A comparison of covered vs bare expandable stents for the

CLINICAL RESEARCH STUDIES

A comparison of covered vs bare expandable stents

for the treatment of aortoiliac occlusive disease

Bibombe P. Mwipatayi, MMed (Surg), FCS (SA), FRACS,

a,b

Shannon Thomas, MBBS (Hons),

Jackie Wong, MPH,

Suzanna E. L. Temple, PhD, MBA,

a,c

Vikram Vijayan, MRCS, FRCS,

Mark Jackson, MD, FRACS,

and Sally A. Burrows, BMath Grad Dip Med Stat,

on behalf of the

Covered Versus Balloon Expandable Stent Trial (COBEST) Co-investigators,* Perth, Western Australia

and Gold Coast, Queensland, Australia

Objective: This trial was conducted to determine if covered stents offer a patency advantage over bare-metal stents in the

treatment of aortoiliac arterial occlusive disease.

Methods: The Covered Versus Balloon Expandable Stent Trial (COBEST), a prospective, multicenter, randomized

controlled trial, was performed involving 168 iliac arteries in 125 patients with severe aortoiliac occlusive disease who

were randomly assigned to receive a covered balloon-expandable stent or bare-metal stent. Patient demographic data,

clinical signs and symptoms, TransAtlantic Inter-Society Consensus (TASC) classiﬁcation, and preprocedure and

postprocedure ankle-brachial index measurements were recorded. The primary end points included freedom from binary

restenosis and stent occlusion of the treated area, as determined by ultrasound imaging or quantitative visual angiogra-

phy, or both. Postprocedural follow-up was at 1, 6, 12, and 18 months.

Results: Aortoiliac lesions treated with a covered stent were signiﬁcantly more likely to remain free from binary restenosis

than those that were treated with a bare-metal stent (hazard ratio [HR], 0.35; 95% conﬁdence interval (CI), 0.15-0.82;

P ⴝ .02). Freedom from occlusion was also higher in lesions treated with covered stents than in those treated with a

bare-metal stent (HR, 0.28; 95% CI, 0.07-1.09); however, this did not reach statistical signiﬁcance (P ⴝ .07). Subgroup

analyses demonstrated a signiﬁcant difference in freedom from binary restenosis for covered stents in TASC C and D

lesions compared with a bare stent (HR, 0.136; 95% CI, 0.042-0.442). This difference was not demonstrated for TASC

B lesions (HR, 0.748; 95% CI, 0.235-2.386).

Conclusions: COBEST demonstrates covered and bare-metal stents produce similar and acceptable results for TASC B

lesions. However, covered stents perform better for TASC C and D lesions than bare stents in longer-term patency and

clinical outcome. ( J Vasc Surg 2011;54:1561-70.)

Aortoiliac arterial occlusive disease is frequently en-

countered in the management of lower limb vascular insuf-

ﬁciency. Surgical reconstruction of the aortoiliac segment

was initially performed using thromboendarterectomy

;

however, advances in graft technology have made bypass

procedures more commonly performed and durable. Distal

aortic bypass has a reported surgical mortality of 4%, with

major complications seen in up to 21% of patients.

2-7

From the Department of Vascular Surgery, Royal Perth Hospital, Perth,

Western Australia

; School of Surgery, Faculty of Medicine, Dentistry and

Health Sciences, University of Western Australia, Perth, Western Austra-

lia

; Lung Institute of Western Australia and School of Medicine and

Pharmacology, University of Western Australia, Perth, Western Australia

;

Department of Vascular Surgery, Gold Coast Hospital, Gold Coast,

Queensland

; and Department of Biostatistics, School of Medicine and

Pharmacology, Royal Perth Hospital and University of Western Australia,

Perth, Western Australia.

*COBEST co-investigators: Dr Marek Garbowski, Dr Glen Benveniste, Dr

Michael Denton, Dr John Anderson, Dr Steve Dubenec, Dr Michael

Neale, Dr Vikram Puttaswamy, Prof John Fletcher, and Prof Geoffrey

White.

Atrium Medical Corporation provided funding for the employment of a

research assistant and had no role in study design, data collection, data

interpretation, or writing of the report. The corresponding author had full

access to all the data in the study, and the authors had ﬁnal responsibility

for the decision to submit for publication.

Competition of interest: Dr Mwipatayi and Dr Jackson have received

research-funding assistance from Atrium Medical Corporation.

Presented at the VIVA meeting, Las Vegas, Nev, October 19-23, 2009; the

LINC meeting, Leipzig, Germany, January 25-28, 2010; C3 meeting,

Baltimore, Md, June 20-24, 2010; ANZSVS Vascular Meeting, Gold

Coast, State, Country, August 2-5, 2010; 10th China National Annual

Congress of Vascular Surgery, Nanjing, China, September 24-26, 2010;

Veith meeting, New York, NY, November 17-21, 2010; and CACVS

2011 Congress, Paris, France, January 27-29, 2011.

Additional material for this article may be found online at www.jvascsurg.org.

Correspondence: Prof B. Patrice Mwipatayi, University of Western Austra-

lia, School of Surgery, Royal Perth Hospital, Department of Vascular

Surgery. Level 2, MRF Bldg, Perth, Western Australia, Australia (e-mail:

[email protected]; [email protected]).

The editors and reviewers of this article have no relevant ﬁnancial relation-

ships to disclose per the JVS policy that requires reviewers to decline

review of any manuscript for which they may have a competition of

interest.

