Author + information
- Received September 9, 2008
- Revision received February 19, 2009
- Accepted April 19, 2009
- Published online July 1, 2009.
- Osamu Iida, MD,
- Shinsuke Nanto, MD, PhD⁎ (, )
- Masaaki Uematsu, MD, PhD,
- Kuniyasu Ikeoka, MD,
- Shin Okamoto, MD and
- Seiki Nagata, MD, PhD
- ↵⁎Reprint requests and correspondence:
Dr. Shinsuke Nanto, Cardiovascular Division, Kansai Rosai Hospital, 3-1-69 Inabaso, Amagasaki 660-8511, Japan
Objectives We investigated the time course of stent patency in the femoro-popliteal artery for as long as 4 years.
Background Stent fracture has been related to poor 2-year patency in the femoro-popliteal artery.
Methods We studied 239 consecutive patients who underwent provisional de novo stenting with nitinol stents for 333 limbs (Luminexx stent [C. R. Bard, Inc., Murray Hill, New Jersey] in 91 limbs; Smart stent [Cordis Corp., Miami Lakes, Florida] in 242 limbs) from April 2004 to December 2007. Stent fracture was determined by X-ray with multiple projections. Patency was assessed by duplex ultrasonography as peak systolic velocity ratio <2.4 or by angiography (% diameter stenosis <50%). Primary patency in those with and without stent fracture at follow-up was assessed along with factors influencing stent fracture.
Results Primary patency was 81%, 74%, 68%, and 65% at 1, 2, 3, and 4 years, respectively. Stent fracture occurred in 14% (78 of 544) per stent and 17% (55 of 333) per limbs. Stent fracture was significantly associated with multiple stent deployments (with fracture = 2.3 ± 0.9 stents vs. without fracture = 1.5 ± 0.7 stents, p < 0.001) and long lesions (with fracture = 208 ± 84 mm vs. without fracture = 121 ± 79 mm, p < 0.001). Primary patency was 68% with fracture versus 83% without fracture at 1 year, p = 0.03; 65% versus 75% at 2 years, p = 0.05; 61% versus 69% at 3 years, p = 0.06; and 61% versus 65% at 4 years, p = 0.07. Neither type 1 nor type 3 fracture affected patency, although type 2 showed the worst patency.
Conclusions Stent fracture worsened the patency during the first 2 years, but it did not apparently affect patency beyond 2 years. In particular, complete stent separation did not affect patency.
The technical and clinical success rates of the endovascular therapy (EVT) for the stenotic and occlusive lesions in the femoro-popliteal artery (FPA) have reached over 95% due to the improvements of new generation devices (1,2). Long-term patency, however, remains as an unsolved issue. Balloon angioplasty alone in the FPA demonstrated low primary patency rates of 40% to 60% after 6 to 12 months (3,4) depending on the clinical stages, lesion severity, and run-off in below-the-knee arteries (1,2). Early experiences with stainless-steel stents showed no beneficial efficacy on the primary patency compared with angioplasty alone (5,6). Accordingly, in the TransAtlantic Inter-Society Consensus (TASC) 2000, stenting was relegated to a bailing-out procedure in the event of failed balloon angioplasty (1).
Recently, nitinol stents have been introduced, which have demonstrated superior primary patency to balloon angioplasty (7,8) or to steel stents (9) in the superficial femoral artery (SFA). Hence, EVT with nitinol stents has become widely applied in the FPA lesion. Despite the improved primary patency, however, stent fracture after deployment of nitinol stents has been recognized as an adverse event in the FPA, where the mechanical factors, such as external forces, vessel compression, torsion, and elongation. Stent fracture was considered as a risk of in-stent restenosis and reocclusion in the SFA stenting in the chronic phase (10), just as it was suggested to be a cause of restenosis in the coronary artery (11). Nonetheless, in our clinical practice experience, stent fracture does not always result in restenosis in the FPA. Although the 2-year patency with and without stent fracture in the FPA has been documented earlier (10), the time course of patency with and without stent fracture beyond 2 years remains unknown. Thus, in this study, we sought to investigate the patency after nitinol stenting in the FPA in relation to stent fracture and fracture morphology over a 4-year period.
We retrospectively analyzed a patient database prospectively maintained from April 2004 to December 2007 that included the 239 consecutive patients who successfully underwent EVT for de novo FPA lesions with provisional nitinol stenting in 333 limbs, and who also gave consent to receive follow-up ultrasound and X-ray in the Cardiovascular Division of Kansai Rosai Hospital. All patients had symptoms due to FPA lesions (Rutherford stage 2 to 6) that affected the quality of life in spite of excises and medications. We excluded those presenting with acute or subacute limb ischemia with nitinol stents previously deployed in the FPA, and those who underwent stenting in the restenotic lesions. We developed the study protocol in accordance with the Declaration of Helsinki, and it was approved by the ethics committee of Kansai Rosai Hospital. All patients gave written informed consent.
