections (9.4%), noted after adjunctive treatment, were mostly minor. There were 3 adverse device events, including 1 pseudoaneurysm, 1 distal popliteal perforation, and 1 instance of device entanglement with an existing stent. The complication rates post Phoenix atherectomy and at final assessment were below 1%, with the exception of dissections.

At 30 days, 2/234 non-CLI patients (0.9%) and 4/235 CLI patients (1.7%) underwent TLR. In the same time period, 1 additional patient in each cohort underwent TVR (Table 3). There were no major amputations in the non-CLI or CLI cohorts. In the CLI cohort, 16/235 (6.8%) underwent minor amputations. Of these, 11 (68.8%) were planned minor amputations. Of the 16 amputations, 15 were at the toe level and 1 patient had a transmetatarsal amputation.

Amputation was also evaluated as a function of history of diabetes, Rutherford category, and WIfI classification (Table 4). Presence of diabetes was not associated with increased frequency of amputation, as 4.5% of diabetic patients and 1.5% of non-diabetic patients underwent amputation through 1 month (P=.06). Moreover, amputation rates through the 1-month follow-up were significantly higher in patients with higher Rutherford category and WIfI classification at baseline. There were 4/278 patients (1.4%) with a very low baseline WIfI classification, 4/122 patients (3.3%) with a low baseline WIfI classification, 4/40 patients (10.0%) with a medium baseline WIfI classification, and 4/16 patients (25.0%) with a high baseline WIfI classification who underwent amputation through the 1-month follow-up.

The mean baseline ABI was 0.74 ± 0.23 for non-CLI patients and 0.76 ± 0.33 for CLI patients. At 1 month, the average ABI increased to 0.98 ± 0.20 in the non-CLI group and 0.96 ± 0.34 in the CLI group. The difference in the change in ABI between the 2 groups was not statistically significant (P=.16). Furthermore, the mean baseline toe brachial index (TBI) for non-CLI and CLI patients was 0.56 ± 0.25 and 0.49 ± 0.28, respectively. At 1 month, the mean TBI increased for both the non-CLI cohort (0.70 ± 0.23) and CLI cohort (0.64 ± 0.29), showing a statistically significant improvement from baseline in both non-CLI (P<.001) and CLI cohorts (P<.001).

The most common Rutherford category at baseline was 3 for the non-CLI group (190/241 patients) and 5 for the CLI group (107/259; Figure 1). At 30 days post intervention, 129 non-CLI patients (55.1%) were classified as Rutherford category 0. Moreover, 63 CLI patients (26.9%) were Rutherford category 0 at 30 days post intervention. The average change in Rutherford category was -1.9 ± 1.2 for non-CLI patients and -1.8 ± 2.0 for CLI patients.

The Phoenix registry, indicative of a real-world population, showed favorable early safety and performance outcomes with a high rate of procedural success and low rates of distal embolization and perforation associated with the Phoenix atherectomy system. The study results presented here further support atherectomy as a safe and effective tool for treating lower-extremity PAD.

In today’s clinical landscape, atherectomy devices are rarely used as stand-alone treatment options for PAD. They are primarily viewed as a vessel-preparation tool with the intention of maximizing plaque debulking. When atherectomy is used prior to angioplasty, the balloon does not need to exert as much pressure on the vessel wall, which increases luminal gain while concurrently decreasing vessel barotrauma, inflammation, dissection, and risk of plaque recoil.⁷ DCBs and drug-eluting stent treatments are also optimized following atherectomy because the drug can be delivered more efficiently and uniformly. Atherectomy can be beneficial because it has the ability to clear plaque in troublesome arterial segments, such as the distal foot arterial segment or flexion points, and potentially avoid the need for a stent.⁷

The Phoenix atherectomy system, which received FDA clearance in January 2014, is a novel debulking system with the ability to access and cut a broad range of lesion types. The device is uniquely designed to continuously capture and remove debulked material by the Archimedes screw, which can reduce the risk of distal embolization. The EASE study, a prospective trial to evaluate short- and mid-term safety and effectiveness of the Phoenix device, yielded positive results with high rates of technical success and a freedom from major adverse events through 30 days.⁸ The Phoenix registry augments the EASE data by providing mid- and long-term outcomes on a broader patient population with PAD.

