Clinical efficacy and safety of the PRO-glide device as a sUture-mediated ClosurE in Thoracic EndoVascular Aortic Repair in patients with previous groin intervention (from the PRODUCE-TEVAR Trial)

Abstract

Background

While the percutaneous approach is increasingly preferred, suture-mediated closure devices have been put into clinical practice to close the femoral artery during procedures requiring a large-sized introducer. However, scar in the groin is considered a contraindication or an exclusion criterion for percutaneous procedures. The aim of our study was to investigate the outcomes and safety of Pro-Glide device as suture-mediated closure device in patients who underwent thoracic endovascular aortic repair with percutaneous femoral access ≥22 F who had previous groin intervention.

Methods

A total of 73 patients who underwent endovascular repair with percutaneous femoral access were retrospectively included in the study. Previous groin intervention was defined as history of open surgical access or large sheath insertion (>18 F) to femoral artery because of endovascular or valvular intervention. Patients were divided into two groups as who had previous groin intervention PGI (+) and had not PGI (−).

Results

A total of 73 patients [60 male (82.2%)] were included in the study. Seventeen patients had PGI, and 56 did not. When groups were compared in terms of sheath sizes, a significantly higher sheath sizes were used in PGI (+) patients (24.5 ± 1.1 F vs. 23.8 ± 0.9 F, p = 0.005). The overall success rate in the femoral approach with pre-close technique was statistically insignificant between two groups (94.1% vs. 96.4%, p = 0.55). One patient in PGI (+) group and two patients in PGI (−) had technical failure for percutaneous femoral approach. One patient (5.9%) in PGI (+) group and one patient (1.8%) in PGI (−) group had femoral complications after the procedures; however, there was no significant difference between the groups in terms of complications (5.9% vs. 1.8%, p = 0.13).

Conclusion

Pro-Glide device may be a safe and less invasive method for femoral access in patients with PGI and might not be considered as a contraindication for patients with history of PGI.

Keywords

TEVAR groin scar Pro-Glide percutaneous closure vascular closure device femoral access

Introduction

During the past three decades, several studies carried out with the gradual development of the percutaneous approach in endovascular repair demonstrated convenience and reduced ambulation time for the percutaneous approach.¹ While the percutaneous approach is increasingly preferred, suture-mediated closure devices have been put into clinical practice to close the femoral artery during procedures requiring a large-sized introducer.² These devices decreased vascular complications in conjunction with a less invasive approach to vascular repair which decreased manual compression and decreased bed rest in these patients.³,⁴ However, local complications such as groin infection, hematoma, pseudoaneurysm might occur after percutaneous femoral access.⁵ The current literature demonstrated risk factors associated with technical failure of the suture-mediated closure devices including obesity, operator experience, sheath size, the anatomy of the femoral artery and anatomy of groin, which included small luminal diameter of femoral artery, calcification in femoral artery, scar tissue in groin area.⁶,⁷ Moreover, previous intervention to the groin was found to be a predictor of technical failure during percutaneous access due to increased fibrosis in the groin.⁸ Hence, in some studies, the scar in the groin is considered a contraindication or an exclusion criterion for percutaneous procedures.^9–11

The aim of our study was to investigate the outcomes and safety of Pro-Glide device as suture-mediated closure device in patients who underwent thoracic endovascular aortic repair with percutaneous femoral access ≥22 F who had previous groin intervention. The outcome and safety of the device were evaluated by investigating post-procedural complications and technical success for femoral artery closure.

Materials and methods

Study population

This retrospective study was conducted in accordance with the principles of the Helsinki Declaration and approved by the local Institutional Review Board. A total of 73 patients who underwent thoracic endovascular repair with percutaneous femoral access in a single tertiary university hospital between December 2015 and March 2020 were included in the study. The inclusion criteria were as follows: availability of all medical records; evaluation by contrast-enhanced computed tomography (CT) carried out during clinical follow-up; patients who were provided closure of the femoral access site using Pro-Glide device. Patients who did not meet all these criteria were excluded from the study. Body mass index was calculated by the following formula: BMI = weight/height² (kg/m²). Presence of coronary artery disease and peripheral artery disease was noted for the patients. Cardiovascular risk factors such as hypertension, hyperlipidemia, diabetes mellitus and smoking status were noted.¹² Glomerular filtration rate (GFR) was calculated according to the creatinine levels measured before endovascular procedure with Cockroft-Gault formula.¹³ Previous groin intervention was defined as history of open surgical access or large sheath insertion (>18 F) to femoral artery because of endovascular or valvular intervention. Patients were divided into two groups as who had previous groin intervention PGI (+) and had no PGI (−). Most of the patients in PGI (+) group included patients who required re-intervention of TEVAR due to various reasons.