After the initial diagnostic angiograms of the lower limb, the indication of EVT was judged by consulting with vascular surgeons. Indication of EVT for the FPA lesions included >70% of vessel diameter stenosis without inflow lesions. Approaches for EVT in the FPA lesions were determined by the operator's discretion. In general, after a 6-F sheath insertion, either a 0.035- or a 0.014-inch wire was advanced into the lesion; unfractionated heparin (5,000 U) was injected into the artery. The lesion was expanded using an optimal balloon for 60 s. A nitinol stent was implanted in accordance with the American College of Cardiology/American Heart Association guidelines in patients presenting with a residual pressure gradient >10 mm Hg, residual stenosis >30%, and/or flow-limiting dissection after balloon dilation. Self-expandable stents with the diameter approximately 2-mm larger than the reference diameter were used. Stents of 8 mm in diameter were used in most cases. Selection of stents was dependent on the terms: Luminexx stents (Bard, Inc., Murray Hill, New Jersey) were used from April 2004 to December 2005, and mean follow-up duration was 28 ± 19 months; Smart stents (Cordis Corp., Miami Lakes, Florida) were used from January 2006 to December 2007, and mean follow-up duration was 17 ± 9 months. Dual antiplatelet therapy (aspirin 100 mg/day, ticlopidine 200 mg/day, or cilostazol 200 mg/day) was started at least 1 week before EVT and continued until the end of follow-up.
Follow-up and outcomes
Clinical evaluations including symptom changes, ankle brachial index, lesion patency evaluated by ultrasound, and stent fracture assessed by X-ray were performed at baseline, approximately at 24 h, 72 h, 1 month, and at every 3 months after procedure. Primary patency was defined as peak systolic velocity ratio <2.4 by duplex ultrasound (12). Stent fracture was defined as clear interruption (>1 to 2 mm) of stent struts identified by X-ray from 4 projections, 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., Research Program of Rocha-Singh et al. (13). In brief, a single-tine fracture was defined as type 1; multiple-tine fractures, type 2; stent fracture(s) with preserved alignment of the components, type 3; stent fracture(s) with mal-alignment of the components, type 4; stent fracture(s) in a trans-axial spiral configuration, type 5. An X-ray from 4 projections was performed at each follow-up period to evaluate the stent fracture by 2 observers who had the experience of more than 500 EVT cases within a year. If different grades of stent fracture were observed within a stent, the severest grade was adopted. Bent knee radiography was not performed. The outcomes of this study were overall primary patency, patency associated with and without stent fracture, patency according to the types of nitinol stents, and the morphology of stent fracture during the follow-up period.
Statistical analysis was performed using SPSS (SPSS Inc., Chicago, Illinois). Data are shown as mean ± standard deviation. Unpaired t test was used to compare the continuous variables between the groups, and chi-square test was used to compare ratios between the groups, with the statistical significance level set at a value of p < 0.05. Primary patency was determined with Kaplan-Meier survival analysis and compared by log-rank testing.
Patient characteristics are shown in Table 1. There were no significant differences in patient characteristics including atherosclerosis risk factors between those with and without stent fracture. During follow-up, 13% (30 of 239) of patients died, and 10 limbs were amputated in 9 patients.
Lower limb and lesion characteristics
Lower limb and lesion characteristics are shown in Table 2. Lesions involving calcification were defined by angiography as readily visible densities noted within the apparent vascular wall. Those showing stent fracture involved less critical limb ischemia, and lower Rutherford stages than those not showing stent fracture. Lesions associated with stent fracture included more severe forms of TASC classification, longer lesion length, and more frequent chronic total occlusions than the lesions without stent fracture.
The intervention procedure is shown in Table 3. Four hundred and seven Smart stents were used in 242 limbs, and 137 Luminexx stents were used in 91 limbs. Stent fracture occurred in 14% (78 of 544) of stents, and 17% (55 of 333) of limbs during the follow-up. Between the different types of nitinol stents, Smart stents showed less fracture rate (13%, 31 of 242) than Luminexx stents (26%, 24 of 91, p = 0.003). The number of stents used per vessel was greater in those with stent fracture than in those without fracture.
Morphology of the stent fracture
Morphology of the stent fracture was as follows: type 1 in 10 limbs, type 2 in 32 limbs, and type 3 in 13 limbs. None of the patients had type 4 nor type 5 stent fracture.