The registry enrolled a diverse group of patients with lesions of varying complexities and presentations. Comparisons between previously published data and the Phoenix registry study results should be made with caution due to differing designs, patient populations, and endpoint definitions. However, in many ways, the Phoenix atherectomy system is comparable to other directional or rotational atherectomy devices. A very high procedural success rate, >97% for all comers, was achieved utilizing additional balloon dilation therapy with POBA in most cases. Moreover, the Phoenix registry results demonstrated low, but comparable rates of dissection and perforation with other atherectomy devices, which range from 2%-3% to 11.3%.^4,9,10 This study also noted rates of distal embolization (0.1%) during the procedure that were on the lower end of the range noted with other atherectomy devices.^4,11 This finding, also confirmed by the EASE trial, is significant as it is an often-cited concern with atherectomy devices. This may be attributed to the clear-and-capture technique utilized by the Phoenix atherectomy system, which decreases the risk of adverse events most typically associated with contemporary atherectomy devices. The use of any embolic protection devices was left to the discretion of the operator; however, it was not captured as part of the study. Patients treated with the Phoenix atherectomy system had a very low rate of TLR and experienced no major index limb amputation at 30 days. Additionally, within the non-CLI group, there were no major or minor index limb amputations. For comparison, the JET registry, which evaluated the Jetstream atherectomy device in non-CLI (Rutherford 1-3) patients, noted 2 instances (0.9%) of TLR and no index-limb amputations within 30 days.¹² Subgroup outcomes were analyzed to discern differences in postprocedure performance. Specific groups of interest included CLI and diabetics. Subset analyses revealed that groups with the more severe disease were often significantly more likely to undergo a target-limb amputation.

Recently, a study of atherectomy outcomes, utilizing the Vascular Quality Initiative database, called into question the use of the therapy in clinical practice due to its higher cost and unclear long-term durability. The study showed that compared with stenting, atherectomy patients had a higher risk of major amputation and major adverse limb events.¹³ However, this study did not fully represent patient characteristics similar to those like the Phoenix registry such as a large BTK and CLI cohort. The results of the Phoenix registry demonstrate the applicability of atherectomy in subjects with a broad spectrum of lower-extremity PAD. It reaffirms that the Phoenix atherectomy system is effective, with marked clinical and hemodynamic improvements, for both CLI and non-CLI patients. Moreover, the excellent safety profile of the device is highlighted by embolization and perforation rates, which are lower than other commercially available atherectomy systems.^4,12,14 The real-world outcomes of Phoenix registry further substantiate the essential role of atherectomy in the endovascular treatment paradigm for PAD. Additional data are warranted to confirm the long-term outcomes of the Phoenix Atherectomy System.

Study limitations. This publication presents the results of a non-randomized, single-arm, early-stage registry, which limits the generalizability of the study results. Furthermore, a randomized controlled trial is required to evaluate the effectiveness of the Phoenix atherectomy system compared with other devices (ie, other atherectomy devices, POBA, stents, drug-coated technology, etc). Moreover, the Phoenix registry only evaluates 12-month outcomes for the CLI cohort, which bars further conclusions on the device’s long-term durability in the non-CLI patient population. Finally, many of the study endpoints were investigator determined as opposed to core-lab adjudicated, which predisposes the study findings to bias or can potentially decrease accuracy because of the non-standardized system for obtaining measurements.

Real-world observational data from the Phoenix registry show the Phoenix atherectomy system is both safe and effective in the treatment of PAD. The results presented here contribute additional support for the use of atherectomy in clinical practice, especially in challenging patient populations. Additional follow-up data analysis is required to ascertain the long-term outcomes associated with the Phoenix atherectomy system and a randomized study is required to confirm the results of this particular device against other devices.

Acknowledgments. The authors would like to acknowledge Margaret Folaron of Philips for editorial support, Tami Crabtree of Advanced Research Associates for statistical assistance, and Kasthuri Nair of Syntactx for technical and statistical support. The authors wish to recognize the work of Deborah K. Blongewicz, RT, CCRP, who had worked for Philips/Volcano for over 13 years in a profession that she loved. She took great pride as the Senior Clinical Project Manager for the Phoenix registry.

• Peripheral Arterial Vessel Prepping with Intravascular Ultrasound and Atherectomy: An Interventionalist’s Perspective
• Intravascular Ultrasound Insights Into Perforation After Coronary Atherectomy
• The Impact of Plaque Rupture on Mismatch Between Intravascular Ultrasound Minimal Lumen Area and Fractional Flow Reserve Measured in an Angiographically Borderline Left Main With Complex Left Anterior Descending Involvement
• Location of Dissections Post Balloon Angioplasty in Patients With Femoropopliteal Arterial Disease as Seen on Intravascular Ultrasound