ABSTRACT: Background: In-stent restenosis (ISR) in the superficial femoral artery (SFA) continues to be the Achilles heel of endovascular treatment of obstructive peripheral arterial disease. Stent fracture (SF) has been identified as one of the possible causes of ISR, but data on the role of SF in the development of ISR remains to be controversial. Methods: 97 consecutive patients (105 limbs) with angiographically confirmed obstructive ISR in the SFA who had previously undergone endovascular revascularization with nitinol stents were studied. Patients with covered stents (Viabahn) were excluded. Stents were evaluated by fluoroscopy in at least 2 orthogonal views. Stent fracture rates, severity, and angiographic relationship to ISR were studied. Logistic regression analysis was performed. Results: Mean time from the stent implantation to presentation with ISR was 15.5±15.3 months. Out of 105 limbs with ISR, SF was present in 31 (30%) limbs, of which only 3 (10%) limbs had SF associated with ISR. Stent fracture occurred more frequently in males, smokers, and in the distal SFA. There were no significant differences in mean stented length or procedural and demographic characteristics between groups with and without SF. Conclusions: Stent fractures in the SFA play a modest role in the development of ISR. In our study, the association between ISR and SF was observed in only 10% of limbs with SF. This represents only 2.9% of the 105 limbs included in our study. Therefore, a vast majority of patients with ISR did not have SF. The strongest predictors of SF were male sex, smoking, and larger stent diameter.  

VASCULAR DISEASE MANAGEMENT 2016;13(1):E7-E16

Key words: nitinol self-expanding stents, stent fracture, in-stent restenosis, superficial femoral artery, peripheral arterial disease

____________________________________________

Maintenance of long-term patency after implantation of nitinol stents in the superficial femoral artery (SFA) continues to be one of the most challenging aspects of endovascular therapy.^1,2 Despite advances in femoropopliteal stent technology, in-stent restenosis (ISR) remains a common clinical problem, with an estimated rate of 20% to 50%.^3-5 One of the proposed mechanisms of ISR in the SFA is stent fracture (SF). Stent fracture in the SFA and popliteal arteries occurs due to various mechanical factors, such as vessel compression, torsion, and elongation, which are unique to this vascular bed.^6,7

Data from the literature provide contradictory evidence of the possible relationship between nitinol SF in the SFA and the development of ISR. Several studies have demonstrated that the presence of stent fractures increases the rate of ISR,^6,7 while other have failed to show statistically significant relationships between SF and ISR.⁸ 

In our practice, we have noted that stent fracture in the SFA does not always result in restenosis. Prompted by this clinical observation and questionable association between SFA ISR and SF, we sought to investigate a possible association between nitinol stent ISR, SF, and SF morphology. 

Methods 

We conducted a retrospective cohort study of 97 consecutive patients (105 limbs) treated for SFA ISR over a 5-year period from January 2008 to January 2013. Patients eligible for this study had to have undergone endovascular treatment with nitinol stent(s) in the SFA for occlusive lesions and subsequently developed SFA ISR with symptoms of lower limb ischemia. In-stent restenosis was suspected when a patient presented with clinical recurrence of ischemic symptoms in a previously stented limb along with the evidence of ISR on noninvasive imaging studies (duplex ultrasonography or CTA). Data from the initial SFA stenting procedure such as type and number of implanted stents, demographic characteristics, lesion type (total occlusion or stenosis), target-lesion length, and runoff score (the number of patent tibioperoneal vessels) were collected. 

During the SFA ISR revascularization procedure, all patients underwent fluoroscopy/CINE using at least 2 different orthogonal views for detection of SF in previously implanted SFA stents. Stent fracture was defined as clear interruption of stent struts (>1 mm) identified by fluoroscopy/CINE with resulting kink or misalignment along the axial length of the stent. Morphology of the stent fracture was classified based on Appendix A: the VIVA Physicians, Inc.⁹ Patients with Viabahn stent-grafts (W.L. Gore) were excluded.  

Angiographic restenosis was defined as >60% reduction of the vessel diameter by visual estimation. The exact location of each stent fracture and ISR in the vessel were visually assessed. All angiographic films were reviewed by 2 independent operators. Stent fracture was considered to be associated with ISR if it was located within 1 cm of the site of ISR and not associated if SF was present but located at non ISR segment. 

This study also assessed demographic and clinical characteristics, as well as angiographic and interventional data, that have been reported in the literature as factors that may potentially compromise stent patency. These variables included the length of stented segment, stent diameter, stent type, time to restenosis, weight, medications, glomerular filtration rate, white blood cell count, lipid and blood glucose levels, history of HTN, dyslipidemia, coronary artery disease, diabetes, hypertension, and smoking. 