0741-5214/$36.00

doi:10.1016/j.jvs.2011.06.097

1561

brought to you by COREView metadata, citation and similar papers at core.ac.uk

provided by Elsevier - Publisher Connector

Although a laparoscopic approach may decrease this mor-

tality and morbidity, this technique has yet to become

common place.

Currently, the decision to perform an endovascular

procedure or open surgical bypass depends on various

factors, most importantly, the severity and anatomic distri-

bution of disease. The TransAtlantic Inter-Society Consen-

sus (TASC) I and II working group published a consensus

of recommendations for management of peripheral arterial

disease.

9,10

The recommendation for patients with focal or

short lesions (ie, TASC A and TASC B lesions) is endovas-

cular intervention with an evidence basis suggesting long-

term patency rates close to 80% at 2 years and 60% at 5

years.

11,12

Lower patency rates with TASC C and D lesions

led to the recommendation for an open surgical approach,

except in patients deemed to be high-risk.

The use of covered stents to treat aortoiliac occlusive

disease has been proposed as a method to reduce intimal

hyperplasia and improve patency rates.

13-15

Although the

results of numerous reports are promising, present evidence

is insufﬁcient to support their widespread application. Cur-

rently, covered stent use is largely reserved for the treat-

ment of aneurysms, ruptures, and arteriovenous ﬁstulas.

16-18

No published randomized controlled trials (RCTs) to date

have compared the efﬁcacy of covered stents vs bare-metal

stents (BMSs) for aortoiliac occlusive disease. We hypoth-

esized that via intimal exclusion, covered stents will provide

a patency advantage and improved clinical outcome in

advanced iliac occlusive disease compared with BMS.

METHODS

Study design. The Covered Versus Balloon Expand-

able Stent Trial (COBEST) was a prospective, multicenter,

RCT (registration number: ISRCTN89458845) compar-

ing the use of the Advanta V12 balloon-expandable cov-

ered stent (Atrium Medical Corp, Hudson, NH) with

commercially available bare-metal stents (BMS) in patients

with aortoiliac occlusive disease. The study was conducted

between January 2006 and December 2008 with 13 phy-

sicians (11 vascular surgeons and 2 interventional radiolo-

gists) at eight major centers across Australia. The study

design allowed an assessment of the safety and efﬁcacy of

covered stents vs BMSs in the treatment of hemodyna-

mically signiﬁcant aortoiliac occlusive disease. In particular,

COBEST aimed to determine if the theoretical advantage

of covered stents resulted in clinical improvements.

The institutional review board and ethics committees at

each hospital approved the study protocol, and written

informed consent was obtained from all patients before

enrollment. The criterion for study entry and exclusion are

listed in Table I.

Study end points. The primary study end point was

the rate of binary restenosis (deﬁned by ⱖ50% reduction in

lumen diameter) and freedom from stent occlusion at 18

months. These were determined by duplex ultrasound

(DUS) imaging of the aortoiliac segment, computed to-

mography angiography (CTA), or catheter biplane digital

subtraction angiography (DSA). Patients were assessed

clinically with an ankle-brachial index (ABI) and aortoiliac

arterial DUS at intervals of 1, 6, 12, and 18 months. If the

DUS scan was inconclusive, a CTA or DSA, or both, were

performed using a prespeciﬁed protocol to determine if the

primary end point had been achieved.

The secondary end points were determined anato-

mically, clinically, and hemodynamically. Anatomic end

points included stent patency, as assessed by the TASC B,

C, and D classiﬁcation, stent integrity, and target vessel

revascularization (TVR), deﬁned as any repeat percutane-

ous intervention or surgical bypass of any segment of the

target vessel. The target vessel was deﬁned as the entire

aortoiliac vessel proximal and distal to the target lesion.

TVR was typically driven by clinical evidence of symptoms

or positive stress-induced signiﬁcant symptoms (treadmill

exercise ABI).

The clinical end points were major amputation above

the ankle and major adverse events resulting in patient

hospitalization or prolongation of existing hospitalization,

signiﬁcant physical disability, or death. The hemodynamic

end point was a change in ABI between baseline and

measurements at 1, 6, 12, and 18 months.

Randomization. Eligible patients were randomized

to receive a covered stent or a BMS by an online computer-

ized randomization program (https://io.atriummed.com/

Cobest/default.aspx). Unstratiﬁed randomization was

used to randomize patients in a 1:1 ratio with a minimiza-

tion algorithm to allow balanced allocation of participants

across intervention groups. All randomized patients were

irrevocably included in the study, regardless of eligibility or

Table I. Inclusion and exclusion criteria for the Covered

Versus Balloon Expandable Stent study

Inclusion criteria

● Men and women aged ⱖ18 years

● Informed consent obtained

● Evidence of TASC B, C, or D lesions

● Hemodynamically signiﬁcant dissections and recurrent stenosis

after angioplasty

Exclusion criteria

● Life expectancy ⬍12 months (patients had to be followed up

for at least 18 months)

● Uncontrolled hypertension

● TASC A lesion

● Pregnant women or women of childbearing potential who

were not using an effective method of contraception

● Prior enrollment in this trial, or a patient who had had any

procedure performed at the aortoiliac level

● Extensive common femoral artery disease or multiple groin

procedures

● Contraindication to aspirin or clopidogrel usage

● Occluded superﬁcial and profunda femoral arteries

● Mental condition rendering the individual unable to

understand the nature, scope and possible consequences of the

study, or a language barrier preventing the individual from

providing informed consent

● Uncooperative attitude or potential for noncompliance with

the protocol requirements, making study participation

impractical

TASC, TransAtlantic Inter-Society Consensus.

