The COMPLIANCE 360° Trial: A Randomized, Prospective, Multicenter, Pilot Study Comparing Acute and Long-Term Results of Orbital Atherectomy to Balloon Angioplasty for Calcified Femoropopliteal Disease

Abstract: Objective. This study compares treatment results of orbital atherectomy (OA) vs balloon angioplasty (BA) for calcified femoropopliteal (FP) disease. BA for calcified FP disease is associated with increased dissection rates and suboptimal results. OA is hypothesized to decrease these acute complications via lesion compliance change. Methods. Fifty patients (65 lesions) with calcified FP disease were randomized to OA plus BA vs BA alone and followed for 12 months. The primary endpoint was freedom from target lesion revascularization (TLR), including adjunctive stenting, or restenosis as evidenced by duplex ultrasound at 6 months. Results. Mean maximum balloon pressure was 4.0 atm in the OA arm vs 9.1 atm in the BA arm (P<.001). In subjects with residual stenosis >30%, the operator chose to stent 2/38 lesions (5.3%) in the OA arm vs 21/27 lesions (77.8%) in the BA arm (P<.001). Freedom from TLR (including adjunctive stenting) or restenosis was achieved in 77.1% of lesions in the OA group vs 11.5% in the BA group (P<.001) at 6 months, and 81.2% vs 78.3% at 12 months, excluding adjunctive stenting (P>.99). Conclusions. Compared to BA alone, OA plus BA yields better luminal gain by improving lesion compliance and decreases adjunctive stenting in the treatment of calcified FP disease. At 12 months, the occurrence of TLR or restenosis was similar in both groups despite the large disparity in stent usage at the time of initial treatment.

J INVASIVE CARDIOL 2014;26(8):355-360

Key words: orbital atherectomy, superficial femoral artery

The adductor canal portion of the superficial femoral artery (SFA) and the popliteal artery are mechanically dynamic vascular segments. Stents placed in this region are subjected to a number of unfavorable forces that increase the risk of stent fracture and restenosis.^1,2 While newer iterations of vascular stents have led to a decrease in the incidence of these complications, none of the randomized stent trials have included the popliteal artery beyond its proximal third, and restenosis remains a problem and accrues over time.³ Furthermore, the superiority of stenting to balloon angioplasty (BA) has been demonstrated in 15 cm SFA lesions; however, stenting for short SFA lesions may offer no benefit over PTA alone.^4,5 In addition, the lack of an effective therapy for in-stent restenosis has led many practitioners to pursue a treatment strategy of provisional stenting for femoropopliteal (FP) disease and therefore BA and atherectomy continue to be used in practice. 

The presence of lesion calcification plagues all methods of endovascular revascularization, especially BA. Fitzgerald et al demonstrated through intravascular ultrasound (IVUS) investigation that three-quarters of lesions in both peripheral and coronary arteries that dissect after BA contain calcium.⁶ Heavy calcification can also lead to stent underexpansion. Interventional tools that can modify lesion calcification while minimizing vessel injury and the need for stent placement (in a provisional stent strategy)would be beneficial. 

The Diamondback 360° Peripheral Orbital Atherectomy System (Cardiovascular Systems, Inc) has been previously well described.^7,8 The chief mechanism of action involves differential sanding with preferential removal of the relatively non-compliant elements of the plaque, particularly calcium, and a lesser effect on the more elastic components, most notably the normal vessel wall. The orbital nature of the device lends itself to the larger diameter of the FP vascular segment.⁸ The investigators hypothesize that vessel preparation utilizing orbital atherectomy (OA) decreases the acute complications associated with BA in calcium-containing FP lesions, reducing the need for adjunctive stenting. 