Femoral access

The procedures were performed via the femoral artery route and were closed with Pre-close technique using Pro-Glide Device (Abbott Laboratories, Chicago, IL, USA). Before procedure, all patients underwent CT evaluation to confirm the diagnosis for aortic and femoral anatomy. The data about common femoral artery (CFA) diameter, CFA distance to skin and calcification grading were determined under the guidance of CT images. CFA depth and cross-sectional diameter were measured from the reference point of mid-femoral head to a straight line from anterior vessel wall to the skin surface overlying the artery. This reference point is typically above the CFA bifurcation. CFA calcification was scored on the basis of presence or absence of atherosclerotic plaque from the superficial epigastric artery to femoral bifurcation (normal = 0, posterior wall plaque without calcium involving <50% of luminal diameter = 1, plaque or calcium involving ≥50% of luminal diameter = 2, anterior wall calcification = 3, circumferential calcification = 4) which is defined in a previous study.¹⁴

Pre-close technique was performed similar to the previous studies.¹⁵ First, a 7 F femoral sheath which was routinely used for coronary interventions, was introduced to CFA. The confirmation of appropriate placement of the sheath was routinely performed with manual contrast injection. After confirming the true access for CFA, a 0.035-inch guide-wire was introduced to CFA and the 7 F sheath was removed. Following removal of the sheath, two Pro-Glide devices were inserted to CFA using the guide-wire. Before device insertion, if the patient had PGI, subcutaneous scarred tissue was properly and gently dissected and prepared for device insertion. After dissecting gently and preparing the subcutaneous tissue, we performed multiple consecutive dilatations to femoral artery and surrounding tissue with dilatators of each increasing diameter. Each of the Pro-Glide sutures were left extra-corporeally and stabilized with mosquitos. One of the Pro-Glides was advanced with 30° of medial rotation and the other one was advanced with 30° of lateral rotation. After both of the devices’ sutures were deployed, 7 F femoral sheath was reintroduced to stabilize the femoral hemostasis.

Procedural data

Endovascular procedures and pre-close techniques were performed by the same two operators (Ö.Ç. and M.E.) in all patients undergoing TEVAR. All patients were intubated during the procedure. During endovascular intervention, 7 F sheath was removed and larger diameter introducer sheath was introduced to the femoral artery for delivering endovascular devices. In a few of the cases, procedures were performed without introducer sheath (sheathless) and only with device catheter itself which was 25 F. After the conclusion of endovascular repair, introducer sheath was removed and while manual compression was applied, both of Pro-Glide knots were cinched over the guide-wire. If the hemostasis was maintained, guide-wire was removed and manual compression was continued to the femoral area. If the hemostasis could not be maintained in femoral artery, then another sheath was replaced in the artery via guide-wire and arteriotomy was repaired surgically. Procedural data during endovascular intervention were noted for every patient. These data included the size of the device used, operating room time, hospital length of the stay, manual compression time to femoral artery after procedure and technical success. Operating room time was defined as period of time starting from skin puncture to final dressing application. Hospital length of the stay was defined as days spent in hospital from admission to hospital until discharge. Manual compression time was defined as the period of time as manual hand compression started to femoral access site just after suturing was performed with Pre-close technique until the bleeding fully stopped from the groin access site. Technical success for femoral artery closure was defined as the closure of arteriotomy without the need for any additional surgical or endovascular procedure regarding hemorrhagic or ischemic complications for Pre-close technique. Access-site-related complications for 30-day such as hematomas with or without transfusions, wound dehiscence requiring dressing changes, seromas, pseudoaneurysms and infections were noted. Moderate to large (2 cm) asymptomatic or subclinical hematomas or seromas and arteriovenous fistulas were also included as complications. After discharge of the patients, first postoperative CT scan including femoral vessels at the one-month of follow-up visit was evaluated for every patient.