A majority of the index procedures were performed through the contralateral femoral approach using 45 cm long 6 Fr or 7 Fr sheaths. Interventions were performed after a therapeutic dose of heparin was given to maintain an ACT greater than 200 throughout the procedure. After crossing the lesion with a wire, it was first dilated with 1:1 balloon dilatation, followed by implantation of a variety of self-expanding stents (stent and balloon choice as per operator discretion) that typically were sized 1 mm larger than the reference vessel diameter. Post dilation was performed with a balloon that was sized 1:1 to the reference vessel diameter. All studied patients, at the end of the procedure, had a good angiographic result with no significant residual stenosis. 

Statistical Analysis

Categorical variables are expressed as counts and percentages of patients. Chi-square test without Yate correction was performed to evaluate differences between categorical groups. Fisher exact test was used to examine the significance of the association (contingency) between two categories if there were 5 or fewer subjects. Continuous variables were reported as mean ± standard deviation (SD). To assess intergroup differences and calculate P values for continuous variables, the two-tailed, independent t test was used.  A P value ≤.05 was considered statistically significant. Univariate and multivariate logistic regression analyses were performed using SAS 9.2. All statistical analysis calculations were performed using SPSS (IBM SPSS Statistics, version 21) statistical software.

Results 

Baseline demographic and clinical characteristics are presented in Table 1. A total of 105 limbs with angiographically confirmed SFA ISR were analyzed. The mean age of the group was 73.3 years of age, 45% were females, and mean duration to ISR was 15.5 months. Seventy-four percent of the patients with ISR presented with lifestyle-limiting claudication, 10% with rest pain, and the remainder with critical limb ischemia with tissue loss (3%) or without tissue loss (11%). 

Angiographic and interventional data are presented in Table 2. In total, 205 SFA stents with a mean stented length of 201.7±103 mm were included in the final analysis. The mean number of stents implanted at the time of index procedure per limb was 1.97±0.92. Up to 4 stents were implanted in both groups: 37.1% of limbs were treated with 1 implanted stent, 35.2% with 2, 21.0% with 3, and 6.7% with 4. The most commonly used stent (62%) for initial implant was the Protégé stent (Covidien). In addition, a variety of other nitinol stents were studied.

Several limbs had more than one SF, which explains why the total number of SFs (44) exceeds the number of limbs (31). Sixty-four percent of limbs from the SF+ group demonstrated 1 fracture, 29% had 2 stent fractures in different segments of that limb and 6.5% of limbs developed 3 fractures. Table 2 demonstrates the percentage of stent fractures in various types of stents.

The mean length of the stented segment did not differ significantly between the two groups; 218.1±101 mm in the SF group (SF+) vs 194.8±103.2 mm in the non-SF group (SF-) (P=.29). On the other hand, larger stent diameter was associated with stent fracture: 6.5±0.5 mm in the SF+ vs 6.2±0.6 mm in the SF- group (P=.003). A majority of the patients in our study received more than one stent (63%). There was no difference in SF between patients who received 1 stent and those who received more than 1 stent (P=.83). The distribution of SF types is presented in Table 2. However, a conclusion regarding stent fracture rates in various nitinol stent types cannot be drawn from these data, because the assignment of stents was nonrandomized and the different types of stents were not equally represented. 

Stent fractures were observed more frequently in males (P=.036). Patients with SF had significantly higher weight than patients without SF (P=.0167). The number of smokers was significantly higher in SF+ when compared to SF- (55% vs 20% respectively, P=.0004). Univariate analysis demonstrated that current smoking (HR 2.75 [1.93-6.57], P=.0004), male gender (HR 2.77 [1.05-6.34], P=.039) and stent width (HR 1.99 [1.24-3.178], P=.004) were associated with SF. Multivariate analysis demonstrated that current smoking alone was significantly associated with SF (HR 2.01 [1.10-5.07], P=.032). 

In relation to the anatomic site, SF most commonly occurred in the distal SFA (50%), followed by the mid (32%) and the proximal (18%) segments of the SFA.

Of the 105 limbs with ISR, 31 limbs (29.5%) had angiographically confirmed SF. Of these 31 limbs, only 3 limbs had SF in the same anatomical segment as the ISR. In the remaining 102 limbs (90%), SF was not related to ISR. All three limbs with stent fracture associated ISR had type 3 fractures (Figure 1). The most common type of stenosis was diffuse in stent restenosis (Figure 2).

One of our patients who presented with ISR in the right SFA also had a previously implanted left SFA stent implanted 5 years prior. During angiography it was noted that the left SFA stent had a severe type 5 fracture with a pseudoaneurysm at the fracture site, with no associated restenosis or left leg symptoms (Figure 3). Another patient presented with severe thigh pain 24 hours after implantation of a nitinol self-expanding stent for prior total occlusion of the SFA. Fluoroscopy on presentation demonstrated a type 4 fracture and angiography demonstrated SFA perforation at the fracture site with active bleeding (Figure 4). This was successfully treated with a covered stent. These examples suggest that the severity and type of stent fracture may not always correlate with clinical presentation and symptoms. 

