Selective stenting in subintimal angioplasty: Analysis of primary stent outcomes
Article Outline
Objective
Subintimal angioplasty (SIA) is being increasingly utilized to treat chronic arterial occlusions. The role of stents in SIA is currently unknown. We performed a retrospective review of selective stent use in SIA to assess outcomes and factors affecting these results.
Methods
A retrospective review of patient information—including demographics, indications, procedures, noninvasive studies, and post-procedural events—was performed on our database for patients undergoing SIA in the superficial femoral and popliteal arteries. Outcomes were calculated only on technically successful SIAs using Kaplan-Meier survival analysis. Continuous and non-continuous data were compared using the Student t test and the z test, respectively. Survival curves were compared using log-rank testing for univariate analysis and Cox hazard-regression analysis for multivariate analysis.
Results
Three-hundred-sixty-eight patients (382 limbs) underwent femoral and/or popliteal SIA for critical limb ischemia or disabling claudication from December 1, 2002 through July 31, 2006. Eighty-four limbs (22%) had a stent placed, while 298 (78%) did not receive a stent. Mean follow-up was 11.7 months (range, 0-45 months). One-year primary and secondary patency for stent vs no-stent group was 50% vs 45% (P = .73) and 70% vs 78% (P = .47), respectively. One-year limb salvage rate for the stent vs no-stent group was 85% vs 90% (P = .61). At 2 years, patients receiving a stent are more likely to undergo open bypass than those without a stent (P = .06). Eighty-three patients underwent 84 SIA with stent placement. The mean number of stents for each case was 1.4 ± 0.7. Univariate analysis revealed that previous ipsilateral bypass surgery significantly decreased 1-year patency: 35% vs 56% (P = .05). SIA performed for disabling claudication had a trend toward improved 1-year patency 58% vs 39% for critical limb ischemia (P = .09). A stent diameter ≥7 mm displayed a trend toward better patency 53% vs 37% for diameter ≤6 mm (P = .08). None of these factors proved significant with multivariate analysis.
Conclusion
Selective stents placed for suboptimal results after subintimal angioplasty produce similar patency rates to primary SIA without stents. Patients receiving stents with prior lower extremity bypass surgery will have worse outcomes than those without. Use of a stent diameter ≤6 mm and indication of critical limb ischemia will likely produce worse results. It appears that other stent variables (location, number, length, and overlap) do not alter patency. Finally, selective stent use after SIA provides excellent limb salvage.
Endovascular treatment for infrainguinal arterial occlusive disease has been rapidly expanding over the last decade. One technique, subintimal angioplasty (SIA), has been increasingly utilized since its inception by Bolia in 1987.1 This technique creates a subintimal channel by dissection and angioplasty. Over the last 20 years, stents have been utilized for suboptimal results and, more recently, routinely after SIA. Analysis of our first 105 SIAs with selective stent use revealed a primary patency of 55% and 35% at 1-year and 3-years, respectively.2 Limb salvage rate of 78% was seen at 3 years.
Several studies have reported the use of stents both selectively and routinely to supplement intraluminal balloon angioplasty (IBA). Recent reports have supported routine stenting after IBA.3, 4, 5 Other studies of IBA have cautioned about their use in longer, complex lesions.6 Whether this intraluminal data can be extrapolated to SIA and the factors that may impact the patency of stents in SIA are still unknown. We performed a retrospective review of our selective stent use in SIAs to assess outcomes and the factors which may affect these results.
Materials and methods
A retrospective review was performed on patients with critical limb ischemia or disabling claudication who underwent a primary SIA from December 1, 2002 through July 31, 2006. Patient demographics, clinic notes, noninvasive vascular studies, angiographic findings, and operative reports were reviewed after approval by the Institutional Review Board.