Statistical analyses

Data were analyzed using the Statistical Package for the Social Sciences, version 24.0 (SPSS Inc., Chicago, IL, USA). Whether the variables show normal distribution, visual (histograms, probability curves) and analytical methods (Kolmogorov–Smirnov and Shapiro–Wilk) were evaluated. Numerical variables showing normal distribution were expressed as mean ± standard deviation (SD), numerical variables not showing normal distribution were expressed as median (interquartile range) and categorical variables as percentage (%). Numerical variables were evaluated using Student t-tests and the Mann–Whitney U-test between the two groups. Chi square or Fisher exact test were used to compare the categorical variables. Throughout the present study, a p-value of <0.05 was considered significant.

Results

A total of 73 patients [mean age: 59.93 Years, 60 male (82.2%)] were included in the study. Demographic and clinical characteristics of the patients are summarized in Table 1. Prior to the procedure, 17 patients had PGI, and 56 did not. In PGI (+) group, 14 patients had open surgical access previously and 3 patients had percutaneous femoral access previously. There was no significant difference between the two groups based on their demographic and clinical characteristics. In PGI (+) group, all the patients had only one time access previously to the femoral artery.

Table 1.

Demographic and clinical characteristics of the patients.

	Previous groinintervention (+)(n = 17)	Previous groinintervention (−)(n = 56)	p value
Age, years	58.4 ± 9.4	60.4 ± 10.1	0.497
Male (n – %)	15 (88.2%)	45 (80.4%)	0.719
Body mass index, kg/m²	26.2 ± 5.6	26.9 ± 6.7	0.312
Diabetes mellitus (n – %)	5 (29.4%)	15 (26.8%)	1.0
Hypertension (n – %)	17 (100.0%)	54 (96.4%)	1.0
Hyperlipidemia (n – %)	2 (11.7%)	6 (10.7%)	0.244
Smoking (n – %)	8 (47.1%)	21 (37.5%)	0.481
Glomerular filtration rate (ml / min)	97 (77–106)	86 (40–103)	0.151
Coronary artery disease (n – %)	6 (35.3%)	8 (14.3%)	0.078
Peripheral artery disease (n – %)	1 (5.9%)	4 (7.1%)	1.0

The anatomical features of the CFA based on CT examination, and properties of the device and sheath dimensions used during the endovascular repair, are shown in Table 2. Additionally, detailed information about the sizes and usage percentages in the groups is given in Figure 1. When groups were compared in terms of sheath sizes, a significantly higher sheath size was used in PGI (+) patients (24.5 ± 1.1 F vs. 23.8 ± 0.9 F, p = 0.005). Also, a higher number of 25 and 26 F sheaths were used during the procedure of the PGI (+) group, while a higher number of 24 F sheaths were used in the PGI (−) group. There were two patients in PGI (+) group who had sheath size of 26 F and none in PGI (−) group (11.8% vs. 0%, p = 0.052). There were 18 patients who had sheath size of 25 F. The procedures to these 18 patients were performed without sheath (sheathless) and the catheter device was introduced directly via femoral artery. There was no significant difference between the groups in terms of anatomic characteristics of CFA. When the patients using 25 F sheath were compared, the percentage of patients in the PGI (+) group was statistically more significant (47.1% vs. 17.9%, p = 0.024), but there was no difference between the two groups in terms of technical success (100% vs. 100%). In a total of 41 patients, 24 F sheaths were used. In contrast to the use of 25 and 26 F sheaths, the use of 24 F sheaths was higher in the PGI (−) group (64.3% vs. 29.4%, p = 0.011). There was no statistical difference between the two groups regarding technical success (94% vs. 100%). Moreover, when groups were compared in terms of the device size used for endovascular repair, the device size was statistically significantly higher in the PGI (+) group in parallel with the sheath size (41.8 ± 4.2 vs. 37.3 ± 4.7, p < 0.001).

Table 2.

Anatomical characteristics of CFA detected in computerized tomography and characteristics of sheath and devices used for patients.