JOURNAL OF VASCULAR SURGERY

December 2011

1562 Mwipatayi et al

if they received the allocated treatment, and were moni-

tored for 18 months. A subgroup of patients had bilateral

lesions and each side of the iliac artery disease (left or right)

was randomized in these patients rather than the patient

per se.

Stents. The investigational devices consisted of the

Advanta V12 covered stent and commercially available

BMSs. The Advanta V12 is encapsulated with expanded

polytetraﬂuroethylene (ePTFE), premounted on an Aner-

tia noncompliant balloon (Atrium Medical Corp), and

compatible with a 6F to 7F sheath. Stents were sized

perioperatively, with covered and BMS oversized by 1 mm

beyond the calculated diameter of the iliac artery treated.

The BMS used was chosen at the discretion of the

interventionist, but Australian Therapeutic Goods Ad-

ministration approval was required. The bare-metal bal-

loon-expandable stents used included Palmaz Genesis

(Cordis Corp, East Bridgewater, NJ) in 32.5%, Express

LD iliac stent (Boston Scientiﬁc, Natick, Mass) in 28%,

Assurant Cobalt iliac stent (Medtronic, Minneapolis,

Minn) in 18.6%, Peiron (Biotronik, Berlin, Germany) in

12.8%, and AVE-Bridge (Medtronic) in 2.3%. The two

self-expandable BMSs used were the Smart (Cordis

Corp) in 3.5% and Edwards Life Stent (Bard Peripheral

Vascular Inc, Tempe, Ariz) in 2.3%. No drug-eluting

stents were used.

Eligible

170 Limbs

Excluded – 2 Limbs

 Missing data

Analysed – 81 Limbs

Lost to follow-up – 1 Limb (1 Patient)

 1 Month - 1 Limb

 6 Months - 0 Limb

 12 Months - 0 Limb

 18 Months - 0 Limb

V12

83 Limbs

Lost to follow-up – 8 Limbs (8 Patients)

 1 Month - 4 Limb

 6 Months - 1 Limb

 12 Months - 1 Limb

 18 Months - 2 Limb

Bare Stent

85 Limbs

Analysed – 78 Limbs

Allocation

Analysis

Follow-Up

Randomised

168 Limbs

(125 Patients)

Enrolment

V12

82 Limbs

Bare Stent

86 Limbs

Baseline

Crossover

–

1 Limb

Fig 1. Flow chart shows the trial proﬁle, including descriptions of the patients enrolled in the study and patients who

were lost to follow-up.

JOURNAL OF VASCULAR SURGERY

Volume 54, Number 6

Mwipatayi et al 1563

Medical therapy. All patients received aspirin (100-

150 mg daily) indeﬁnitely and clopidogrel (75 mg daily) for

a minimum of 1 month after the intervention. A 300-mg

loading dose of clopidogrel was given during or after the

intervention.

Study procedures. Access to the iliac lesion was

achieved at the investigator’s discretion via an antegrade

approach from the contralateral femoral artery with the use

of a dedicated 6F or 7F “cross-over” sheath or via a

retrograde (ipsilateral) approach. A brachial approach was

used at the discretion and experience of the investigator

using a Shuttle sheath (Cook Medical, Bloomington, Ind).

After sheath placement, a heparin bolus (3000-5000 units)

was administered intravenously.

DSA was performed to assess and conﬁrm the DUS

imaging or CTA ﬁndings. The patency status of the ipsilat-

eral femoral, popliteal, and infrapopliteal arteries was doc-

umented. After the target lesion was successfully crossed

with a hydrophilic 0.018-inch or 0.035-inch wire, patients

were randomized and received a covered balloon-expandable

stent or a BMS.

At baseline and each follow-up, all patients were strat-

iﬁed by symptoms according to categories outlined previ-

ously by Rutherford et al.

The status of the runoff vessels

was determined using an objective scoring system.

The

superﬁcial and deep femoral arteries, which are the runoff

vessels for iliac artery procedures, were evaluated by the

presence of an occlusion or stenosis, or if they were patent.

Mandatory clinical evaluations combined with mea-

surement of ABIs were performed at each postprocedural

follow-up at 1, 6, 12, and 18 months. DUS imaging of the

aortoiliac segment was also performed at each visit. Binary

restenosis was deﬁned by a doubling of peak systolic veloc-

ities (PSV) across a lesion or a PSV ⬎300 cm/s with

monophasic Doppler waveforms in the distal common

femoral artery.

20-22

Medical events, hospitalizations, access

interventions, and adverse events were documented.

Statistical analysis. Sample size was calculated using

the method proposed by Armitage et al.

Because there

has been no previous RCT comparison between groups

with the two types of stent, an initial limit of noninferiority

between the two groups was estimated to be 10% of the

true population restenosis rate, which we assumed to be 6%.

With power set at 80% and ␣ set at 5% (on a two-sided test),

the number of iliac arteries required was calculated to be 70

in each group, giving a total of 140 limbs. To compensate

for patients lost to follow-up, we included an additional 14

limbs. Thus, the total target sample size was calculated to

be at least 154 limbs. One patient did not receive the

allocated treatment (cross-over treatment). This was a con-

scious decision by the surgeon because the randomized

stent was not available.