During this period, 639 consecutive SIAs were performed on 591 patients with chronic arterial occlusive disease. The occluded segments ranged from the common iliac artery to the tibial arteries. This study focuses only on those limbs in which a successful SIA was performed in the superficial femoral and popliteal segments; hence, evaluating results of 382 limbs in 368 patients. In our practice, patients with disabling claudication or critical limb ischemia are routinely offered endovascular therapy as a first-line therapy for chronic arterial occlusions, and patients invariably choose this option after discussion of surgical alternatives.
All procedures were performed by vascular surgeons. Most procedures were performed in an angiographic suite and the remaining in an operating room with non-fixed fluoroscopy if deep sedation or general anesthesia was required or if a concomitant open operation was required. There was no established protocol for the SIA technique or use of stents. The general principles of creating a SIA channel and use of a stent for a suboptimal result were practiced by all physicians. The common femoral artery contralateral to the treated limb was the preferred means of angiographic access. After angiography and identification of an arterial occlusion, patients were systemically heparinized. The occluded segments were typically approached by placement of a longer sheath over the aortic bifurcation and in proximity to the occlusion. A soft, hydrophilic 0.035-inch guidewire in combination with a 4F or 5F angled hydrophilic catheter (Glidecath, Terumo Medical Corporation, Somerset, NJ) were the most commonly used to perform the subintimal dissection. Other common catheters utilized during SIA include the Bernstein catheter (Cook Medical; Bloomington, Ind) and the Quick-Cross catheter (Spectranetics Corporation, Colorado Springs, Colo). Passage of the guidewire along the medial, lateral, anterior, or posterior border of the arterial wall, guidewire advancement in a helical course across the lesion, and injection of contrast demonstrating a dissection plane were all indications of subintimal guidewire location. After confirmation of catheter re-entry into the true lumen just distal to the arterial occlusion, balloon angioplasty was used to dilate the subintimal channel. After SIA was completed, an angiogram was performed to assess the results. Stents were selectively deployed within segments of the channel for: (1) suboptimal angioplasty, defined as residual stenosis greater than 30%; (2) dissection flaps; and (3) calcification. Calcification was used to denote lesions which were very difficult to traverse in a subintimal channel and could visualize severe calcified plaque on angiogram. Also, stents were deployed only in suboptimal segments. When two or more stents were used, they could be placed in a contiguous and non-contiguous position depending upon focality and location of the suboptimal segment. Self-expanding nitinol stents were utilized for the majority of cases (n = 111; 92%); the remaining cases used balloon-expandable stents (n = 4; 3%) or stent type could not be identified (5%). Balloon expandable stents were utilized when accurate deployment was necessary, such as at the bifurcation of the common femoral artery into the superficial femoral artery (SFA) and profunda femoral artery.
Adjunctive procedures utilized during the procedure for the “stent” group included mechanical atherectomy (2), mechanical thrombectomy (2), thrombolytic infusion (3), cryoplasty (2), and laser atherectomy (2). Adjunctive procedures were utilized to address a suboptimal result determined by angiography. Approximately 3% of the no-stent group required an adjunctive procedure; whereas, 8% in the stent group (excluding tissue plasminogen activator [tPA]). Technical success was defined as the creation of a subintimal channel bypassing the occlusion, with successful re-entry into the true lumen and subsequent angioplasty. Our immediate technical success has been previously published at 87%.2 For this study, only patients with a technically successful SIA were evaluated. After the procedure, patients received clopidogrel for at least 1 month and aspirin indefinitely. Patients were allowed to resume ambulation 6 to 8 hours after the procedure and were typically discharged home within 24 hours.
Clinical follow-up at 1 month, 3 months, and 6 to 12 months after the procedure was routine and included physical examination and measurement of ankle-brachial indices (ABI). Duplex scan examination of the subintimal channel and further follow-up were obtained at the discretion of the treating surgeon. The majority of surveillance for patency was obtained through duplex scan examination and ABIs; the rest was obtained via physical exam. “Non-operative” candidates were defined as either prohibitive surgical risk secondary to medical condition or were without surgical bypass options. Any additional endovascular procedures to maintain or restore patency of the subintimal channel were recorded, as were all open surgical revisions, bypasses, and major amputations performed through July 31, 2007.