	Previous GroinIntervention (+)(n = 17)	Previous GroinIntervention (−)(n = 56)	p value
CFA diameter, mm	10.5 ± 1.5	10.6 ± 1.9	0.958
Sheath size diameter, F	24.5 ± 1.1	23.8 ± 0.9	0.005
22	2 (11.8%)	10 (17.9%)	0.720
24	5 (29.4)	36 (64.3%)	0.011
25	8 (47.1%)	10 (17.9%)	0.024
26	2 (11.8%)	0 (0.0%)	0.052
Skin to CFA distance, mm	37.1 ± 12.3	33.1 ± 12.8	0.194
Calcification grade of femoral artery	0 (0–1)	1 (0–1)	0.516
Device size used for endovascular repair	41.8 ± 4.2	37.3 ± 4.7	<0.001

Note. The significant p-values are in bold.

Figure 1.

Figure shows distribution of sheath sizes in prior groin intervention – (blue) and prior groin intervention + (red) cohorts.

Procedural data, complications and clinical outcomes are shown in Table 3. There was no significant difference between the groups in terms of operating room time, hospital stay, technical success and complication rates. Every patient needed two Pro-Glides for suture closure during the procedure. However, because of anatomical difficulties, one patient in PGI (+) and two patients in PGI (−) group needed a third device during the procedure. However, there was no statistical difference between the groups in terms of number of closure devices used. The overall success rate in the femoral approach with pre-close technique was not statistically significant between the two groups (94.1% vs. 96.4%, p = 0.55). The relationship between sheath sizes and technical success rates for percutaneous femoral approach in groups is given in Figure 2. One patient in PGI (+) group and two patients in PGI (−) had technical failure for percutaneous femoral approach. The technical success rate of pre-close technique with Pro-Glide was 1/2 (50%) for femoral access with 26 F and 15/15 (100%) for femoral access with <26 F in patients who had history of previous groin intervention. In patients with technical failure, femoral arterial puncture could be performed and procedures were concluded; however, these three patients had to undergo surgical open repair due to persistent bleeding after the procedure during their hospital stay in the first 48 h. None of the patients needed covered stents or femoral grafts. All of these three patients had interventions via ≥24 F size of sheath in femoral artery. One patient in PGI (+) group and one patient in PGI (−) group had femoral complications after the procedures. PGI (+) group had higher rate of complication when compared to PGI (−) group; however, there was no statistical significance. (p = 0.13). Both of the patients who had femoral complications had hematoma in femoral area in post-operative 30 days follow-up and these patients did not need any surgical intervention for hematoma.

Table 3.

Procedural data, clinical outcomes and procedural complications.

	Previous groinintervention (+)(n = 17)	Previous groinintervention (−)(n = 56)	p value
Operating room time, min	87.7 ± 6.9	88.9 ± 10.0	0.898
Manual compression time, min	8.6 ± 1.4	8.2 ± 1.3	0.232
Number of Pro-Glides used
• 2 devices	16 (94.1%)	54 (96.4%)	0.456
• 3 devices	1 (5.9%)	2 (3.6%)	0.367
Technical success for femoral artery closure	16 (94.1%)	54 (96.4%)	0.554
Complications	1 (5.9%)	1 (1.8%)	0.133
• Groin infection	0 (0.0%)	0 (0.0%)	0.121
• Pseudoaneurysm	0 (0.0%)	0 (0.0%)
• Hematoma / seroma	1 (5.9%)	1 (1.8%)

Figure 2.

Figure shows percentages of technical success depending on sheath sizes in prior groin intervention – (blue) and prior groin intervention + (red) cohorts.

Discussion

In our study, we have compared the outcome of Pro-Glide device usage in patients with previous groin intervention who underwent TEVAR with large-size sheath. The main finding of our study was that there was no significant difference between patients with previous groin intervention and without in terms of percutaneous femoral access ≥22 F size of sheath.