Methods

Study design and patients. The COMPLIANCE 360° is a prospective, randomized, multicenter pilot study comparing the impact of vessel compliance change after use of OA vs BA alone in treating FP disease. A total of 50 patients were enrolled at 9 United States centers. Since this was a pilot study, enrollment was limited to 50 patients. Each site enrolled an average of 5.5 patients/center (range, 2-13 patients/center). Eligible patients were 18 years or older, had peripheral arterial disease with Rutherford class 2-4 symptoms and de novo FP lesions of ≥70% stenosis with fluoroscopically visible calcium, and gave informed consent. All patients had to have at least 1 patent run-off vessel. There were no limitations with regard to lesion length or number of lesions within the defined vascular segment. Distal popliteal lesions were included. Only one limb per patient was allowed. The protocol required intraluminal guidewire passage and absence of thrombus. Major exclusion criteria included an anticipated life span of less than 1 year; known allergy to heparin, aspirin, and clopidogrel, or sensitivity to contrast media; chronic renal failure; cardiac arrhythmias; congestive heart failure exacerbation; and myocardial infarction. Patients were recommended to be on an antiplatelet agent before the procedure, preferably clopidogrel, and combined aspirin + clopidogrel for at least 4-6 weeks post procedure. 

Randomization. Randomization occurred after each subject met all inclusion criteria after angiography (digital subtraction) and no exclusion criteria. The subject was then enrolled in the study. The subjects were equally randomized to OA plus BA or BA alone. To ensure proper randomization, sealed randomization envelopes were provided to each site. Crossover to the other randomization arm was not allowed. Note that patients, not lesions, were randomized. A lesion was defined as distinct from another in the same limb if there was an interval segment of any length between adjacent diseased segments that did not require (or undergo) any treatment. Lesions with any fluoroscopically visible calcium, as determined by the operator, were included as it was hypothesized that even low lesion calcium content may contribute to suboptimal BA results. A calcium grading system was devised that accounted for both arc and relative length of calcium within the diseased segments as estimated by angiography. The relationship of calcium arc and dissection has been previously demonstrated.⁶ An assumption inherent in this grading system is that calcium located on both sides of a lesion at a given location represents an arc of at least 180°; clearly, this appearance could occur as a result of two discontinuous arcs as well. 

Procedures. In patients who received BA treatment as an adjunct to OA, the choice of balloon was selected by the operator. Semicompliant balloons and 1:1 artery-to-balloon sizing were recommended in both arms. The method of balloon inflation was standardized in both arms. The recommended balloon pressure was ≤4 atm. In the OA arm, if a waist was still visible after inflation with 4 atm, as viewed in two planes, the operating physician could use OA again in this arm for the removal of calcified plaque in the specific segment after angiographically verifying no dissection was present. In the BA arm, it was recommended that balloon pressure start at a low pressure of 3 to 4 atm and increase to higher pressure to achieve full balloon expansion with no balloon waist visible as viewed in two planes. Each slow, gradual, low-pressure inflation was recommended to be held for at least 60 seconds. The lowest pressure required to achieve complete balloon expansion was determined, which was considered a good proxy for lesion compliance. Balloon expansion was considered complete when no balloon waist was visible when viewed in two planes. The two planes were defined as the two working views that minimized the degree of vessel overlap and demonstrated the stenosis in its most severe view. After the prescribed therapy of OA plus BA or BA alone, residual stenosis was estimated visually by the operator. Lesions failing to achieve a residual stenosis of ≤30% by angiography were recommended to be adjunctively stented, at the discretion of the operator. Most studies have used this criterion for stent placement,^4,9 but ultimately it depends on the operator’s decision based on the patient. 

Follow-up. Patient demographics, lesion characteristics including calcium grade, and a number of procedural variables were recorded. Clinical follow-up was obtained at 1, 6, and 12 months along with ankle-brachial index and Rutherford class, which were also obtained at baseline. Duplex ultrasound was performed at 6 and 12 months and restenosis was defined as a peak systolic velocity ratio (PSVR) of ≥2.5. Angiographic results were determined by visual estimation by the operator. Duplex ultrasound results were required and performed by the investigators’ institutions. PSVR was determined by using the greatest systolic velocity within the treated segment in the numerator and the systolic velocity in the nearest “normal” segment proximal to the lesion in the denominator.

