Role of Superficial Femoral Artery Stents in the Management of Arterial Occlusive Disease: Review of Current Evidence

Abstract

The objectives of this study were to compare the 1-year patency of superficial femoral artery (SFA) stent placement with percutaneous transluminal angioplasty (PTA) alone and to attempt to define the role of stents in arterial occlusive disease. Literature searches of the Embase, Medline, and Cochrane databases identified relevant articles, which were split into two subgroups: those containing case-controlled matches for stenting and PTA and those considering only stent placement. The review conformed to the QUORUM statement. One-year patency rates were 219 of 383 (57%) in the stented group and 319 of 607 (53%) in the PTA group for matched cases (odds ratio 1.206 [95% CI 0.932–1.559; p = .115]). The patency of stents deployed secondarily was 554 of 909 (61%). The combined 1-year patency of primary and secondary SFA stents for matched and unmatched case series was 816 of 1,282 (64%). One-year patency rates following stent placement or PTA alone in well-matched patient groups demonstrated no significant difference. This would suggest that the routine use of primary SFA stenting should be undertaken only in selected cases and should mainly be used in “bailout” situations or for complex lesions where surgery is precluded. However, the studies used displayed a high degree of heterogeneity, and some used stent technology that is now considered obsolete. Ongoing randomized controlled trials will provide a more definitive answer to this important problem.

Keywords

infrainguinal stent peripheral vascular disease superficial femoral artery angioplasty superficial femoral artery stent

Peripheral artery disease (PAD) describes stenotic, occlusive, or aneurysmal lesions of the aorta and its branch arteries, excluding the coronary arteries.¹ This review focuses on stenotic and occlusive disease of the peripheral arteries of the lower extremities, specifically the superficial femoral artery (SFA). Chronic lesions within the SFA tend to be atherosclerotic in origin, with a second significant subset being secondary to thromboembolism.¹ Consequently, the major risk factors for PAD are also those for atherosclerosis, namely, smoking, diabetes mellitus, hypertension, hyperhomocysteinemia, and dyslipidemia. These risk factors, and diabetes in particular, are becoming more prevalent within modern Western society, affecting approximately 27 million people in North America and Europe.² Around 30% of Americans over the age of 70 years are currently affected.³

Increasing numbers of patients present to vascular surgeons with intermittent claudication. This is characterized by lower limb pain brought on by a predictable level of exertion, relieved by rest and caused most often by an inability of the vascular tree to supply the increased metabolic demand of active muscle. Although claudication itself is generally a chronic stable symptom, 1 to 2% of patients will progress to critical limb ischemia within 5 years. A quarter of these patients will undergo amputation, and a further quarter will die from a cardiovascular cause.¹ The economic impact of these events is significant and increasing. For these reasons, the optimal management of PAD requires definition.

It was initially anticipated that minimally invasive endovascular interventions in the femoropopliteal region would result in reduced hospital stay, lower anesthetic risk, and fewer postprocedural complications for patients undergoing management of severe PAD. In practice, robust evidence of the superiority of these techniques has not been forthcoming, and in particular, the precise role of SFA stenting has yet to be defined. The lack of well-conducted randomized controlled trials (RCTs) comparing stent placement to the other established managements for PAD is a major problem. The aim of this review was to collect all current published evidence regarding the role of SFA percutaneous transluminal angioplasty (PTA) and SFA stent placement and to draw some conclusions regarding the optimal use of each technique.

Methods

Literature Searches

A literature search of Embase, Medline, and Cochrane databases was carried out for articles detailing the outcome of stenting and/or angioplasty for SFA stenoses or occlusions. The search terms used were “superficial femoral artery AND stent,” “superficial femoral artery AND angioplasty,” “superficial femoral artery AND stenosis,” and “superficial femoral artery AND occlusion,” and a sensitive but less specific search of the key words “peripheral AND stent” was conducted.