There are several studies evaluating the safety of Pro-Glide or any other suture-mediated closure devices in endovascular procedures. Moreover, there is plenty of knowledge about comparing data between suture-mediated closure devices and surgical approach. In most of these studies, previous groin intervention or groin scar initiated as exclusion criterion.¹⁶,¹⁷ In the past, scars in the groin were considered as a contraindication for femoral percutaneous procedures.¹⁸,¹⁹ It was suggested that scar and fibrosis were deflecting the needle and causing failure or inappropriate puncture of femoral artery which indeed caused technical failure of femoral access. However, in a study, Eisenack et al.⁸ previously reported that percutaneous approach for endovascular repair with size of sheaths between 14 F and 24 F found that presence of scar in groin increased the likelihood of post-procedural late repairs of femoral artery and they commented that scar in the groin area increased the risk of complications in terms of percutaneous femoral access. In their study, clinical and procedural parameters were not stated in terms of technical failure such as sizes of the sheaths failed and anatomical characteristics of the femoral artery. Additional to that, Eisenack et al. used another brand of suture-mediated closure device which was Prostar XL. On the other hand, Rachel et al.²⁰ compared the outcome of percutaneous femoral access with surgical femoral access, and they found in multivariate analysis that there was no correlation between scar tissue in femoral area and successful percutaneous femoral access. However, Rachel et al. used smaller size of sheaths different to our study. Their study included patients who underwent introducing sheath sizes between 16 and 22 F and in sheath sizes were between 22 and 26 F in our study. Additionally, Rachel et al. also performed femoral access with Prostar XL. It was known that Prostar XL was associated with higher rates of major and minor bleeding risks.²¹ In another study, Zakko et al.²² researched association between obesity and percutaneous femoral access using Pro-Glide and their finding demonstrated that scar in groin which was related to prior groin operation was not associated with technical failure of percutaneous femoral access in obese patients. These studies show us that the data about association between scar in groin and percutaneous femoral access are scarce. Our study is the first study in literature evaluating outcome of percutaneous femoral access with previous groin intervention and comparing with the patients without previous groin intervention who underwent TEVAR with size of sheaths ≥22 F.

During intervention to scar tissue in femoral area, not only needle insertion but also sheath insertion is strenuous. One of the most crucial parts for intervention to scar tissue is preparing the scarred subcutaneous area for appropriate sheath insertion. There are different methods that can be used for easier sheath insertion to scarred femoral area. In a study, Pecoraro et al.²³ administrated predilation with balloon angioplasty to femoral artery to facilitate large size sheath through femoral scar tissue. In our experience, before Pro-Glide placement for Pre-close technique, we properly and gently dissected the subcutaneous tissue and after reaching to femoral artery, we performed multiple consecutive dilatations to femoral artery with dilatators of each increasing diameter. This approach gives us easier insertion of Pro-Glide and large size sheath through scar tissue in femoral area with less traumatic touch to groin.

In PGI (+) and PGI (−) groups, technical success was 94.1% and 96.4%, respectively and total complication rate was 2.7%. When compared to literature, even though our study had larger size of sheaths, our finding was similar to the findings in review of previous studies.²⁴

Study limitations

There were several limitations in this study. First, and foremost, the retrospective nature of the study inherently limits the generalizability of our results. Second, this study is a rather small sized, single center study which might limit the generalizability and reliability of our results. Third, we only used Pro-Glide device so we could not generalize the results of our study to all suture-mediated closure devices. Fourth, we did not use routine ultrasound during puncture of femoral artery, and routine ultrasound usage might decrease technical failure during femoral access. Finally, the follow-up period was only 30 days and we did not have the data in long-term outcomes in our study.

Conclusion

PGI might not be a contraindication for Pre-close technique using Pro-Glide device to introduce large size sheath even up to 26 F. Moreover, the Pro-Glide device may be a safe and less invasive method for the closure of the femoral access site in these patients. However, larger randomized studies are required to confirm the efficacy and safety of the Pro-Glide device in patients with PGI.

Footnotes

Acknowledgement

None.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Ahmet Anıl Şahin

Ali Rıza Demir

References

Torsello

Kasprzak

Klenk

, et al. Endovascular suture versus cutdown for endovascular aneurysm repair: a prospective randomized pilot study. J Vasc Surg 2003; 38: 78–82.

Hogg

Kibbe

MR.

Percutaneous thoracic and abdominal aortic aneurysm repair: techniques and outcomes. Vascular 2006; 14: 270–281.

Vinayakumar

Kayakkal

Rajasekharan

, et al. 24h and 30 day outcome of Perclose Proglide suture mediated vascular closure device: an Indian experience. Indian Heart J 2017; 69: 37–42.

Nasu

Tsuchikane

Sumitsuji

, et al. The safety and efficacy of “pre-closure” utilizing the Closer suture-mediated vascular closure device for achievement of hemostasis in patients following coronary interventions: results of the second Perclose Accelerated Ambulation and Discharge (PARADISE II) Trial. J Invasive Cardiol 2005; 17: 30–33.