Discussion

Stent fracture is a well-known phenomenon in peripheral arterial disease, and it is the result of a sophisticated interaction between the high mobility of lower extremities, dynamic forces from surrounding musculature and flexion points, plaque morphology (substantial calcification and fibrosis), stent characteristics, and features of the stented segment that may result in external stent compression.¹⁰

Scheinert et al reported a stent fracture rate of 25% in their study, the mean lesion length was 157 mm.⁶ The observed stent fracture rate of 29.5% in our study is slightly higher. This is most likely due to a longer stented segment (mean 202 mm). 

In several studies, SF has been implicated as the cause of ISR. Iida et al retrospectively analyzed 585 patients (742 limbs) with 1,333 nitinol stents implanted in the de novo SFA lesions. SF was observed in 14% of limbs, and was significantly associated with early ISR (P=.0004).² Another study by Soga et al demonstrated a significantly higher SFA restenosis rate in patients with stent fractures (67% vs. 18%, P=.0034).¹¹ In both studies, it is unclear whether the stent fracture correlated with the anatomic site of ISR. In the study by Scheinert et al, SF was detected in 45 of 121 treated limbs and was associated with ISR. In their study, they found SF to be angiographically associated with ISR of >50% in 32.8% of limbs, complete occlusion in 34.4%, and no obstruction in 32.%.⁶ While these findings are compelling, the majority of stents investigated in this study were SMART stents, which constituted only 10% of the stents in our study. Whether the type of stent used has any effect on the association between SF and ISR is not well studied and will require direct head-to-head comparison.

On the other hand, the SF in the FAST study, the RESILIENT randomized control trial, and the SIROCCO I and SIROCCO II trials was not related to clinical impairment or ISR. The RESILIENT trial randomized patients to balloon angioplasty and Lifestent nitinol self-expanding stent (Bard Peripheral Vascular) arms. Of the 163 patients who underwent stenting, stent fracture was noted in 9 patients (4 type 1 and 5 type 4 SF), none of which were associated with ISR or target lesion revascularization.³ In the FAST study, which studied stent integrity in 83 patients who underwent SFA intervention with the Bard Luminexx stent, SF was noted in 10 patients. SF was not associated with ISR (20.0% vs 28.8% respectively, P=.719).¹² In the SIROCCO I trial, of the 43 patients who underwent SMART bare nitinol stenting, 8 developed SF at 18 months with no evidence of ISR.⁸ Similarly, in the SIROCCO II trial, 2 of the 50 patients randomized to bare nitinol stenting developed SF with no associated ISR.¹³

In our study, SF was only modestly associated with ISR, while a majority of ISR was unrelated to SF. Although a majority of stent fractures may not be clinically significant or result in ISR, they may occasionally result in complications such as pseudoaneurysms and perforation as illustrated earlier.

It has been hypothesized that stent overlap is one of the reasons for increased incidence of SF secondary to uneven stresses on certain areas of the stent. Schlager et al retrospectively evaluated 286 patients with Wallstents (Boston Scientific) nitinol stents used in 170 patients. They came to conclusion that the stented segment length was significantly associated with stent fractures (P=.046).⁷ The length of stented segment in our study was almost twice as long as in the Schlager et al study, and a majority of patients received more than one stent. In spite of the highly complex lesions, stent fracture was neither significantly associated with stented length nor with the number of stents used.

Scheinert et al prospectively studied 180 limbs treated with a variety of self-expanding nitinol stents for treatment of SFA obstructions. In their study, the type of stent significantly impacted the development of SF with rates ranging from 27% to 53% for various nitinol stents.⁶ Schlager et al also demonstrated that SF was related to the type of stent. In patients with Wallstents, there was no significant association between ISR and SF (P=.57). On the other hand, fractures showed a significant association with restenosis in patients with SMART stents (Cordis) (P=.008).⁷ Therefore, the incidence of SF and its association with ISR may be different for varied stent designs and need to be individually studied in a prospective fashion.  

There is an observed predisposition of the distal third of the SFA to deform more than the proximal or middle segments. This might be due to lower muscle tension proximal to the adductor canal and age-related loss of elasticity of the arterial wall.^14,15 This association has been demonstrated in a small clinical study with patients treated with the Luminexx where SF were angiographically confirmed in 11 of 40 limbs (28%), with a more frequent occurrence in the distal SFA (P=.04).10 The above findings were also confirmed by Davaine et al in a prospective study of 58 patients (62 limbs) treated with nitinol stents. Stent diameter (P=.04), lesion location in the distal SFA (P=.05), and length of femoropopliteal segment were positively associated with stent fractures.¹⁶ In our study, stent fractures were distributed unequally in the SFA with increased incidence of stent fractures in the distal segment of the SFA. 
The link between smoking and PAD was demonstrated more than 100 years ago. The occurrence of claudication was 3 times more common among smokers than nonsmokers.¹⁷ Our data demonstrated an increased risk of developing SF in both males and females if the patient was a smoker (P=.0004). There was an almost threefold higher prevalence of current smokers in SF+ group compare to SF- group (17 [55%] vs 15 [20%], respectively). Smokers may have more rigid arterial wall and decreased vessel elasticity compared to nonsmokers, which in turn may increase compression forces on the stent, resulting in higher incidence of SF. 