Table II. Characteristics of the study patients at baseline

according to treatment groups

Characteristic

V12 stent Bare stent

(n ⫽ 83) (n ⫽ 85)

Patients, No. 62 63

Age, years 65.34 ⫾ 1.43 67.21 ⫾ 1.29

Male sex 67.7 57.1

Race

White 96.8 95.2

Aboriginal 1.6 1.6

Asian 1.6 3.2

Side of lesion

Left 59.0 42.4

Right 39.8 56.5

Central 1.2 1.1

Rutherford-Becker

Category 1 3.2 4.8

Category 2 27.4 30.2

Category 3 40.3 50.8

Category 4 24.2 12.7

Category 5 4.8 1.6

Risk factors

Diabetes 24.2 28.6

Hypertension 56.5 71.4

Smoking 59.7 58.7

Hyperlipidemia 41.9 46.0

Chronic renal failure 3.2 3.2

COAD 9.7 9.5

Coronary artery disease 37.1 38.1

Hypercoagulable states 6.5 3.2

Pre-op antiplatelets 88.7 93.7

Aspirin 87.1 84.1

Clopidogrel 9.7 19.0

COAD, Chronic obstructive airway disease.

Continuous data are expressed as mean ⫾ standard error of the mean, and

categoric data as number or percentage.

Table III. Clinical and radiologic characteristics of the

study patients at baseline according to treatment groups

Characteristic

V12 stent Bare stent

(n ⫽ 83) (n ⫽ 85)

Preoperative assessment

Resting ABI (side of lesion) 0.65 ⫾ 0.03 0.63 ⫾ 0.03

No. 75 78

Duplex scan

Performed 62 (74.7) 65 (76.5)

⬎50% stenosis 87.1 81.5

Occlusion 12.9 18.5

Angiogram

Performed 67 (80.7) 73 (85.9)

TASC B 50.7 72.6

TASC C 34.3 20.5

TASC D 14.9 6.8

CT angiogram

Performed 32 (38.6) 36 (42.4)

TASC B 56.3 63.9

TASC C 31.3 30.6

TASC D 12.5 5.6

Runoff

Performed 74 (89.1) 73 (85.9)

0 0 1.4

1 12.2 5.5

2 20.3 21.9

3 23.0 37.0

4 44.6 34.2

ABI, Ankle-brachial index; TASC, TransAtlantic Inter-Society Consensus.

Data are number (%) or mean ⫾ standard error of the mean.

JOURNAL OF VASCULAR SURGERY

December 2011

1564 Mwipatayi et al

Because this patient did not receive the allocated treat-

ment, we undertook two main analyses. An intention-to-

treat analysis was performed for all baseline and primary end

point data for all randomized patients irrespective of

whether the patient received the randomized study device

or the patient’s compliance with the study protocol. A

per-protocol analysis was performed for all subgroup anal-

yses and secondary end point data because the one cross-

over patient’s data did not create any difference between

the intention-to-treat analysis and the per-protocol analy-

sis.

Baseline characteristics of the two groups were de-

scribed using means and standard errors (SE) for continu-

ous data and proportions for qualitative variables. Freedom

from binary restenosis and freedom from stent occlusion at

1, 6, 12, and 18 months were analyzed with the use of

Kaplan-Meier survival estimates. Cox proportional hazards

models were used to determine if differences between the

estimates were signiﬁcant after establishing the requisite

proportional hazards assumptions were met. Alternative

end points, occlusion, and binary restenosis at 18 months,

were analyzed using logistic regression.

Subgroup analyses of both restenosis end points were

performed by testing the interaction of subgroup and ran-

domized treatment. There were insufﬁcient numbers of

vessels with an occlusion for subgroup analysis to be per-

formed. Logistic regression was also used to investigate the

differences in proportions of patients with limb amputation

between the two randomized groups. Results are reported

as hazard ratios (HR) or odds ratios (OR) with 95% conﬁ-

dence intervals (CI).

When two iliac arteries from the same patient were

randomized, it was necessary to recognize that the vessels

were genetically identical and therefore the outcomes

would be correlated. To adjust for the intracluster correla-

tion, robust sandwich estimators, as proposed by Lin and

Wei (1989),

were used to produce cluster adjusted SEs

and P values for all logistic regression and survival analyses.

Analysis of the change in ABI indexes and TVR over time

was performed using a hierarchic longitudinal mixed-models

analysis. The interaction between randomized treatment

and time was used to determine if there were differences

in the treatment slopes over time for the linear ABI model.

The small number of TVR cases precluded investigation of

the interaction in the logistic model.

All hypotheses tests were performed using two-sided

tests, and the critical value for statistical signiﬁcance was set

at a value of P ⬍ .05. Analyses were conducted using PASW

18 (SPSS, Chicago, Ill), SAS 9.2 (SAS Institute Inc, Cary,

NC), and Stata 11 (StataCorp LP, College Station, Tex)

statistical software.

RESULTS

The trial recruited 125 patients (168 individual iliac

arteries). Randomization assigned 83 vessels to receive a

covered stent and 85 to receive a BMS (Fig 1). Patients

were assessed at the prespeciﬁed intervals for 18 months.