Patency of the SIA was defined by at least one of the following criteria: flow through the vessel demonstrated by angiography or duplex ultrasonography scan, maintenance of an ABI greater than 0.10 above the pre-procedural value, or maintenance of a palpable pedal pulse that was absent before the procedure in an asymptomatic patient. Resolution of symptoms was not considered an indication of patency. Any follow-up intervention necessitating open surgical revision or bypass was reported as such and was not included in primary assisted or secondary patency. Symptomatic improvement in patients with critical limb ischemia was defined as the resolution of rest pain or healing of ulcers and gangrenous wounds after debridement or minor amputation. Symptomatic improvement in patients with claudication was defined as improvement in walking distance, as determined through follow-up visits. Loss of symptomatic improvement was used to calculate the maintenance of claudication relief.
Continuous data are expressed as mean standard deviation (SD) and were compared by using the Student t test. Non-continuous data are expressed as percentages and were compared by using the z test comparison for proportions. A P < .05 was considered statistically significant. Patency, limb salvage, symptomatic improvement, and freedom from surgical bypass were determined with Kaplan-Meier survival analysis and compared by log-rank testing. Multivariate analysis was performed by Cox proportional-hazards regression. Patency is presented by efficacy analysis and not intention to treat.
Results
Analysis of no-stent vs stent
Three hundred sixty-eight patients (382 limbs) underwent femoral and/or popliteal SIA for critical limb ischemia or disabling claudication from December 1, 2002 through July 31, 2006. Eighty-four limbs (22%) had a stent placed, while 298 (78%) did not receive a stent. Mean follow-up was 11.7 months (range, 0-45 months). Patient demographics are presented in Table I. Patients who received a stent had a mean age of 67.6 ± 13.2, while those who did not receive a stent had a mean age of 69.2 ± 11.7 (P = NS). The stent group had statistically significant more patients with hyperlipidemia (48% vs 35% [P < .05]) and a history of smoking than the non-stent group (58% vs 46% [P = .05]). There was also a trend for more patients in the stent group who had undergone a prior lower extremity bypass than the non-stent group (27% vs 18% [P = .07]).
Table I. Patient demographics and risk factors for peripheral arterial disease
| Variable | No-stent (limbs-298; 285 pts)† | Stent (limbs-84; 83 pts) | P-value |
|---|---|---|---|
| Age (years) | |||
 Mean ± SD | 69.2±11.7 | 67.6±13.2 | .27 |
| Range | 38.4-98.9 | 40.6-98.4 | — |
| Male gender | 52%(155) | 61%(51) | .16 |
| Indication | |||
| Critical limb ischemia | 58%(172) | 49%(41) | .15 |
| Rest pain | 24%(70) | 20%(17) | .53 |
| Ulceration | 25%(75) | 17%(14) | .10 |
| Gangrene | 9%(27) | 12%(10) | .44 |
| Claudication | 42%(125) | 50%(42) | .19 |
| Risk factors | |||
| Hypertension | 68%(194) | 71%(58) | .76 |
| Diabetes mellitus | 50%(142) | 43%(36) | .32 |
| Coronary artery disease | 53%(152) | 48%(40) | .35 |
| History of smoking | 46%(132) | 59%(49) | .05 |
| End-stage renal disease | 11%(32) | 12%(10) | .67 |
| Hyperlipidemia | 35%(99) | 47%(39) | .04 |
| Previous LE bypass | 18%(52) | 27%(22) | .07 |
| Nonoperative candidate | 14%(41) | 10%(8) | .27 |
| Location of SIA | |||
| SFA only | 50%(148) | 57%(48) | .23 |
| SFA-popliteal | 47%(139) | 39%(33) | .23 |
| Popliteal only | 3%(11) | 4%(3) | .96 |
†Patient demographics calculated by number of patients; procedural characteristics calculated by number of limbs. |
One-year primary patency for the stent vs the no-stent group was 50% vs 45%, respectively (P = .73) (Fig 1). The 1-year secondary patency for the stent vs the no-stent group was 70% vs 78% (P = .47), respectively (Fig 2). Twenty-eight percent (n = 79) of patients in the no-stent group underwent a re-intervention, while 25% (n = 21) in the stent group required a re-intervention (P = NS). There was no statistically significant difference in technical success rates between the two groups. In patients with critical limb ischemia, the 1-year limb salvage rate for the stent vs the no-stent group was 85% vs 90% (P = .61), respectively (Fig 3). Only 1 patient treated for claudication required an amputation 35 months after the procedure. Maintenance of claudication relief for the stent vs the no-stent group was 83% vs 89% (P = .54). At 2 years, there was a trend for better freedom from bypass surgery in the no-stent group vs the stent group (83% vs 65%, respectively; P = .06) (Fig 4).