Endpoints. The primary endpoint was defined as freedom from target lesion revascularization (TLR), including the need for adjunctive stenting or restenosis (PSVR ≥2.5 on duplex ultrasound) per lesion at 6 months. Six months was chosen due to a concern of loss to follow-up over a longer time frame given the small sample size and the likelihood of capturing the majority of events composing this endpoint within 6 months. Secondary endpoints included changes in ankle-brachial index and Rutherford Class from baseline to 6 and 12. Mean maximal balloon inflation pressure required for complete balloon expansion and the adjunctive stent rate were also collected. Lesion treatment success was defined as ≤30% residual stenosis without stent placement.

Statistical analysis. Due to the pilot nature of the study, no sample size calculations were performed. Baseline patient-level comparisons between groups were conducted using Wilcoxon rank-sum tests for continuous variables and Fisher’s exact test for categorical variables. Significance testing for patient-level endpoints was conducted using Fisher’s exact test or general linear model, where appropriate. Significance testing on the lesion level was conducted using generalized estimating equations with an exchangeable working correlation structure to account for correlated outcome data, where multiple lesions were treated within a single patient. Continuous variables were analyzed with generalized estimating equation models under a normal distribution and an identity link function, and categorical variables (including the primary endpoint) were analyzed under a binomial distribution and a logit link function. All P-values are two-sided, and statistical significance was set at the .05 level. Analyses were performed by an independent biostatistician using SAS software, version 9.2 (SAS Institute, Inc).

Results

A total of 50 patients were randomized in the study with 25 patients (38 lesions) to the OA arm and 25 patients (27 lesions) to the BA arm. The disparity in lesion numbers between the respective arms occurred by chance. Patient demographics are noted in Table 1. Significantly more patients in the OA group were diabetic (P=.05).

Lesion characteristics are presented in Table 2. The mean lesion length was 55.9 mm in the OA group and 87.3 mm in the BA group. (P=.09). Eight lesions (21%) were chronic total occlusions in the OA group vs 5 lesions (18.5%) in the BA group. The BA group had a greater severity of baseline stenosis, although minimum lumen diameter as assessed visually by the operator was not significantly different between the two groups. Mean angiographic calcium scores (Table 2) were similar (2.61 in the OA group vs 2.37 in the BA group). One lesion in the OA arm did not contain visible calcium, but was included because it was treated along with other calcified lesions in that limb. There were more lesions in the OA group exposed to poor run-off (defined as 1 run-off vessel vs 2 or 3 run-off vessels) (P=.05). The majority of crowns used were 2.00 mm or 2.25 mm solid crowns (37/45), with multiple crowns used in some lesions (Table 3). There were no reports of hemolysis as manifested by visibly apparent hemoglobinuria. One occurrence of distal embolization occurred in the OA arm. Mean maximal balloon pressure was less in the OA group (4.0 atm) vs the BA group (9.1 atm), which was statistically significant (P<.001). The criterion for lesion treatment success (residual stenosis ≤30% without stent placement) after assigned treatment was met in 86.8% of the lesions in the OA group vs 18.5% in the BA group (P<.001). Only 2 lesions were stented in the OA group, while 21 lesions were stented in the BA group (P<.001). There was a considerable improvement in minimum lumen diameter noted after OA (Table 3) although debulking, per se, was not the goal. Procedural complications are presented in Table 3. 

The primary endpoint was determined at 6-month follow-up and consisted of freedom from TLR including the need for adjunctive stenting or restenosis (PSVR ≥2.5 on duplex ultrasound). At 6 months, 77.1% of lesions in the OA group (27/35) vs 11.5% in the BA group (3/26; P<.001) were free from TLR (including adjunctive stenting) and restenosis. At 12-month follow-up, freedom from TLR (excluding initial adjunctive stenting) or restenosis was achieved in 81.2% of lesions (6/32) in the OA group vs 78.3% (5/23) in the BA group (P>.99). All stents used were nitinol, ranged in size, and were from a variety of manufacturers. Restenosis or TLR in patients receiving stents did not correlate with lesion location, stent size, or manufacturer. Changes in ankle-brachial index and Rutherford class were similar between the two groups at all follow-up times (Table 4). 