One patient in the covered-stent group and eight in the

BMS group were lost to follow-up.

At baseline, demographic and clinical characteristics

were similar between the two groups, with no signiﬁcant

differences in the frequency of atherothrombotic risk fac-

Fig 2. Kaplan-Meier curves are shown for freedom from binary restenosis for the intention-to-treat population. The

sandwich variance estimate method (Lin and Wei) was used to determine the log-rank test adjusted for clustering (P ⫽

.0243). CI, Conﬁdence interval; HR, hazard ratio.

JOURNAL OF VASCULAR SURGERY

Volume 54, Number 6

Mwipatayi et al 1565

tors or coexisting cardiovascular conditions (Table II).

Most patients in both groups were already receiving anti-

platelet agents, and only three patients were taking oral

anticoagulation medications (Table II). In the BMS group,

Palmaz Genesis, Express LD, Assurant Cobalt, and Peiron

stents were used predominantly.

Preoperative imaging demonstrated 72.6% in the BMS

group had TASC B lesions vs 50.7% in the covered-stent

group (Table III).

Procedural complications included hemorrhage, ﬂow-

limiting dissection, lymph leak, and seroma formation, with

no statistical signiﬁcance between the groups. Intravascular

anticoagulation was administered to 97.6% of patients dur-

ing the procedure. No conversions to open aortoiliac sur-

gery during the perioperative period were reported. Proce-

dural success, deﬁned as ⬍30% residual stenosis during the

angiographic intervention, was achieved in 100% of cases in

both groups.

Primary outcome. Aortoiliac lesions treated with a

covered stent were signiﬁcantly more likely to remain free

from binary restenosis at 18 months than those that were

treated with a BMS (HR, 0.35; 95% CI, 0.15-0.82; P ⫽

.02). Binary restenosis was documented in 8 arteries in the

covered-stent group and in 20 in the BMS group (Fig 2).

Complete occlusions of the stented lesion during the

follow-up period occurred in 3 patients in the covered-stent

group and in 10 patients in the BMS group; however, this

difference was not statistically signiﬁcant (HR, 0.28; 95%

CI, 0.07-1.09; P ⫽ .07; Fig 3).

Subgroup analyses. The subgroup analysis speciﬁcally

looked at freedom from binary restenosis according to the

initially designated TASC lesions. A signiﬁcant interaction

was found between the TASC classiﬁcation and the stent

type for binary restenosis at 18 months (P ⫽ .03). None of

the factors investigated were signiﬁcant (Fig 4). This study

was not powered to investigate interactions; hence, further

research is required to conﬁrm these results. The Kaplan-

Meier survival estimates showed a statistically signiﬁcant

beneﬁt when covered stents were used in TASC C and D

Fig 3. Kaplan-Meier curves are shown for freedom from stent occlusion for the intention-to-treat population. The

sandwich variance estimate method was used to determine the log-rank test adjusted for clustering (P ⫽ .0874). CI,

Conﬁdence interval; HR, hazard ratio.

Fig 4. Odds ratios of 18-month restenosis of ⬎50% in selected

patient subgroups. For most of the variables, the 95% conﬁdence

interval crosses the line of identity, indicating lack of statistical

signiﬁcance. The probability values for interaction indicate that

only TransAtlantic Inter-Society Consensus (TASC) variables in-

teracted on the level of statistical signiﬁcance with the treatment

modality.

JOURNAL OF VASCULAR SURGERY

December 2011

1566 Mwipatayi et al

lesions compared with BMSs (HR, 0.136; 95% CI, 0.042-

0.442; Fig 5; Appendix Fig, online only). However, this

signiﬁcant difference in freedom from binary restenosis was

not demonstrated for TASC B lesions (HR, 0.748; 95% CI,

0.235-2.386; Fig 6).

The 44 patients who had bilateral iliac artery stenting

were randomized to receive 43 covered stents (SE, 0.473;

95% CI, 16.399-18.252) and 45 BMSs (SE, 0.970; 95%

CI, 12.304-16.106). The same stent was allocated to both

limbs in 20 of the 44 patients. The overall freedom from

Fig 5. Kaplan-Meier curves are shown for freedom from binary restenosis for the type of stent used according to the

TransAtlantic Inter-Society Consensus (TASC) C/D group. The sandwich variance estimate method was used to

determine the log-rank test adjusted for clustering (P ⫽ .0056). CI, Conﬁdence interval; HR, hazard ratio.

Fig 6. Kaplan-Meier curves are shown for freedom from binary restenosis for the type of stent used according to the

TransAtlantic Inter-Society Consensus (TASC) B group. There is no statistical difference between the two stents in

patients with TASC B lesion. CI, Conﬁdence interval; HR, hazard ratio.

JOURNAL OF VASCULAR SURGERY

Volume 54, Number 6

Mwipatayi et al 1567

binary restenosis demonstrated that there was signiﬁcantly

less restenosis in covered stents compared with the BMSs

(HR, 0.14; 95% CI, 0.03-0.61; P ⫽ .009). The interaction

of TASC group and randomized stent was not signiﬁcant

(P ⫽ .23).

Secondary outcomes. ABI measurements during the

follow-up period were not signiﬁcantly different between

the two groups at 1 and 6 months. However, the difference

became statistically signiﬁcant at 12 months (P ⫽ .014) and

had marginal signiﬁcance at 18 months (P ⫽. 06), suggest-

ing long-term improvement in ABIs in the covered-stent

group (Table IV).