Fig 1.
Primary patency of stent vs no-stent group. One-year patency stent 50% vs no-stent 45% (P = .73).
Fig 2.
Secondary patency of stent vs no-stent group. One-year patency stent 70% vs 77% (P = .47).
Fig 3.
Limb salvage for CLI patients of stent vs no-stent group. One-year stent 85% vs no-stent 90% (P = .61).
Fig 4.
Freedom from bypass surgery. Two-year rate stent (65%) vs no-stent (83%) (Log rank P = .06).
Two patients (2.3%) in the stent group encountered procedural complications which both required endovascular therapy; a distal embolus treated with thrombolysis and a fractured glidewire retrieved with a snare wire. Eleven patients (3.7%) in the no-stent group had procedural complications: hematoma (3), pseudoaneurysm (2), arterial perforation (2), distal embolus (1), arteriovenous fistula (1), wound infection (1), and a contralateral iliac artery dissection (1). Two complications required surgical intervention; evacuation of a retroperitoneal hematoma and repair of an iliac artery dissection. Periprocedural mortality was similar between the two groups (1.2% stent and 0.7% no-stent; P = NS). During the study, there were 50 deaths; 6 in the stent group and 44 in the no-stent group. The majority of deaths occurred in patients treated for critical limb ischemia (44), whereas only six deaths occurred in patients with claudication. The average time to death was 10.7 ± 9.7 months (range, 0.2-35.2).
Analysis of stent group
Eighty-three patients underwent 84 SIA with primary stent placement. Stent characteristics are presented in Table II. The mean number of stents used for each case was 1.4 ± 0.7 (range, 1-5). The mean stent diameter utilized during the study was 7 mm (range, 5 mm-8 mm). The total length of the stent used in each case averaged 89 mm ± 62 (range, 15 mm-300 mm). In those patients who had multiple stents deployed, 73% required overlap. The other 27% with multiple stents did not overlap secondary to stent placement in non-contiguous, treated segments. For example, two stents placed separately in the proximal and distal segment of the SFA would not overlap.