Articles were limited to those published in the English language. Scrutiny of references allowed the identification of any articles not picked up by the searches. The selected articles were reviewed in more detail, and exclusion criteria were applied. Articles excluded were those that considered the use of adjunctive procedures (laser, ultraviolet, and cryotherapy) and those that did not address the subject of the review.

Validity Assessment

Articles meeting the inclusion criteria were considered in terms of the quality of the study (S.P. and P.J.E.H.). These criteria were defined through patient numbers involved, the strength of adherence to the original study protocol, the numbers of patients crossing over between study arms, and their findings. Articles were not excluded if the numbers were very small or if a high proportion of patients crossed over between groups, but this was noted. The degree of adjustment for patient comorbid risk factors within individual studies was recorded only as either present or absent. The primary end point considered was primary patency at 1 year postprocedure.

Data Abstraction

Data abstraction was carried out by two authors (S.P. and P.J.E.H.). In addition to the 1-year patency, for each article, data were collected regarding the demographics of the populations studied, details of the lesions treated (stenotic or occlusive), the type of stent used where employed, and details of any adjunctive thrombolytic therapies. Articles providing details of case-control matched sets of stented and nonstented occlusions were considered sufficiently robust to calculate the odds ratio (OR) with regard to outcome in addition to subgroup analyses. No adjustment was made in this review for TASC classification beyond that reported in the original articles. One-year patency was defined as a < 50% reduction in postoperative luminal diameter. Freedom from binary restenosis was also used to measure patency. This was defined as a reduction in luminal diameter of ≥ 50% postoperative diameter on angiography.

Study Characteristics

Having performed the literature searches, we found that the data were too heterogeneous to perform a meaningful meta-analysis, although this would have technically been possible. This decision was made in conjunction with a medical statistician. A major factor underlying this was the paucity of high-quality RCTs of primary stent placement versus primary PTA alone. The articles remaining after the exclusion criteria were applied were systematically reviewed using a process that conformed to the QUOROM guidelines (a consensus statement of standards when undertaking systematic reviews) through the review structure, despite the lack of RCTs.

Results

From the literature searches, 26 potentially relevant articles were retrieved (Figure 1). Of these, seven were excluded from review; three were reviews considering outcomes from all endovascular interventions at a variety of different anatomic sites (iliac arteries, infrapopliteal arteries, and SFAs) and so were unsuitable for inclusion by our criteria^4–6; one was a German article with only an English extended abstract, which was not automatically excluded as being non-English during the literature search⁷; one involved the use of adjuvant laser therapy and angioplasty⁸; one RCT compared angioplasty to vein bypass surgery⁹; and the remaining article considered the rate of stent fractures as the primary end point.¹⁰

Figure 1.

QUOROM flow chart.

After these exclusions, 19 articles remained, of which 9 provided data for both PTA alone and angioplasty with SFA stent placement, and in 7 of these, patients were randomized to stent or nonstent groups.^11–17 Demographic data on these groups were not provided consistently enough to comprehensively compare the characteristics of each group, but they were broadly similar (Table 1). In the remaining two articles, stents were employed selectively if the angioplasty results were considered to be radiologically suboptimal.^18,19 A further nine articles provided data on stent placement only. The data from these articles were considered with those from the stenting arm of the previously selected studies to give an indication of overall stent outcome and a subgroup analysis. Some of the studies included stents that are no longer employed in the femoropopoliteal territory (ie, Palmaz, Strecker, wall, polytetrafluoroethylene, and steel stents). We decided to report these studies in order to produce a comprehensive review of subject, but they were separated to facilitate subgroup analysis.

Table 1.

Demographics Derived from Series Reporting a Comparison of Outcomes in Patients with SFA Disease Treated with Stenting or PTA Alone

	Stented Patients	PTA Patients
Age (yr)	67.3 (65–71)	67.7 (63.5–72)
Sex (% male)	63 (59–67)	50 (24–75)
Hypertension (%)	66 (27–83)	68 (24–92)
Diabetes (%)	40 (10–66)	32 (14–50)

PTA = percutaneous transluminal angioplasty; SFA = superficial femoral artery.