The association between male gender and stent fracture is an interesting one. This may possibly be related to a more active lifestyle and larger thigh muscle mass that increases dynamic compression forces on the stent structure. Iida et al demonstrated that patients who walked more than 5,000 steps a day had higher incidence of stent fracture (P=.0027).¹⁰

This study was a single-center retrospective analysis that had all limitations associated with this type of study design. However, the existing data on the association between SF and ISR is very limited and contradictory. To our knowledge, our study analyzed the largest number of patients with angiographically confirmed ISR and in our opinion is a valuable contribution to existing data. A randomized controlled trial would be an ideal approach to study this further, but this is not methodologically feasible. 

Our study demonstrates that in a group of unselected patients with clinically significant and angiographically confirmed ISR in nitinol self-expanding stents, SF was associated with ISR in only 10% of cases. This real world study shows that association between SF and ISR is very modest. The strongest predictors of stent fracture in our study were male sex, smoking, and larger stent diameter. In addition, SF occurred more commonly in distal SFA.  

Editor’s note: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. The authors report no disclosures related to the content herein.

Manuscript received September 11, 2015; manuscript accepted October 21, 2015.

Address for correspondence: Anvar Babaev, MD, PhD, NYU School of Medicine, 530 First Avenue, HCC 14, New York, NY 10016, United States. Email: anvar.babaev@nyumc.org  

References

1. Matsumura JS, Yamanouchi D, Goldstein JA, et al. The United States StuDy for EvalUating EndovasculaR TreAtments of Lesions in the Superficial Femoral Artery and Proximal Popliteal By usIng the Protege EverfLex NitInol STent SYstem II (DURABILITY II). J Vasc Surg. 2013;58(1):73-83 e1.
2. Iida O, Uematsu M, Soga Y, et al. Timing of the restenosis following nitinol stenting in the superficial femoral artery and the factors associated with early and late restenoses. Catheter Cardiovasc Interv. 2011;78(4):611-617.
3. 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.
4. Schillinger M, Sabeti S, Loewe C, et al. Balloon angioplasty versus implantation of nitinol stents in the superficial femoral artery. N Engl J Med. 2006;354(18):1879-1888.
5. 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.
6. 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.
7. Schlager O, Dick P, Sabeti S, et al. Long-segment SFA stenting--the dark sides: in-stent restenosis, clinical deterioration, and stent fractures. J Endovasc Ther. 2005;12(6):676-684.
8. Duda SH, Bosiers M, Lammer J, 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):701-710.
9. Allie DE, Hebert CJ, Walker CM. Nitinol stent fractures in the SFA: the biomechanical forces exerted on the SFA provide a “stiff” challenge to endovascular stenting. Endovascular Today. 2004; July/August: 22-34.
10. Iida O, Nanto S, Uematsu M, et al. Effect of exercise on frequency of stent fracture in the superficial femoral artery. Am J Cardiol. 2006;98(2):272-274.
11. Soga Y AK, Yamada T. Frequency and Impact of Stent Fractures after Stent Implantation of Superficial Femoral Artery. Circulation. 2006;114: II_683. 
12. Krankenberg H, Schluter M, Steinkamp HJet 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. 
13. Duda SH, Bosiers M, Lammer J, 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):331-338. 
14. Young MD, Streicher MC, Beck RJ, van den Bogert AJ, Tajaddini A, Davis BL. Simulation of lower limb axial arterial length change during locomotion. J Biomech. 2012;45(8):1485-1490. 
15. Cheng CP, Choi G, Herfkens RJ, Taylor CA. The effect of aging on deformations of the superficial femoral artery resulting from hip and knee flexion: potential clinical implications. J Vasc Interv Radiol. 2010;21(2):195-202.
16. Davaine JM, Querat J, Guyomarch B, et al. Incidence and the clinical impact of stent fractures after primary stenting for TASC C and D femoropopliteal lesions at 1 year. Eur J Vasc Endovasc Surg. 2013;46(2):201-212.
17. Norgren L, Hiatt WR, Dormandy JA, Nehler MR, Harris KA, Fowkes FG. Inter-Society Consensus for the Management of Peripheral Arterial Disease (TASC II). J Vasc Surg. 2007;45 Suppl S:S5-S67.