Table II. Characteristics of stent group
| Variable | Data |
|---|---|
| Number of stents | |
| Mean | 1.4±0.7 |
| Range | 1-5 |
| Stent diameter | |
| Mean | 7.0mm±0.9 |
| Range | 5mm-8mm |
| Stent length | |
| Mean | 89mm±62 |
| Range | 15mm-300mm |
| Stent location⁎ | |
| SFA | 79%(67) |
| SFA-popliteal | 10%(8) |
| Popliteal | 11%(9) |
| Stent overlap† | 73%(22) |
| Indication for stent⁎⁎ | |
| Suboptimal angioplasty | 60%(50) |
| Dissection flap | 38%(32) |
| Severe calcification | 8%(7) |
⁎SFA (superficial femoral artery) indicates stents place in the proximal, mid, and distal segments of SFA. |
†Stent overlap-calculated for patients with multiple stents (n = 30). |
⁎⁎Some limbs had more than one indication for the stent, ex. Suboptimal result and severe calcification in the same or separate segments. |
The locations of stent deployment were the superficial femoral artery only 79% (67), the SFA-popliteal segments 10% (8), and the popliteal artery only 11% (9). For purposes of patency analysis, the femoro-popliteal segments were divided into: (1) proximal (SFA-proximal and middle segments; n = 34) and (2) distal (SFA-distal and popliteal; n = 39). Eleven patients were not included in this data; 7 had stents placed throughout the channel, while the other 4 patients had stents placed at the proximal and distal aspects of the SIA channel. The indications for stent use were suboptimal angioplasty (60%), dissection flap (38%), and severe calcification (8%). In some cases, multiple indications were given for the same segment or for two separate lesions. The amount of calcium in the occluded arterial segment was not quantified. Also, calcification was always used in conjunction with a suboptimal angioplasty as an indication for stent use. Univariate analysis revealed that a patient who had a previous, ipsilateral, lower extremity bypass surgery had significantly decreased 1-year patency (35%) vs those without a previous bypass (56%) [P = .05]. Also, patients who had a SIA performed for disabling claudication had a trend toward improved 1-year patency (58%) vs critical limb ischemia (39%) (P = .09). In addition, a stent diameter ≥7 mm displayed a trend toward better patency 53% vs 37% for a stent diameter ≤6 mm (P = .08). Univariate analysis did not reveal any other procedural or demographic factors which affected patency rates in SIA with stents (Table III). Multivariate analysis did not reveal any significant factors affecting patency.
Table III. Univariate analysis of risk factors for loss of primary patency with stent use
| Risk factor | HR | 95% CI | P-value |
|---|---|---|---|
| CAD | 0.71 | 0.35-1.39 | .31 |
| Female | 1.36 | 0.70-2.71 | .35 |
| African American | 0.77 | 0.39-1.52 | .46 |
| HTN | 1.25 | 0.58-2.70 | .57 |
| Age >80 | 0.76 | 0.62-2.74 | .49 |
| Smoking | 1.14 | 0.58-2.22 | .70 |
| Hyperlipidemia | 1.39 | 0.71-2.78 | .33 |
| End-stage renal disease | 0.99 | 0.80-3.23 | .99 |
| Diabetes mellitus | 1.12 | 0.44-1.76 | .72 |
| Prior bypass surgery | 1.92 | 0.99-4.76 | .05 |
| Critical limb ischemia | 1.75 | 0.91-3.56 | .09 |
| Non-operative candidate | 1.88 | 0.68-7.90 | .18 |
| Stent length | |||
| <60 | 0.81 | 0.40-1.62 | .55 |
| >100 | 1.14 | 0.63-2.11 | .93 |
| >150 | 1.32 | 0.61-3.11 | .44 |
| Stent number >2 | 1.43 | 0.73-2.94 | .28 |
| Stent location- distal | 0.79 | 0.37-1.59 | .48 |
| Stent overlap | 1.20 | 0.38-1.77 | .61 |
| Stent diameter <6 mm | 1.82 | 0.93-4.06 | .08 |
| Stent indication - dissection | 1.55 | 0.80-3.22 | .18 |
Discussion
Endovascular therapy of peripheral arterial disease continues to expand into more complex lesions. One technique, SIA, is increasingly being used for such therapy. Previously published series have shown 1-year primary patency and limb salvage rates of 51-80% and 85-94%, respectively.1, 2, 7, 8, 9 Our results in this study affirm these outcomes and show that there is no statistical significant difference between femoro-popliteal SIA without stents and with stents placed selectively.