The numbers in parentheses denote the range of means among the studies.

The results of articles providing matched case-controlled series of stented and nonstented cases are given (Table 2A and Table 2B, Table 3). These articles considered 1,452 interventions with 564 SFA stents being placed and 888 limbs being treated with PTA alone. Seven of the nine articles considered reported the primary end point of 1-year patency.^{11–13,15–17,19} One article only provided a combined patency rate for the stented and nonstented arms and so did not provide data suitable for further consideration within this review.¹⁸

Table 2A

Results from Studies Comparing Matched Patients Only for SFA PTA versus SFA Stent Implantation

Study	Area	Stent Type	Antithrombotic Therapy	Primary or Secondary Stenting	Number	Stented	Nonstented	Stented 1 yr Patency	Nonstented 1 yr Patency
Krankenberg et al (2007)¹¹	Hamburg University Cardiovascular Centre	Bard Luminexx 3 Vascular Stent (self-expanding nitinol)	300 mg clopidogrel in stent arm postprocedure, all limbs 100 mg aspirin indefinitely	Primary	244	123	121	69/101 (freedom from duplex restenosis)	62/101 (freedom from duplex restenosis)
Schillinger et al (2006)¹²		Self-expanding nitinol stent	Clopidogrel for 3/12 and aspirin indefinitely	Primary	104	51	53	32/51 (63%) (absence of restenosis on duplex scanning)	20/53 (37%) (absence of restenosis on duplex scanning)
Surowiec et al (2005)¹⁸	University of Rochester (1986–2004)	Self-expanding nitinol stent	All on antiplatelet agents and at the discretion of the surgeon, abciximab 3.75 μg/kg/h for 12 h	Secondary	380	141	239	75% (285/380) only recorded as a combined patency for PTA and stent, so data not included for analysis in this review
Schmeider et al (2008)¹⁹	Division of Vascular Surgery, Eastern Virginia Medical School	Self-expanding nitinol stent (111) or balloon-expandable stents	Clopidogrel for 1/12, then aspirin indefinitely	Primary	382	84	298	42/84 (50%) (mixture of duplex scanning and angiography)	134/298 (45%) (mixture of duplex scanning and angiography)
Total 1 yr patency								143/236 (61%)*	216/452 (48%)*

PTA = percutaneous transluminal angioplasty; SFA = superficial femoral artery.

Primary restenosis rates are calculated by life table analysis unless otherwise stated.

This table incorporates studies that used contemporary devices.

*Excludes Surowiec et al¹⁸ as their study did not report separate patency rates.

Table 2B

Results from Studies Comparing Matched Patients Only for SFA PTA versus SFA Stent Implantation

Study	Area	Stent Type	Antithrombotic Therapy	Primary or Secondary Stenting	Number	Stented	Nonstented	Stented 1 yr Patency	Nonstented 1 yr Patency
Zdanowski et al (1999)¹³	Lund University, Lund, Sweden	Strecker stent	160 mg aspirin postoperatively indefinitely	Primary	32	15	17	6/12 (freedom from binary restenosis)	6/8 (freedom from binary restenosis)
Saxon et al (2003)¹⁴	San Diego Vascular Institute	Hemobahn nitinol and ePTFE stents	Clopidogrel for 1/12 postoperatively and aspirin indefinitely	Secondary	28	15	13	13/15 (2 yr patency data from duplex scanning)	3/12 (2 yr patency data from duplex scanning)
Becquemin et al (2003)¹⁵	Hospital Henri Mondor, Paris	Palmaz	Low–molecular-weight heparin for 24 h, then aspirin (300 mg/d) or ticlopidine indefinitely	Primary	140	65	75	44/65 (68%) from 1 yr angiography	49/75 (65%) from 1 yr angiography
Cejna et al (2001)¹⁶	University Clinicsof Vienna	Palmaz	Aspirin 100 mg od for life and 750 U heparin for 2 d	Primary	91	46	45	29/46 (63%) angiographic 1 yr patency	28/45 (63%) angiographic 1 yr patency
Vroegindeweij et al (1997)¹⁷		Palmaz	Heparin for 48 h, aspirin 80 mg/d indefinitely	Primary	51	24	27	15/24 (62%) color flow duplex scanning	20/27 (74%) color flow duplex scanning
Total 1 yr patency								94/147 (64%)*	103/155(66%)*

ePTFE = expanded polytetrafluoroethylene; PTA = percutaneous transluminal angioplasty; SFA = superficial femoral artery.