TVR during the study period demonstrated that there

was less reintervention in the covered-stent group com-

pared with the BMS group (OR, 21; 95% CI, 0.07-0.64;

P ⫽ .006). Most of the reinterventions were performed at

12 and 18 months (Fig 7). Three patients from the BMS

group underwent aortobifemoral bypass grafting.

Four amputations were performed during the study

period. Two patients with covered stents had below knee

amputation at 18 months, which was likely related to

uncontrollable diabetes mellitus and progression of infra-

genicular arterial disease. Two patients with BMSs had

above knee amputations at 18 months due to a combina-

tion of factors. The difference between the two groups was

not statistically signiﬁcant (OR, 1.02; 95% CI, 0.089-

11.73; P ⫽ .984).

A left brachial artery approach was used in 10 patients,

with 1 patient requiring the use of the Outback LTD

re-entry catheter (Cordis Corp), with a successful outcome.

None of these patients had any major complications (no

cerebral ischemic event or vessel perforation) related to the

approach used. Multiple associated procedures were per-

formed during iliac artery stenting: femoropopliteal bypass

graft in ﬁve patients, common femoral artery endarterec-

tomy in four, and superﬁcial femoral artery stenting in four.

External iliac artery stenting was performed in 14 patients,

with successful outcome at 12 and 18 months.

DISCUSSION

The COBEST trial is the ﬁrst RCT to compare covered

stents and BMSs in the management of aortoiliac occlusive

disease. Previous studies have compared different types

of BMSs deployed in iliac arteries and failed to demons-

trate any signiﬁcant differences in technical success and

follow-up outcomes.

25-27

The only published randomized

study comparing primary angioplasty, followed by selective

or primary stent placement in patients with iliac artery

obstructive disease, is the Dutch Iliac Stent Trial (DIST).

The short-term and long-term DIST results indicated that

selective stent placement should be considered as the treat-

ment of choice.

Angioplasty is a less invasive alternative treatment to

open bypass surgery and proven efﬁcacy for the treatment

of patients with focal iliac artery stenosis. The procedural

technical success rate has improved signiﬁcantly (up to

95%), especially when adjunctive stent placement is used.

Patency rates of 80% to 90% after 5 years have been re-

ported for short iliac stenoses, which is comparable with

patency results for open surgery.

28,29

For complex, multi-

focal, or totally occluded atherosclerotic segments of iliac

arteries (C and D lesions), TASC recommends surgery as

the procedure of choice.

Since their ﬁrst applications by Dotter, endovascular

metallic stent placement has been widely used for the

treatment of patients with aortoiliac occlusive disease.

30,31

Vorwerk et al

32,33

reported 4-year patency rates of 78% and

82% for primary and secondary stenting procedures, respec-

tively, in 100 patients. Cikrit et al

reported a 5-year

patency rate of 63% in 38 limbs that were treated by Palmaz

stent placement, whereas Palmaz et al

initially reported a

92% patency rate at 9 months. Primary 4-year patency rates

as high as 86% were recently reported.

25,31

Covered stents are able to exclude plaque and endothe-

lium, thereby potentially mitigating late luminal loss by

halting migration and proliferation of vascular smooth

muscle cells and inﬂammatory cells through open stent

struts.

This may result in a reduction of restenosis caused

by luminal encroachment from extracellular matrix deposi-

tion intimal hyperplasia.

15,36

BMSs do not provide a

boundary that excludes the underlying plaque from the

lumen, and stent oversizing relative to the reference lumen

causes an increase in neointimal growth. Covered stents

may also offer the beneﬁt of being less thrombogenic than

BMSs. In the United States, no covered stents are currently

approved by the U.S. Food and Drug Administration for

application in the iliac arteries for occlusive disease. How-

ever, homemade covered stents, iliac limbs from modular

aortic stent grafts, and commercially manufactured covered

stents approved for other nonvascular indications have

been placed in the iliac arteries. Because of the bulk of the

additional graft material, covered stents require larger de-

livery systems than BMSs. This may expose the patient to a

higher risk of groin complications.

The restenosis observed in the covered-stent group was

predominantly located at the ends of the stent or outside

the stent. We hypothesis that this may be secondary to

compliance mismatch due to the use of the rigid end of the

covered stent and balloon dilatation causing the increased

stent–artery interface intimal hyperplasia. In addition, the

stent may not cover the lesion fully, resulting in progression

of the atherosclerotic process by activation from balloon

inﬂation but with the end of the balloon inﬂated ﬁrstly

Table IV. Resting ankle-brachial index (ABI) on side of

the lesion

Resting

ABI

V12 stent Bare stent

PNo. Mean ⫾ SEM No. Mean ⫾ SEM

Baseline 75 .65 ⫾ .03 78 .63 ⫾ .03 .639

1 month 71 .91 ⫾ .03 72 .91 ⫾ .03 .927

6 months 73 .89 ⫾ .02 74 .88 ⫾ .03 .653

12 months 75 .94 ⫾ .02 79 .85 ⫾ .03 .014

18 months 70 .94 ⫾ .02 73 .86 ⫾ .03 .07

SEM, Standard error of the mean.

JOURNAL OF VASCULAR SURGERY

December 2011

1568 Mwipatayi et al

outside of the stent causing a process known as the “dog

bone” effect.