While the role of stent placement after SIA is unknown and has not been widely studied, its role in intraluminal therapy has been evaluated. Only two published studies on SIA report use of stents and their impact. The largest reported experience is from Treiman and all other studies are hard to interpret their results. Physicians are currently using intraluminal stent data to justify stent use in SIA. Schillinger et al4 reported decreased rates of restenosis at 6 months (23% vs 43%) and 1 year (37% vs 63%) in patients receiving stents vs those with just balloon angioplasty in the superficial femoral artery; however, there were no differences between the groups for relief of claudication or limb salvage. Lower rates of re-intervention and recurrent stenosis in the femoro-popliteal segments have been seen up to 2 years in primary stenting after angioplasty compared to angioplasty alone.5 Ferreira et al3 have also reported 5-year assisted patency rates of 90% after IBA with the stent in the SFA. It appears that stents aid patency rates; however, other studies have shown their limitations. Surowiec et al6 have shown that patency rates in the SFA after IBA with stents decrease as the Trans-Atlantic Inter-Society Classification (TASC) grade increases. In addition, the femoral artery stenting trial (FAST) trial has shown no benefit of stent over IBA at 1 year for short SFA lesions.10 As we continue to define the role of stents in IBA, the applicability of these results to SIA remain questionable. First, the lesion morphology after angioplasty is different. The atherosclerotic plaque remains in the flow channel after intraluminal angioplasty, whereas a subintimal flow channel is devoid of exposed plaque except at entry and re-entry points. This could alter arterial wall remodeling after SIA from those observed with IBA. Hence, the role of neo-intimal hyperplasia after stent placement in a subintimal channel is unknown. Second, the types of lesions in each group are different. Whereas IBA can be used for segments with stenosis or short occlusions, SIA is used for segments with short or long occlusions. Also, as suggested by the worsening stent results with increasing TASC lesion, the majority of SIA are performed for TASC C and D lesions. Hence, the results from IBA stents can't be readily extrapolated to SIA.
Our results have shown that prior bypass surgery can adversely affect outcomes in patients with stent use. Walker et al11 also discovered poor patencies in 12 patients undergoing SIA without stents after occluded vascular grafts. Worse patency may be related to progression of atherosclerosis in that arterial segment or to neo-intimal hyperplasia near the anastomotic site. This study also showed a trend toward worsening patency in patients with critical limb ischemia compared to claudication. Other papers on SIA have also shown similar trends.12, 13, 14, 15 There also appears to be a trend for worse patency with smaller stents. This finding, however, may be confounded by vessel lumen. Hence, smaller stents may just reflect placement of stents in smaller or more distal vessels.
Treiman et al16 studied 29 patients who received routine stents after SIA and found good short-term patency rates of 85% and 64% at 1- and 2-years, respectively. The long-term patency rates, however, declined precipitously to 18% and 9% at 3- and 4-years.16 No further subanalysis of factors affecting outcomes was performed secondary to small numbers. Several important points need to be discussed. First, long-term outcomes after SIA are unknown. We have shown a primary patency of 55% and 35% at 1 year and 3 years, respectively, using selective stenting.2 While our early primary patency rates are not equal to Treiman's at 1 year, our primary patency rates are superior at 3 years. Therefore, stents may improve short-term patency due to prevention of recoil, but sacrifice long-term patency by inducing neo-intimal hyperplasia.
Second, it has been our view that stents may hinder any secondary interventions. Our study, however, shows that re-interventions can be undertaken with similar success rates as those limbs without stents. Third, one putative benefit of SIA is the maintenance of and possible opening up of collaterals.17 The fate of these collaterals could be jeopardized by placing stents. In our study, there was a trend for SIA patients receiving stents (35%) to require more open bypass surgery than those without stents (17%) at 2 years (P = .06). This trend could reflect our bias that endovascular reinterventions may be more difficult after stent failure. Hence, patients with failed stents may undergo open bypass surgery without an endovascular salvage attempt. Finally, Treiman's study lined the whole subintimal channel with stents, whereas we selectively placed stents in segments as needed. Hence, comparisons of SIA with stent cannot be made between the studies. The benefits and drawbacks of either approach are uncertain and remain an important question to address.