Primary restenosis rates are calculated by life table analysis unless otherwise stated.

This table shows studies that used devices that are not considered useful in the SFA in contemporary practice.

*Excludes Saxon et al¹⁴ as follow-up was at 2 years.

The patency of SFA stents at 1 year ranged from 42 of 84 (50%)¹⁹ to 69 of 101 (68%),¹¹ whereas that for the PTA alone ranged from 20 of 53 (37%)¹² to 6 of 8 (75%).¹³ Across all case-controlled studies involving primary stenting, 1-year patency rates were 219 of 383 (57%) in the stent placement subgroup and 319 of 607 (53%) in the PTA subgroup, giving an OR of 1.206 (95% CI 0.932–1.559; p = .115) in favor of stent placement, but this finding was not statistically significant. When compared to the best available evidence of nonmatched PTA series,¹ the results remained nonsignificant (OR 0.927 [95% CI 0.731–1.176]; p = .534).

One study that employed 2-year patency as a primary end point suggested that patency was greater in the stented group, but the numbers were small-13 of 15 (87%) in the stent arm and 3 of 12 (25%) in the PTA arm-and nonsignificant (OR ± 95% CI 3.467 [0.84–13.87]; p = .087).¹⁴ These values have not been pooled as 1-year patency was the primary end point. A summary of the pooled data is given in Table 3.

Table 3.

Combined Outcomes of All Studies Reviewed

Summary of Combined Studies Reporting PTA versus Stenting
Total number: 1,452
Stenting vs PTA: 564/888
Patency of stents at 1 yr: 219/383 (57%; range 50–68%)
Patency of PTA at 1 yr: 319/607 (53%; range 37–75%)
Odds ratio of stenting vs PTA: 1.2 (95% CI 0.93–1.559; p = .115): nonsignificant difference
Summary of All Series Reporting Stenting of the Superficial Femoral Artery
Total number: 1,488 (1 yr patency available for 1,282)
Primary vs secondary: 455/1,033
Combined 1 yr patency: 816/1,282 (64%; range 22–90%)
Primary stenting subgroup 1 yr patency: 262/373 (70%; range 50–90%)
Secondary stenting subgroup 1 yr patency: 554/909 (61%; range 22–87%)

PTA = percutaneous transluminal angioplasty.

Comparing stents that are no longer used in the lower limb with contemporary devices showed a patency rate of 94 of 147 (64%) versus 143 of 236 (61%) at 1 year, respectively, with an OR of 1.15 (95% CI 0.753–1.7662; p > .05) in favor of older devices; however, this was not statistically significant. The rate of 1-year patency of the PTA groups was 103 of 155 (66%) and 216 of 452 (48%), respectively, giving a confidence interval of 2.16 (95% CI 1.478–3.168; p < .05) in favor of the PTA performed in the newer group. This gave an OR for stenting versus PTA of 1.68 (95% CI 1.220–2.313; p < .05) in the contemporary device group in favor of stenting and 0.895 (95% CI 0.558–1.438; p > .05) in favor of PTA (but not statistically significant) in the historical group. This suggested that the newer stents may perform slightly better in comparison with PTA in the contemporary setting when compared to older studies.

When considering studies of matched and unmatched patient groups, 1,488 SFA stents were placed. Once again, the demographic data available were limited (Table 4), but data on the anatomic location of lesions, the scoring systems used for disease severity, and the use of antiplatelet agents from each article are given (Table 5A and Table 5B).