Discussion

Femoropopliteal disease continues to present endovascular treatment challenges to interventionists. While drug-eluting nitinol stents may ultimately influence practice,⁹ provisional stenting continues to be the chief modus operandi in this vascular segment despite evidence that supports primary stenting.¹⁰ While a good BA result can lead to reasonable patency in some lesions, suboptimal acute results frequently occur, including vessel dissection and recoil, and stents have provided an expedient solution for these problems. However, the dynamic nature of this vascular segment and the resulting forces exerted upon these implanted devices appears to increase stent fractures, which contributes to the problem of restenosis.⁹ Accordingly, other tools continue to be sought to improve current interventional results. It has been well established that lesion calcium is responsible for complications associated with BA and stenting. In the Resilient and Zilver trials, 48.3% and 57% of the lesions contained calcium in the percutaneous transluminal angioplasty (PTA) arm, respectively. Of the patients in the PTA arm, 40.3% and 50.4% required bail-out stenting.^4,9 In the COMPLIANCE 360° study, 100% of the lesions were calcified, so it seems reasonable that 80% of the lesions in the BA arm required stents. Historically, these trials have demonstrated that stents perform better than BA alone. In this study, including more stents in the BA arm would have been expected to favor patency for that arm. 

In broader terms, it is the relative non-compliance of calcified plaque that often prevents uncomplicated vessel expansion with BA and the juxtaposition of non-compliant and compliant tissue that leads to dissection. Orbital atherectomy is extremely effective in removing this relatively non-compliant plaque and has been shown to be very safe.^7,8 By achieving improved vessel compliance, OA allows for more effective and less complicated balloon angioplasty (Figure 1). Plaque modification rather than maximal debulking is the OA objective in this vascular segment. Pretreatment with OA may possibly result in decreased vessel wall injury via diminished deep wall dissection along with reduced barotrauma afforded by lower balloon pressures. At 12-month follow-up, subsequent TLR or restenosis were similar in the OA and BA groups, even though only two stents were placed in patients in the OA group. Avoidance of in-stent restenosis, a difficult problem to treat, is a key advantage of the OA therapeutic approach. 

Study limitations. A limitation of this study was the small number of patients, which resulted in some differences in baseline characteristics. There was a trend for a greater number of diabetic patients in the OA group (P=.05). More lesions were treated in the OA arm (n = 38) than in the BA arm (n = 27). The mean lesion length was greater in the BA arm (87.3 mm) vs the OA arm (55.9 mm), although this was not statistically significant. However, percent stenosis was statistically significant (P<.01) with mean stenosis of 92.8% in the BA arm and 83.9% in the OA arm. There was a significantly greater number of lesions in the OA group exposed to poor run-off (defined as 1 run-off vessel vs 2 or 3 run-off vessels), which has been shown to correlate with poorer long-term patency.^11,12 Procedurally, the mean procedure time was 96.3 minutes in the OA arm and 69.6 minutes in the BA arm. The increased OA procedure time was attributed to a greater number of lesions treated. On average, the lesion treatment time was similar in both groups.

The main limitation of the study is the lack of independent angiographic and duplex ultrasound core lab adjudication. In addition, patients were lost to follow-up during the course of the study. Operator discretion to adjunctively stent lesions failing to achieve residual stenosis of ≤30%, rather than being protocol driven, may have introduced bias into the study. Leaving stent placement to the discretion of the operator could theoretically have led to fewer stents placed in the OA arm or greater number of stents placed in the BA arm. This bias would have positively impacted the 12-month BA restenosis rate since stenting is superior to PTA.⁹ However, the results of the study demonstrate the same restenosis rates between the OA and BA arms, so it is unlikely that any operator bias regarding stent placement would have favorably influenced the OA results. A larger trial comparing OA with adjunctive low-pressure BA to primary stenting for calcified FP disease would be useful in further defining the comparative effectiveness of these two therapeutic approaches.

Conclusions

The primary endpoint of freedom from TLR (including adjunctive stenting) or restenosis at 6 months was met in 77.1% of lesions in the OA group vs 11.5% in the BA group. Compared to BA alone for the treatment of calcium-containing FP lesions, OA pretreatment likely improves lesion compliance and leads to better luminal gain with lower balloon pressures, resulting in a marked reduction of adjunctive stenting. Patency at 12 months with OA therapy is similar to a provisional stent strategy despite minimal stent usage. Avoidance of in-stent restenosis and preserving future treatment options, by not placing a stent, are key advantages of the OA therapeutic approach.