Overall primary patency per limb is shown in Figure 1. The primary patency remained 65% at 4 years. Primary patency in those associated with stent fracture and in those without stent fracture is shown in Figure 2. Primary patency was worse in those associated with stent fracture than in those without stent fracture at 1 year (68% with fracture vs. 83% without fracture, p = 0.03), but it was not significantly different between the groups beyond 2 years (65% vs. 75% at 2 years, p = 0.05; 61% vs. 69% at 3 years, p = 0.06; and 61% vs. 65% at 4 years, p = 0.07). Primary patency with and without stent fracture in terms of different types of nitinol stents was also evaluated. Primary patency of the Luminexx stent with and without stent fracture was 61% versus 84% at 1 year, p = 0.04; 61% versus 80% at 2 years, p = 0.05; 56% versus 75% at 3 years, p = 0.08; and 56% versus 70% at 4 years, p = 0.09. Primary patency of the Smart stent with and without stent fracture was 75% versus 83% at 1 year, p = 0.26; and 69% versus 71% at 2 years, p = 0.26. Types of nitinol stents did not affect the primary patency during follow-up. The effects of stent fracture morphology on the primary patency are shown in Figure 3. Although type 2 (multistrut) stent fracture revealed poor outcomes (p < 0.01), neither type 1 (single strut, p = 0.68) nor type 3 (complete separation, p = 0.18) affected the primary patency compared with those without stent fracture.
Of 544 nitinol stents implanted in 333 limbs, stent fracture occurred in 14% (78 of 544) of stents, and in 17% (55 of 333) of limbs during the follow-up period of 4 years. Primary patency was 81%, 74%, 68%, and 65% at 12, 24, 36, and 48 months, respectively, which was similar to earlier studies (8,9). Stent fracture was associated with long lesions, and was also associated with the number of stents used. Stent fracture worsened the patency during the first 2 years, but did not apparently affect the patency beyond 2 years. Although type 2 stent fracture revealed poor outcomes, neither type 1 nor type 3 stent fracture affected the primary patency compared with those without stent fracture.
Nitinol stents have become widely applied not only in the simple TASC A and B lesions, but also in the complex TASC C and D lesions in the FPA (14,15). There is increasing evidence that patency of nitinol stents is superior to those of early metallic stents, and to balloon angioplasty alone (7–9). Improvement in long-term patency by using nitinol stents was expected. Earlier reports, however, suggested long lesion length and a large number of nitinol stents were the predictors of nitinol stent fracture in the FPA, and subsequently resulted in stent failure in the chronic phase (10). We have also reported that vigorous exercise adversely affected stent fracture in patients implanted with nitinol stents in the SFA (16). The findings that claudicaton was more prevalent in those with stent fracture than in those without fracture in this study and that stent fracture was significantly associated with long lesions were compatible with these previous observations.
Although long-term patency with respect to stent fracture after EVT is a clinically important issue, none of the earlier studies have addressed the prognosis beyond 2 years. This study is the first to investigate the outcomes of stent fracture beyond 2 years in the new nitinol stent era in relation to the morphology of stent fracture in a single center experience using only nitinol stents.
The incidence of stent fracture in this study was apparently lower than that observed in the FESTO (FEmoral STenting in Obstructions) trial (10). This may be attributed to the differences in life-styles and physique in the Eastern population. In fact, body mass index in patients included in this study was approximately 22 kg/m2. It may be interesting to speculate that stent fracture is adversely affected by the muscle volume through compression and expansion mechanisms. Alternatively, underestimation of stent fracture may occur in severely calcified lesions, where the stents tended to be underexpanded.
Study limitations and weaknesses
This study uses a retrospective and nonrandomized analysis in spite of using a prospectively maintained database and lack of sufficient patients followed in a single center. Only 2 types of nitinol stents were used in this study, which were the only nitinol stents available for clinical use in Japan. The difference in the structure of segments and number of connecting bridges between these 2 stents may be responsible for the different stent fracture rate and the patency. Further investigation is needed using new generation stents in the future.
It is interesting to speculate that there may be a difference in the outcomes depending on the morphology of stent fracture. Benign stent fracture may include type 1 and 3 fractures, and potentially harmful stent fracture includes type 2.
Stent fracture worsened the patency during the first 2 years, but it did not apparently affect patency beyond 2 years, although the study lacked statistical power to confirm this. A large-scale, multicenter, prospective study on this subject is warranted based on the results of this study.
The authors acknowledge the expertise of Drs. Hirokuni Akahori, Kenji Kawamoto, Masamichi Yano, Nobuaki Tanaka, and Haruyo Yasui in performing catheterization.
- Abbreviations and Acronyms
- endovascular therapy
- femoro-popliteal artery
- superficial femoral artery
- TransAtlantic Inter-Society Consensus
- Received September 9, 2008.
- Revision received February 19, 2009.
- Accepted April 19, 2009.
- American College of Cardiology Foundation
- Norgren L.,
- Hiatt W.R.,
- Dormandy J.A.,
- et al.,
- TASC II Working Group
- Minar E.,
- Pokrajac B.,
- Maca T.,
- et al.
- Schillinger M.,
- Sabeti S.,
- Dick P.,
- et al.
- Scheinert D.,
- Scheinert S.,
- Sax J.,
- et al.