Table 4.

Demographics Derived from Series Reporting Outcomes of Patients with SFA Disease Treated with Stenting Alone

	Mean Values
Age (yr)	68.3 (62–75)
Sex (% male)	64 (56–71)
Hypertension (%)	74 (62–77)
Diabetes (%)	44 (31–62)

SFA = superficial femoral artery.

The numbers in parentheses denote the range of means among the studies.

Table 5A

Primary Patency Rates for Stenting, Either Primary or Secondary

Study	Limbs	Data Source	Primary or Secondary Stenting	Lesion Types	Stent Type	Antiplatelet Therapy	Primary 1 yr Patency
Krankenberg et al (2007)¹¹	123	Hamburg University Cardiovascular Centre	Primary	Single SFA lesion or chronic limb ischemia	Bard Luminexx 3 self-expanding nitinol stent	300 mg clopidogrel in stent arm postprocedure, all limbs 100 mg aspirin indefinitely	69/101, (68%) 1 yr freedom from duplex binary restenosis
Schillinger et al (2006)¹²	51		Primary	Severe claudication or chronic limb ischemia owing to stenosis or occlusion of the SFA	Self-expanding nitinol stent	Clopidogrel for 3/12 and aspirin 75 mg indefinitely	Absence of restenosis on duplex scanning 32/51 (63%)
Surowiec et al (2005)¹⁸	141	University of Rochester	Secondary	TASC I grade A, 48%; B, 18%; C, 22%; D, 12%	Nitinol	All on antiplatelet agents and, at the discretion of the surgeon, abciximab 3.75 μg/kg/hr for 12 h	106 (75%) (extrapolated) duplex scanning
Schmeider et al (2008)¹⁹	84	Division of Vascular Surgery, Eastern Virginia Medical School	Secondary	SFA or popliteal, “disabling claudication or critical limb ischemia”	Nitinol	Clopidogrel for 1/12 then aspirin indeffinitely	42 (50%) mixture of duplex scanning and angiography
Duda et al (2005)²⁰	29	The Jewish Hospital, Berlin; from 8 recruiting centers across Europe	Primary	Symptomatic peripheral artery disease; 66.7% SFA occlusions	Sirolimus-eluting nitinol stent; bare nitinol stent	Ticlopidine or clopidogrel for 1/12 and then aspirin 75 mg for at least 1 yr	Freedom from binary restenosis at 6/12: 29/29 (100%) duplex scanning
	28						Freedom from binary restenosis at 6/12: 26/28 (92.3%) duplex scanning
Ferreira et al (2007)²¹	74	Rio de Janeiro, Brazil	Primary	Rutherford category 2 or above, not responsive to 6/12 medical therapy	Self-expanding nitinol stent	Clopidogrel 75 mg qd for 6/12 followed by 300 mg aspirin 4 times daily indefinitely	67 (90%) duplex scanning
Abahji et al (2006)²²	71	Ludwig-Maximilians University, Munich, Germany	Secondary	SFA occlusion or stenosis	Operator dependent	Aspirin and clopidogrel for 1/12 then aspirin indefinitely	38 (59% ) (extrapolated) duplex scanning
Vogel et al (2003)²⁴	41	University of Medicine and Dentistry of New Jersey	Secondary	Symptomatic femoropopliteal disease	Nitinol stents	75 mg clopidogrel postoperatively and indefinitely	34 (84%) duplex scanning
Cheng et al (2001)²⁶	60	University of Hong Kong Medical Centre	Secondary	52% occlusion, 48% stenosis, 43% critical limb ischemia		Ticlopidine 250 mg twice daily	38 (62%) duplex scanning
Ihnat et al (2008)²⁷	109	University of Arizona Health Science Center and Southern Arizona Veterans Affairs Health Care System	Secondary	Chronic femoropopliteal artery occlusive disease	Self-expanding nitinol stents	Clopidogrel for 1/12 (75 mg od) followed by aspirin indefinitely	57 (52%) at 3 yr follow-up; duplex scanning
Sabeti et al (2004)²⁸	104	University of Vienna Medical School	Secondary	Severe intermittent claudication or critical limb ischemia	Nitinol	75 mg clopidogrel followed by aspirin for at least 1/12	78 (75%) duplex scanning and angiography
Total 1 yr patency							504/650 (78%)*

SFA = superficial femoral artery.

