The endovascular treatment of peripheral artery disease (PAD) in the superficial femoral artery (SFA) today offers several options that reflect astounding technologic advances and ingenuity. After plain balloon angioplasty came nitinol stents, stent grafts, drug-eluting stents, and more recently, drug-coated balloons, some in conjunction with various atherectomy devices. These devices have all been heralded at some point as safe, effective, and able to give better results compared with “previously published cohorts.” However, given this plethora of options, surprisingly little high-quality evidence is available as to which method is in fact superior to another and how exactly and in which lesions or patients this benefit can be obtained.

Nitinol stents have been advocated as the therapy of choice in intermediate or longer SFA lesions, especially after failed plain or drug-coated balloon angioplasty. Thus, regardless of whether a primary or provisional use of stents is preferred, the nitinol stent in the SFA will remain a part of every endovascular armamentarium for the foreseeable future.

However, every implanted self-expanding stent will exert a continuous force onto the surrounding vessel wall, termed chronic outward force (COF), for the remainder of its functional life. COF can be made higher by choosing a larger stent size for a given vessel size and by choosing a stiffer stent that exerts a higher expansive force when implanted in a vessel even slightly below the stent's nominal diameter. Although high COF was previously thought to be beneficial to overcome external compression and stent collapse, there is now anecdotal evidence that too much COF triggers inflammation and neointimal proliferation in the vessel, leading to early restenosis and occlusion. This evidence is available in animal studies as well as in clinical subgroup analyses from past randomized controlled trials.1-5


To assess the effect of COF in an animal experiment, nitinol stents were implanted in the peripheral arteries in porcine models.6 Either high-COF (LifeStent, BD Interventional) or low-COF stents (Astron® Pulsar®*, BIOTRONIK) were used. After 28 and 90 days, the treated arteries were histopathologically evaluated. In the high-COF group, neointimal area was significantly larger at both time points compared to the low-COF stent. Angiographic evaluation revealed a trend for lower lumen narrowing for the low-COF group at 28 and 90 days.


The aforementioned animal results have been partly confirmed in a retrospective analysis of patients who received de novo nitinol stent implantation and had an available follow-up CT 12 to 24 months after the index procedure (Funovics et al, unpublished data). A correlation between relative amount of stent oversizing and restenosis could be detected with a correlation coefficient (R2) of 0.216 for male patients and 0.329 for female patients (Figure 1).

Figure 1. Correlation between stent oversizing and restenosis after 12 to 24 months in a retrospective data set of patients after de novo nitinol stent implantation in the SFA. The correlation coefficients for men (blue line) and women (brown line) are 0.216 and 0.329, respectively.

To further investigate the effect of COF on restenosis, the BIOFLEX-COF randomized controlled trial was designed to prospectively assess the influence of stent COF on patency at 1 and 2 years after implantation.

Patients with de novo SFA lesions were randomized in two groups. The high-COF group received a LifeStent stent with generous oversizing, while the low-COF group received a Pulsar®-18 stent (BIOTRONIK) with minimal oversizing. In the high-COF group, the stents are not only stiffer but also sized approximately 1-mm larger at a given vessel diameter (Table 1). This was done to create the highest difference in COF to feasibly test the force’s effect. The relation between COF and diameter of the stent is shown in Figure 2.

Figure 2. Relation between COF (N/mm, y axis) and stent diameter (mm, x axis) of Pulsar-18 and competitor stents.

After implantation, the diameter of the implanted stent was measured at every millimeter along its longitudinal axis in two planes; from the curves shown in Figure 2, the COF could be calculated at every millimeter. The outcome variable is the amount of neointima in the stent measured at CTA after 1 and 2 years. Similarly, the in-stent restenosis is measured at every millimeter along the stent axis and the percentage of lumen loss is calculated (Figure 3).

Figure 3. CTA (coronal reformation) of an implanted stent at 1-year follow-up. The bright white lines represent the stent surface, and the grey area inside the stent is the perfused lumen. Adjacent to the inner stent wall, there are black neointima formations that narrow the perfused lumen to a various extent.

A total of 89 patients are included in the study (high COF, n = 44; low COF, n = 45), and their demographics represent a typical cohort of PAD patients, with main comorbidities of diabetes (51% and 37%), smoking (58% and 43%), hypertension (88% and 98%), and hyperlipidemia (88% and 91%) in the high- and low-COF groups, respectively.

Mean lesion length was 127 and 167 mm in the high- and low-COF groups, respectively, and a second lesion was treated in three cases for each group. Occlusions were treated in 12 patients and 17 patients, with a mean length of 98 and 100 mm, respectively in the high- and low-COF groups.

The preliminary comparison of COF shows that the study could meet its technical goal in creating two groups of otherwise similar characteristics that differ markedly in the stent COF.7 While the high-COF group showed a median total COF of 55.69 N per patient, the low-COF group showed only 14.89 N, a 3.7-fold lower COF (P < .001). The preliminary results show lower restenosis in the low-COF group by a clinically relevant amount.

One-year results are expected to be published in the coming months, and we are confident that our approach to measure neointima on CTA will prove more sensitive in detecting early changes in lumen loss, even before clinical events are noted.


BIOFLEX-COF represents the first prospective randomized trial to assess COF of nitinol stents in the SFA, and we believe we will gain valuable insight as to which sizing and design can be recommended in the future of SFA stenting.

Astron, BIOTRONIK, Passeo, and Pulsar are trademarks or registered trademarks of the BIOTRONIK Group of Companies. All other trademarks are the property of their respective owners.

1. Zhao HQ, Nikanorov A, Virmani R, et al. Late stent expansion and neointimal proliferation of oversized nitinol stents in peripheral arteries. Cardiovasc Intervent Radiol. 2009;32:770-726.

2. Saguner AM, Traupe T, Raber L, et al. Oversizing and restenosis with self-expanding stents in iliofemoral arteries. Cardiovasc Intervent Radiol. 2012;35:906-913.

3. Wressnegger A, Kaider A, Funovics M. Self-expanding nitinol stents of high versus low chronic outward force in de novo femoropopliteal occlusive arterial lesions (BIOFLEX-COF trial): study protocol for a randomized controlled trial. Trials. 2017;18:594.

4. Burket M. BIOFLEX-I increased outward force of SE-BMS in the SFA. Presented at: Charing Cross Symposium 2019; April 16, 2019; London, United Kingdom.

5. Dake M. Insights from Global Zilver PTX experience: what are the predictors for clinical failure and TLR? Presented at: LINC 2019; January 23, 2019; Leipzig, Germany.

6. Funovics M. Chronic outward force (COF) of self-expanding arterial stents correlates with neointimal proliferation in swine. Presented at: CIRSE 2017; September 16-20, 2017; Copenhagen, Denmark.

7. Funovics M. Differences in clinical outcomes of low COF stent vs high COF stent proven in clinical practice. Presented at: CIRSE 2019; September 7-10, 2019; Barcelona, Spain.

*Data generated for Astron Pulsar and Pulsar-18; Astron Pulsar, Pulsar-18, Pulsar-18 T3 and Pulsar-35 have equivalent stent platforms; therefore, the preclinical and clinical results are valid for the Pulsar range.

Prof. Martin A. Funovics, MD
Division of Cardiovascular and Interventional Radiology
Department of Biomedical Imaging and Image-Guided Therapy
Medical University of Vienna
Vienna, Austria
Disclosures: Research grant, travel support, and lecture fees from BIOTRONIK.