Acknowledgments. The authors thank Brad J. Martinsen, PhD, Robert Kohler, MS, and Ann Behrens, BS, of Cardiovascular Systems, Inc, for editing and critical review of this manuscript.

References

1. Smouse HB, Nikanorov A, LaFlash D. Biomechanical forces in the femoropopliteal arterial segment. Endovascular Today. 2005;4:60-66.
2. Scheinert D, Scheinert S, Sax J et al. Prevalence and clinical impact of stent fractures after femoropopliteal stenting. J Am Coll Cardiol. 2005;45(2):312-315.
3. Iida O, Nanto S, Uematsu M, Ikeoka K, Okamoto S, Nagata S. Influence of stent fracture on the long-term patency in the femoro-popliteal artery: experience of 4 years. JACC Cardiovasc Interv. 2009;2(7):665-671.
4. Laird JR, Katzen BT, Scheinert D, et al. Nitinol stent implantation versus balloon angioplasty for lesions in the superficial femoral artery and proximal popliteal artery: twelve-month results from the RESILIENT randomized trial. Circ Cardiovasc Interv. 2010;3(3):267-276.
5. Krankenberg H, Schluter M, Steinkamp HJ, 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):285-292.
6. Fitzgerald PJ, Ports TA, Yock PG. Contribution of localized calcium deposits to dissection after angioplasty. An observational study using intravascular ultrasound. Circulation. 1992;86(1):64-70.
7. Safian RD, Niazi K, Runyon JP, et al. Orbital atherectomy for infrapopliteal disease: device concept and outcome data for the OASIS trial. Catheter Cardiovasc Interv. 2009;73(3):406-412.
8. Adams GL, Khanna PK, Staniloae CS, Abraham JP, Sparrow EM. Optimal techniques with the Diamondback 360° System achieve effective results for the treatment of peripheral arterial disease. J Cardiovasc Transl Res. 2011;4(2):220-229. Epub 2011 Feb 11.
9. Dake MD, Ansel GM, Jaff MR, et al. Paclitaxel-eluting stents show superiority to balloon angioplasty and bare metal stents in femoropopliteal disease: twelve-month Zilver PTX randomized study results. Circ Cardiovasc Interv. 2011;4(5):495-504. Epub 2011 Sep 27.
10. Schillinger M, Sabeti S, Dick P, et al. Sustained benefit at 2 years of primary femoropopliteal stenting compared with balloon angioplasty with optional stenting. Circulation. 2007;115(21):2745-2749. Epub 2007 May 14.
11. Clark TW, Groffsky JL, Soulen MC. Predictors of long-term patency after femoropopliteal angioplasty: results from the STAR registry. J Vasc Interv Radiol. 2001;12(8):923-933.
12. Davies MG, Saad WE, Peden EK, Mohiuddin IT, Naoum JJ, Lumsden AB. Percutaneous superficial femoral artery interventions for claudication — does runoff matter? Ann Vasc Surg. 2008;22(6):790-798.

________________________________________

From the ¹St. Francis Heart and Vascular Center, Topeka, Kansas; ²Memphis Heart Clinic, Southaven, Missouri; and ³Spectrum Health Medical Group, Grand Rapids, Michigan.

Funding: The COMPLIANCE 360° Study was sponsored by Cardiovascular Systems, Inc. 

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Dattilo reports research grants, consulting fees, and speaking/teaching honoraria from Cardiovascular Systems, Inc. Dr Himmelstein reports consulting fees and speaking/teaching honoraria from Cardiovascular Systems, Inc. Dr Cuff reports no disclosures. 

Manuscript submitted November 7, 2013, provisional acceptance given January 26, 2014, final version accepted February 7, 2014.

Address for correspondence: Raymond Dattilo, MD, FACC, Flint Hills Heart, Vascular and Vein Clinic, 3905 Vanesta Drive, Manhattan, KS 66503. Email: icardmd@aol.com