Hajibandeh

Antoniou

Child

, et al. Percutaneous access for endovascular aortic aneurysm repair: a systematic review and meta-analysis. Vascular 2016; 24: 638–648.

Kim

Shin

, et al. Efficacy and safety of the preclose technique following percutaneous aortic stent-graft implantation. J Endovasc Ther 2013; 20: 350–355.

Nara

Watanabe

Kozuma

, et al. Incidence, predictors, and mid-term outcomes of percutaneous closure failure after transfemoral aortic valve implantation using an expandable sheath (from the optimized transcatheter valvular intervention [OCEAN-TAVI] Registry). Am J Cardiol 2017; 119: 611–617.

Eisenack

Umscheid

Tessarek

, et al. Percutaneous endovascular aortic aneurysm repair: a prospective evaluation of safety, efficiency, and risk factors. J Endovasc Ther 2009; 16: 708–713.

Teh

Sieunarine

van Schie

, et al. Use of the percutaneous vascular surgery device for closure of femoral access sites during endovascular aneurysm repair: lessons from our experience. Eur J Vasc Endovasc Surg 2001; 22: 418–423.

10.

McDonnell

Forlee

Dowdall

, et al. Percutaneous endovascular abdominal aortic aneurysm repair leads to a reduction in wound complications. Ir J Med Sci 2008; 177: 49–52.

11.

Howell

Villareal

Krajcer

Percutaneous access and closure of femoral artery access sites associated with endoluminal repair of abdominal aortic aneurysms. J Endovasc Ther 2001; 8: 68–74.

12.

Mahmood

Levy

Vasan

, et al. The Framingham Heart Study and the epidemiology of cardiovascular disease: a historical perspective. Lancet 2014; 383: 999–1008.

13.

Drinka

Langer

The Cockroft-Gault formula. J Am Geriatr Soc 1989; 37: 820.

14.

Hayashida

Lefèvre

Chevalier

, et al. Transfemoral aortic valve implantation new criteria to predict vascular complications. JACC Cardiovasc Interv 2011; 4: 851–858.

15.

Lee

Brown

Nelson

, et al. Total percutaneous access for endovascular aortic aneurysm repair ("Preclose" technique). J Vasc Surg 2007; 45: 1095–1101.

16.

Jean-Baptiste

Hassen-Khodja

Haudebourg

, et al. Percutaneous closure devices for endovascular repair of infrarenal abdominal aortic aneurysms: a prospective, non-randomized comparative study. Eur J Vasc Endovasc Surg 2008; 35: 422–428.

17.

Watelet

Gallot

Thomas

, et al. Percutaneous repair of aortic aneurysms: a prospective study of suture-mediated closure devices. Eur J Vasc Endovasc Surg 2006; 32: 261–265.

18.

Can

Torsello

Umscheid

The arterial approaches for endovascular aortic grafting. In: A

Branchereau

Jacobs

(eds) Endovascular aortic repair: the state of the art. Oxford: Blackwell Publishing, 2008, pp.57--61.

19.

Traul

Clair

Gray

, et al. Percutaneous endovascular repair of infrarenal abdominal aortic aneurysms: a feasibility study. J Vasc Surg 2000; 32: 770–776.

20.

Rachel

Bergamini

Kinney

, et al. Percutaneous endovascular abdominal aortic aneurysm repair. Ann Vasc Surg 2002; 16: 43–49.

21.

Giordano

Corcione

Ferraro

, et al. Comparison of ProGlide vs. prostar in patients undergoing transcatheter aortic valve implantation. Minerva Cardioangiol 2019; 67: 443–449.

22.

Zakko

Scali

Beck

, et al. Percutaneous thoracic endovascular aortic repair is not contraindicated in obese patients. J Vasc Surg 2014; 60: 921–928.

23.

Pecoraro

Krishnaswamy

Steuer

, et al. Predilation technique with balloon angioplasty to facilitate percutaneous groin access of large size sheath through scar tissue. Vascular 2017; 25: 396–401.

24.

Malkawi

Hinchliffe

Holt

, et al. Percutaneous access for endovascular aneurysm repair: a systematic review. Eur J Vasc Endovasc Surg 2010; 39: 676–682.