There are several limitations of our study. First, its retrospective design contributed to incomplete data collection for some preprocedural and postprocedural variables. This could result in underestimation of patency based upon clinical follow-up. Second, our study is the largest series of SIA undergoing selective stenting, but some variables had small numbers after partitioning. Therefore, any subanalysis performed could underestimate the effect a variable imparts on patency. Third, the incomplete and short-term follow-up limits the applicability to long-term outcomes. Fourth, this study was not designed to compare routine placement of stents vs placement of selective or no stents. Our practice utilizes selective placement of stents for a suboptimal result after SIA. While there are defined criteria for suboptimal results, completion angiogram interpretation after SIA remains quite subjective; hence, leading to variable stent use by different surgeons. Fifth, a wide variety and lengths of stents were utilized during the study. The variability of performance between the different types and lengths of stents may have confounded the outcomes in the stent group. In addition, the difference in stent lengths reflects both differences in the patient's disease processes and our bias to stenting only suboptimal segments. Finally, there is a possibility of a type 2 error, especially in subgroup analysis. This study, however, presents the largest reported series on selective stenting after SIA; hence, the possibility of a type 2 error is small.
Our study shows that selective stent placement can salvage poor outcomes after SIA. This study neither affirms nor refutes routine use of stents after SIA. A randomized study comparing routine vs selective stent placement is required to answer this important question.
Conclusions
Selective stents placed for suboptimal results after subintimal angioplasty produces similar patency rates to primary SIA without stenting. Patients with prior lower extremity bypass who receive a stent after SIA will have worse patency than those without prior bypass. Use of a stent diameter ≤6 mm and indication of critical limb ischemia will likely produce worse patency. Also, it appears that other specific stent variables (location, number, length, and overlap) do not alter patency. Finally, selective stent use after SIA provides excellent limb salvage.
Author contributions
References
- . Femoro-popliteal and crural artery recanalization using subintimal angioplasty. Semin Vasc Surg. 1995;8:253–264
- Subintimal angioplasty for the treatment of claudication and critical limb ischemia: 3-year results. J Vasc Surg. 2007;46:959–964
- Superficial femoral artery recanalization with self-expanding nitinol stents: long-term follow-up results. Eur J Vasc Endovasc Surg. 2007;34:702–708
- Balloon angioplasty versus implantation of nitinol stents in the superficial femoral artery. N Eng J Med. 2006;354:1879–1888
- Sustained benefit at 2 years of primary femoropopliteal stenting compared with balloon angioplasty with optional stenting. Circulation. 2007;115:2745–2749
- Percutaneous angioplasty and stenting of the superficial femoral artery. J Vasc Surg. 2005;41:269–278
- . Subintimal angioplasty in the management of critical lower-extremity ischemia: value in limb salvage. Perspect Vasc Surg Endovasc Ther. 2005;17:11–20
- Subintimal angioplasty of femoropopliteal artery occlusions: the long-term results. Eur J Vasc Surg. 1994;8:148–155
- Does subintimal angioplasty have a role in the treatment of severe lower extremity ischemia?. J Vasc Surg. 2003;37:386–391
- 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;116:285–292
- . Subintimal angioplasty of native vessels in the management of occluded vascular grafts. Eur J Vasc Endovasc Surg. 2001;22:41–43
- . Subintimal angioplasty as a treatment of femoropopliteal artery occlusions. Eur J Vasc Endovasc Surg. 2003;25:578–582
- . Subintimal angioplasty for superficial femoral artery occlusion: poor patency in critical ischaemia. Ann R Coll Surg Engl. 2005;87:361–365
- Factors affecting patency of subintimal infrainguinal angioplasty in patients with critical lower limb ischemia. Eur J Vasc Endovasc Surg. 2006;32:668–674
- . Subintimal angioplasty of long superficial femoral artery occlusions. J Vasc Interv Radiol. 2003;14:997–1010
- . Results of percutaneous subintimal angioplasty using routine stenting. J Vasc Surg. 2006;43:513–519
- Fate of collateral vessels following subintimal angioplasty. J Endovasc Ther. 2004;11:269–273
Competition of interest: none.
PII: S0741-5214(08)00931-2
doi:10.1016/j.jvs.2008.05.080
© 2008 The Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved.