Calculated by life table analysis unless otherwise stated.

Also included are the stent types, lesion types/sizes, long-term antiplatelet therapy, and data source. Studies using contemporary devices are shown.

*Excluding Ihnat et al²⁷ as 3-year patency was used and Duda et al²⁰ as 6-month patency was used.

Table 5B

Primary Patency Rates for Stenting, Either Primary or Secondary

Study	Limbs	Data Source	Primary or Secondary Stenting	Lesion Type	Stent Type	Antiplatelet Therapy	Primary 1 yr Patency
Zdanowski et al (1999)¹³	15	Lund University, Lund, Sweden	Primary	Femoropopliteal occlusions	Strecker stent	160 mg aspirin postoperatively indefinitely	6/12 at 1 yr (freedom from angiographic binary restenosis) (50%)
Saxon et al (2003)¹⁴	15	San Diego Vascular Institute	Secondary	Claudication or ischemia owing to SFA or above-knee popliteal lesion	Hemobahn nitinol and ePTFE stents	Clopidogrel for 1/12 postoperatively and aspirin indefinitely	2 yr patency on duplex scanning 13/15 (87%)
Abahji et al (2006)²²	71	Ludwig-Maximilians University, Munich, Germany	Secondary	SFA occlusion or stenosis	Operator dependent	Aspirin and clopidogrel for 1/12 then aspirin indefinitely	42 (59%) (extrapolated) duplex scanning
Gray et al (1997)²³	58	Cleveland Clinic Foundation	Secondary	SFA lesions from 6–35 cm in length	Operator dependent: Wallstent (75%), Palmaz (21%), or both (4%)	Warfarin while stent was patent	13 limbs (22%) duplex scanning
Conroy et al (2000)²⁵	61	Administration Medical Center, Long Beach, CA	Secondary	Mean occlusion length was 13.5 cm	Mixed Wallstents	Warfarin for 1/12 then aspirin 75 mg for 1/12	29 limbs (47%) duplex scanning
Cheng et al (2001)²⁶	60	University of Hong Kong Medical Centre	Secondary	52% occlusion, 48% stenosis, 43% for critical limb ischemia		Ticlopidine 250 mg twice daily	38 (62%) duplex scanning
Sabeti et al (2004)²⁸	123	University of Vienna Medical School	Secondary	Severe intermittent claudication or critical limb ischemia	Stainless steel	75 mg clopidogrel followed by aspirin for at least 1/12	66 (54%) duplex scanning and angiography
Lenti et al (2007)²⁹	166	University of Perugia, Ospedale Civiledi Imperia	Secondary	115 limbs for intermittent or long (> 3 cm) occlusion, 51 limbs for stenosis; 51 of these were in the SFA, 115 in the proximal popliteal artery	Spiral, covered polytetrafluoroethylene stent	Clopidogrel or ticlopidine for 1/12 followed by aspirin indefinitely	106 (64%) duplex scanning
Total	569						294/539 (55%)*

ePTFE = expanded polytetrafluoroethylene; SFA = superficial femoral artery.

Calculated by life table analysis unless otherwise stated.

This table shows studies that used devices that are not considered useful in the SFA in contemporary practice. Also included are the stent types, lesion types/sizes, long-term antiplatelet therapy, and data source.

*Excludes Zdanowski et al¹³ as 6-month patency was reported and Saxon et al¹⁴ as 2-year patency was reported.