We have identiﬁed three major limitations in this study:

First, the different stent types used in the BMS group may

have affected the results and led to bias. It was impractical in

the context of a multicenter study to limit the BMS type.

We suggested current practices and protocols within trial

centers should continue unchanged, with the type of BMS

selected left to the individual operator.

Second, the BMS group had fewer patients with a

TASC D lesion (7%) compared with the covered-stent

group (16.1%), which, despite the randomization, may

have inﬂuenced the results.

Third, DUS was the principal imaging tool used for

follow-up, which may have introduced bias because it is

highly operator-dependent. We decided to use DUS imag-

ing because we were reluctant to unnecessarily use invasive

investigations if not clinically required.

CONCLUSIONS

There is increasing evidence from single-center clinical

investigations that patients with complex aortoiliac lesions,

including chronic iliac artery occlusions and occlusion of

the aortoiliac bifurcation, can be treated safely and effec-

tively with a covered stent. The COBEST results demon-

strate that for patients with severe aortoiliac arterial occlu-

sive disease, there is an increased freedom from restenosis

and occlusion with covered stents compared with BMSs at

12 and 18 months. However, long-term durability data

(5-year follow-up) is desirable.

We thank Dr Michael Phillips for his statistical advice.

AUTHOR CONTRIBUTIONS

Conception and design: BM

Analysis and interpretation: BM, JW, ST

Data collection: MJ, JW, ST

Writing the article: BM ST, JW, ST, VV

Critical revision of the article: VV, ST,

Final approval of the article: BM, VV

Statistical analysis: BM, ST SB

Obtained funding: BM, MJ

Overall responsibility: BM

REFERENCES

1. Wylie EJ. Thromboendarterectomy for arteriosclerotic thrombosis of

major arteries. Surgery 1952;32:275-92.

2. de Vries SO, Hunink MG. Results of aortic bifurcation grafts for

aortoiliac occlusive disease: a meta-analysis. J Vasc Surg 1997;26:

558-69.

3. Littooy FN, Steffan G, Steinam S, Saletta C, Greisler HP. An 11-year

experience with aortofemoral bypass grafting. Cardiovasc Surg 1993;1:

232-8.

4. Schneider JR, Besso SR, Walsh DB, Zwolak RM, Cronenwett JL.

Femorofemoral versus aortobifemoral bypass: outcome and hemody-

namic results. J Vasc Surg 1994;19:43-55.

5. Huber TS, Harward TR, Flynn TC, Albright JL, Seeger JM. Operative

mortality rates after elective infrarenal aortic reconstructions. J Vasc

Surg 1995;22:287-93.

6. Erdoes LS, Bernhard VM, Berman SS. Aortofemoral graft occlusion:

strategy and timing of reoperation. Cardiovasc Surg 1995;3:277-83.

7. Passman MA, Taylor LM, Moneta GL, Edwards JM, Yeager RA,

McConnell DB, et al. Comparison of axillofemoral and aortofemoral

bypass for aortoiliac occlusive disease. J Vasc Surg 1996;23:263-9.

8. Rouers A, Meurisse N, Lavigne JP, Francart D, Quaniers J, Desiron Q,

et al. Potential beneﬁts of laparoscopic aorto-bifemoral bypass surgery.

Acta Chir Belg 2005;105:610-5.

Fig 7. Kaplan-Meier curves for freedom from target vessel revascularization during the study period showed fewer

reinterventions occurred in the covered-stent group (P ⫽ .006 at 12 months). CI, Conﬁdence interval; OR, odds ratio;

TVR, target vessel revascularization.

JOURNAL OF VASCULAR SURGERY

Volume 54, Number 6

Mwipatayi et al 1569

9. Dormandy JA, Rutherford RB. Management of peripheral arterial dis-

ease (PAD). TASC Working Group. TransAtlantic Inter-Society Con-

sensus (TASC). J Vasc Surg 2000;31:S1-296.

10. Norgren L, Hiatt WR, Dormandy JA, Nehler MR, Harris KA, Fowkes

FG, et al. Inter-Society Consensus for the Management of peripheral

Arterial Disease (TASC II). J Vasc Surg 2007;45(Suppl 1):S5-67.

11. Tetteroo E, van der Graaf Y, Bosch JL, van Engelen AD, Hunink MG,

Eikelboom BC, et al. Randomized comparison of primary stent place-

ment vs primary angioplasty followed by selective stent placement in

patients with iliac artery obstructive disease. Lancet 1998;351:1153-9.

12. Bosch JL, Hunink MG. Meta-analysis of the results of percutaneous

transluminal angioplasty and stent placement for aortoiliac occlusive

disease. Radiology 1997;204:87-96.

13. Marin ML, Veith FJ, Cynamon J, Parsons RE, Lyon RT, Suggs WD, et

al. Effect of polytetraﬂuoroethylene covering of Palmaz stents on the

development of intimal hyperplasia in human iliac arteries. J Vasc Interv

Radiol 1996;7:651-6.

14. Palmaz J. Intravascular stents: tissue-stent interactions and design con-

siderations. Am J Roentgenol 1993;160:613-8.

15. Diethrich EB. Polymeric covers and coats for metallic stents: micropo-

rous PTFE and the inhibition of intimal hyperplasia. J Endovasc Surg

1995;2:266-71.

16. Henry M, Amor M, Henry I, Klonaris C, Tzvetanov K, Buniet JM, et al.

Percutaneous endovascular treatment of peripheral aneurysms. J Car-

diovasc Surg (Torino) 2000;41:871-83.

17. Ohki T, Veith FJ. Five-year experience with endovascular grafts for the

treatment of aneurysmal, occlusive and traumatic arterial lesions. Car-

diovasc Surg 1998;6:552-65.

18. Wiesinger B, Beregi JP, Oliva VL, Dietrich T, Tepe G, Bosiers M, et al.

PTFE-covered self-expanding nitinol stents for the treatment of severe

iliac and femoral artery stenoses and occlusions: ﬁnal results from a

prospective study. J Endovasc Ther 2005;12:240-6.

19. Ahn SS, Rutherford RB, Becker GJ, Comerota AJ, Johnston KW,

McClean GK, et al. Reporting standards for lower extremity arterial

endovascular procedures. Society for Vascular Surgery/International

Society for Cardiovascular Surgery. J Vasc Surg 1993;17:1103-7.

20. Zeller T, Tiefenbacher C, Steinkamp HJ, Langhoff R, Wittenberg G,

Schlüter M, et al. Nitinol stent implantation in TASC A and B superﬁcial

femoral artery lesions: the Femoral Artery Conformexx Trial (FACT). J

Endovasc Ther 2008;15:390-8.

21. Krankenberg H, Schlüter M, Steinkamp HJ, Bürgelin K, Scheinert D,

Schulte KL, et 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;

116:285-92.

22. Diehm N, Pattynama PM, Jaff MR, Cremonesi A, Becker GJ, Hopkins

LN, et al. Clinical endpoints in peripheral endovascular revasculariza-

tion trials: a case for standardized deﬁnitions. Eur J Vasc Endovasc Surg

2008;36:409-19.

23. Armitage P, Berry G, Matthews J. Statistical methods in medical re-

search. Oxford: Blackwell Science; 2002. p. 137-46.

24. Lin DY, Wei LJ. The robust inference for the Cox proportional hazards

model. J Am Stat Ass 1989;84:1074-8.

25. Palmaz JC, Laborde JC, Rivera FJ, Encarnacion CE, Lutz JD, Moss JG.

Stenting of the iliac arteries with the Palmaz stent: experience from a

multicenter trial. Cardiovasc Intervent Radiol 1992;15:291-7.

26. Strecker EP, Hagen B, Liermann D, Schneider B, Wolf HR,

Wambsganss J. Iliac and femoropopliteal vascular occlusive disease

treated with ﬂexible tantalum stents. Cardiovasc Intervent Radiol 1993;

16:158-64.

27. Long AL, Sapoval MR, Beyssen BM, Auguste MC, Le Bras Y, Raynaud

AC, et al. Strecker stent implantation in iliac arteries: patency and

predictive factors for long-term success. Radiology 1995;194:739-44.

28. Motarjeme A, Keifer JW, Zuska AJ. Percutaneous transluminal angio-

plasty of the iliac arteries: 66 experiences. AJR Am J Roentgenol

1980;135:937-44.

29. Scheinert D, Schröder M, Ludwig J, Bräunlich S, Möckel M, Flachs-

kampf FA, et al. Stent-supported recanalization of chronic iliac artery

occlusions. Am J Med 2001;110:708-15.

30. Szilagyi DE, Elliott JP Jr, Smith RF, Reddy DJ, McPharlin M. A

thirty-year survey of the reconstructive surgical treatment of aortoiliac

occlusive disease. J Vasc Surg 1986;3:421-36.

31. Sacks D, Marinelli DL, Martin LG, Spies JB, Society of Interventional

Radiology Technology Assessment Committee. Reporting standards

for clinical evaluation of new peripheral arterial revascularization de-

vices. J Vasc Interv Radiol 2003;14:S395-404.

32. Vorwerk D, Guenther RW, Schürmann K, Wendt G, Peters I. Primary

stent placement for chronic iliac artery occlusions: follow-up results in

103 patients. Radiology 1995;194:745-9.

33. Vorwerk D, Günther RW, Schürmann K, Wendt G. Aortic and iliac

stenoses: follow-up results of stent placement after insufﬁcient balloon

angioplasty in 118 cases. Radiology 1996;198:45-8.

34. Cikrit DF, Gustafson PA, Dalsing MC, Harris VJ, Lalka SG, Sawchuk

AP, et al. Long-term follow-up of the Palmaz stent for iliac occlusive

disease. Surgery 1995;118:608-13.

35. Farb A, Sangiorgi G, Carter AJ, Walley VM, Edwards WD, Schwartz

RS, et al. Pathology of acute and chronic coronary stenting in humans.

Circulation 1999;99:44-52.

36. Palmaz JC. Intravascular stents: tissue-stent interactions and design

considerations. Am J Roentgenol 1993;160:613-8.

Submitted Feb 19, 2011; accepted Jun 28, 2011.

Additional material for this article may be found online

at www.jvascsurg.org.

JOURNAL OF VASCULAR SURGERY

December 2011

1570 Mwipatayi et al

Appendix Fig (online only). Kaplan-Meier survival estimates are shown for patients with TransAtlantic Inter-Society

Consensus (TASC) B, C, and D lesions undergoing placement of the bare-metal stent and the covered (V12) stent.

JOURNAL OF VASCULAR SURGERY

Volume 54, Number 6

Mwipatayi et al 1570.e1