Of these procedures, 455 were carried out as primary interventions^{11–13,15–17,20,21} and 1,033 were placed secondarily to radiologically unsuccessful primary PTA.^{14–19,22–29} Of the 1,488 limbs stented, 1-year patency data were recorded for 1,282 limbs. The combined 1-year patency after either primary or secondary stenting was 816 of 1,282 (64%). The end point of 1-year patency was recorded in all but one of the primary stenting studies,^11–13,21 and the remaining article recorded patency at 6 months.²⁰ The 1-year patency in primary stent subgroups, where reported, ranged from 6 of 12 (50%)¹³ to 67 of 74 (90%),²¹ with a pooled 1-year patency of 262 of 373 (70%).

The 1-year patency rate recorded in 9 of the 11 articles involving stent deployment secondary to a radiologically unsuccessful PTA ranged from 13 of 58 (22%)²³ to 13 of 15 (87%),¹⁴ with a combined 1-year patency of 554 of 909 (61%). The other two articles considered 2- and 3-year patency,^14,27 with patency rates of 13 of 15 (87%) and 57 of 109 (52%), respectively.

Within the articles considered in this review, two had two treatment arms, each for a different stent type. One of these articles used the stents as a primary intervention and the other as a secondary intervention.^20,28 The first of these articles demonstrated that at 6 months post–stent placement, bare nitinol stents had a patency of 26 of 28 (92%), whereas sirolimus-eluting stents had a patency of 29 of 29 (100%).²⁰ The other article compared 1-year patency of nitinol and stainless steel stents; nitinol stents had a 1-year patency of 78 of 104 (75%), and the steel stents had a 1-year patency of 66 of 123 (54%).²⁸ A subgroup analysis separating studies that used older stents that are generally considered to be obsolete in the SFA and more contemporary devices showed the patency rate at 1 year in the more recent studies to be 504 of 650 (78%) versus 294 of 539 (55%) in the “obsolete” group. This gives an OR of 2.88 (95% CI 2.240–3.70, p < .05) in favor of the newer stents.

Discussion

This review demonstrated that the 1-year patency for SFA stent placement when compared to PTA alone, in well-matched patient groups, did not support the primary employment of SFA stents, with these cases being no more likely to be patent at 1 year than PTA alone. When all stenting procedures in all of the studied articles were combined (primary and secondary, matched and unmatched), a patency rate of 64% was shown at 1 year, which was comparable to established 1-year patency of PTA alone of 59%.¹

The suggestion was that there was no benefit of primary SFA stenting against PTA alone on clinical grounds. However, the inclusion of series in this review that used Striker and Palmaz stents that are now perhaps historical might be considered a limitation of this review. Furthermore, these data were heterogeneous, with both primary and secondary stenting being considered in addition to a wide variety of stent types. It is possible that the results of this review do not report modern practice, where secondary stenting is employed in preference to primary stenting. In addition, the use of self-expanding nitinol stents would be the standard of care in most departments. Stent graft use in the femoropopliteal segment remains controversial, although in some units, it is gaining popularity, and these might be expected to behave differently to a self-expanding stent. Their use and outcomes should be reported separately when registry or trial data are sufficient and complete. Subgroup analysis of the obsolete stent group in comparison with the contemporary device group showed no difference between the two groups. It did show that the results for PTA in the newer studies were apparently poorer. This meant that the new stent group performed relatively more successfully in comparison with PTA than the older group. This could be because the threshold for using PTA on more complex lesions is lower or that PTA has become less successfully employed over time.

One-year patency in the articles considered was defined as a < 50% reduction in postoperative luminal diameter; however, the method of measurement varied considerably between articles. Some articles recorded patency as freedom from binary restenosis, which may be defined as reduction in the luminal diameter of ≥ 50% postoperative diameter on angiography¹³ through either direct measurement or simple observation. This method may be a significant source of observation bias. In other articles, the patency of vessels was measured with duplex ultrasonography, with the same quantitative parameter of 50% luminal diameter reduction.^{11,12,14,18,20–27,29} Two of the articles assessed patency first with duplex ultrasonography followed by angiography if a reduction in diameter of over 50% was identified on duplex scans.^19,28 Although the results from these different procedures may vary, they were considered comparable within this review, but a need for consensus in recording is highlighted.

The variation in morphology, location, etiology, and severity of lesions treated in the various studies made interpreting the results more difficult. Severity of disease was difficult to consistently quantify as each article included a variety of staging criteria (TASC score, Rutherford scale, or other measures). Each study also had different inclusion and exclusion criteria. Differences in the indications for the deployment of stents in different centers have become apparent when designing clinical trial protocols. In particular, the nonuniform length of lesion treated and the employment of stents for stenotic lesions over occlusions might bias the results from some studies. Given the paucity of data, this heterogeneity was impossible to adjust for, so all PAD patients suffering from the spectrum of symptomatic PAD were considered as a single group.

It was noted that studies reporting unmatched patient populations, or those reporting outside the framework of an RCT, were more likely to report the beneficial results of SFA stent use over PTA alone. This applies in particular to selected single-center series, which may present excellent results that might not be reproducible in the wider community. It could be that this reflects a degree of publication bias, which must be considered when interpreting data such as these that incorporate new technologies.

The use of antiplatelet agents after stent placement is important as platelet-mediated thrombosis is a major cause of potentially preventable stent occlusion. Many groups employed postoperative aspirin, with the dose varying between 75 and 300 mg.²¹ Three studies did not use aspirin postoperatively but selectively employed warfarin,²³ clopidogrel,²⁴ and ticlopidine.²⁶ The patency rates at 1 year post–stent placement in these studies were 22%, 84%, and 62%, respectively. This may suggest that warfarin had a detrimental effect on stent patency, although this observation was based on a small sample size and could have been influenced by unidentified confounding factors. Overall, the consensus was that aspirin should be continued indefinitely after stent placement with a shorter course of another agent after stent placement (usually a 1-month course of clopidogrel^11,12,21).

The technical difficulty of salvage procedures following the failure of either stenting or PTA may be determined in part by which procedure is employed primarily. Repeat PTA tends to be uncomplicated and is widely performed. Reintervention after stenting may be a more challenging undertaking and carry a higher risk to patients of subsequent bypass surgery or amputation. The new generation of stents developed specifically for use in the SFA may reduce these risks, but this will depend on definitive RCT evidence of improved short- and long-term patency. Concern also remains over cost-effectiveness.³⁰

Primary stenting may be considered different from secondary stenting, where, by definition, the lesion is often more challenging as PTA has already achieved suboptimal technical results. This implies a role for stenting in a bailout situation. Additionally, it is accepted that some lesions (long, calcific, with thrombus, prone to dissection) are better treated with primary stenting when the endovascular solution is the only one available; however, when fit for surgery, these patients might be better served with a bypass procedure.

The past use of SFA stents when early results were relatively equivocal may reflect a trend toward introducing new technology before the development of appropriate reporting standards. In an effort to improve the quality, safety, and cost-effectiveness of health care, novel interventions should be introduced through well-designed trials within a framework of quality assurance and accurate reporting.

Conclusion

These data suggest no statistically significant difference in 1-year primary patency outcome between the patients undergoing PTA and those being treated with stents. This would suggest that the routine use of stents in the femoropopliteal region cannot be supported as a primary procedure. However, there was a significant degree of heterogeneity among the studies, reflecting the poor quality of data available. Some of the earlier series are now effectively historical owing to the use of obsolete stents. As such, this conclusion may not reflect accurately the state-of-the-art as practiced in advanced centers and should be interpreted with caution. There are also not enough data to suggest which treatment modality particular lesions would be most amenable to. The next generation of self-expanding stents will no doubt improve on the last, but potentially with increased cost; therefore, proving cost-effectiveness will be vital. The results of RCTs such as Zilver PTX and Durability II are awaited, and it is hoped that these will help further clarify the role of infrainguinal stents.

Footnotes

Acknowledgment

Financial disclosure of authors and reviewers: None reported.

References

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.