Comparison of Two Vascular Closure Device Strategies for Transfemoral Transcatheter Aortic Valve Replacement Using Suture-Based and Collagen-Plug-Based Techniques and Associated Vascular Complications

© 2024 HMP Global. All Rights Reserved.
Any views and opinions expressed are those of the author(s) and/or participants and do not necessarily reflect the views, policy, or position of the Journal of Invasive Cardiology or HMP Global, their employees, and affiliates

J INVASIVE CARDIOL 2024. doi:10.25270/jic/24.00251. Epub October 23, 2024.

Objectives. In transfemoral (TF) transcatheter aortic valve replacement (TAVR), vascular closure devices (VCD) with suture-based vs collagen-plug-based techniques are commonly used to achieve postprocedural hemostasis. This study evaluated the differences in vascular complications between 2 VCD strategies.

Methods. This retrospective study included consecutive patients at the George Washington University Hospital who underwent TF TAVR between May 2015 and December 2019 (n = 267). The VCDs utilized were Perclose ProGlide (PP; Abbott) and Angio-Seal (ASe; Terumo). Patients were divided into 2 groups: (1) those who received 1 PP plus 1 ASe (1 PP + 1 ASe; n = 145), and (2) those who received more than 1 PP plus 1 ASe (> 1 PP + 1 ASe; n = 122). Chi-square and t tests were used to examine differences in baseline characteristics and rates of vascular complications. 

Results. The mean age of the patients was 80 ± 9 years, 70.0% were male, the mean Society of Thoracic Surgeons score was 4.6 ± 3.8, and 18.0% had peripheral vascular disease. Patients who received 1 PP + 1 ASe were associated with fewer vascular complications compared with those who received > 1 PP + 1 ASe (5.5% vs 13.1%; P = .03). Bleeding complications were seen in 4.8% of the 1 PP + 1 ASe cohort compared with 8.2% of the > 1 PP + 1 ASe cohort (NS; P = .26). Occlusion/stenosis complications were seen in 0.7% of the 1 PP + 1 ASe cohort compared with 4.1% of the > 1 PP + 1 ASe cohort (NS; P = .06).

Conclusions: For TF TAVR, a VCD strategy using 1 PP + 1 ASe was associated with significant reduction in total vascular complications compared with the use of more VCDs.

Transcatheter aortic valve replacement (TAVR) procedures have grown significantly in the United States since their inception and, since 2013, are now more frequently performed than surgical aortic valve replacement (SAVR).¹ Guidelines now recommended TAVR for severe aortic stenosis (AS) in patients with low to high surgical risk.^2-8

Preoperative TAVR workup and procedure techniques have also evolved over time. Valve sizing has been optimized with the use of computed tomography, and procedures are more often preformed under conscious sedation with less use of general anesthesia. Improved technology with smaller sheath sizes has led to a greater than 95% use of the percutaneous transfemoral approach.⁹ Despite these advances, vascular complications during TAVR remain one of the most common procedural complications. While the rates of vascular complications are as low as 2% to 4%, it has been shown that up to a quarter of vascular access site complications may be related to the secondary arterial access.^5,6,10

In addition to the primary cannulation site for valve access, secondary arterial access for TAVR is obtained either via the contralateral femoral artery or the radial artery. Secondary arterial access has many purposes: femoral angiography can guide primary femoral puncture access; use of a pigtail catheter in the ascending aorta can mark the annular plane for aortography to guide valve position, deployment, and assess for potential procedural complications related to coronary anatomy; and peripheral angiography can identify vascular complications at the end of the case and be used to perform necessary peripheral interventions.^11-15

Few studies have specifically compared the effectiveness of different vascular closure device (VCD) strategies,^16-20 however, our program experience has provided us with a unique opportunity to fill this gap in the literature. Early in our experience, the default strategy for vascular closure of the TAVR delivery site included the placement of 2 PP plus 1 ASe. Later on, after February 2018, we changed the closure strategy to placement of 1 PP plus 1 ASe. Currently, our closure strategy for the TAVR access site is to routinely use 1 PP and 1 ASe unless there is significant patient obesity, in which case it is up to the discretion of the implanter to use a second PP upfront in case the first PP fails to deploy. This change in strategy allows for a comparison of these 2 approaches. Thus, the purpose of this study is to evaluate our experience with 2 distinct VCD approaches using the PP and the ASe, as well as the impact on vascular access site complications during TAVR.

Patient population

This is a retrospective, observational evaluation of prospectively collected baseline, procedural and post-procedural data stored in our institution’s TAVR database. All patients at the George Washington University Hospital who underwent transfemoral TAVR from May 2015 through December 2019 were included in this retrospective study (n = 267). The VCDs utilized were the Perclose ProGlide (PP) suture-mediated closure device (Abbott) and the Angio-Seal (ASe) device (Terumo). Patients were divided into 2 groups: (1) those who received 1 PP plus 1 ASe (1 PP + 1 ASe; n = 145), and (2) those who received more than 1 PP plus 1 ASe (> 1 PP + 1 ASe; n = 122).  Exclusion criteria were procedural deaths unrelated to vascular access.

The internal institutional review board of the George Washington University Hospital approved this study, and waiver of informed consent was obtained due to the retrospective, deidentified nature of the study. The eligibility of the patients to undergo TAVR after appropriate screening was based on standard clinical evaluation and the consensus of our comprehensive Heart Team.

Outcomes and definitions

The primary endpoint of the study was to compare immediate perioperative vascular access complications, as defined by the Valve Academic Research Consortium (VARC)-2 initiative, between the 2 VCD groups.²¹ These complications included both major and minor vascular complications as defined by (VARC)-2. Vascular complications were divided into 2 major groups, bleeding or stenosis/occlusion, and were defined as any complication that required further vascular intervention.

Procedure

The TAVR procedures were conducted either in the hybrid or regular catheterization suite of the George Washington University Hospital. Cases were routinely performed under general anesthesia or conscious sedation with transthoracic or transesophageal guidance as per operator discretion.

Primary TF access for delivery sheath insertion was initially obtained under fluoroscopic guidance and over time under routine ultrasound and fluoroscopic guidance with initial placement of a 6-French (Fr) Glidesheath (Terumo). The TF access was subsequently upsized to a 14-Fr, 16-Fr, or 18-Fr delivery sheath per conventional TAVR procedure after upfront placement of either 1 or 2 PP via a pre-close technique. Delivery sheaths were sized and placed according to the type of valve delivery system utilized, which included the Edwards XT, Edwards Sapien, Edwards Sapien Ultra (Edwards Lifesciences), and the CoreValve (Medtronic). From May 2015 until February 2018, the standard approach was to use 2 PP via a pre-close technique plus 1 ASe. After February 2018, we primarily used 1 PP via a pre-close technique, unless 2 PP were deemed necessary per operator discretion, plus 1 ASe. Secondary access was obtained in the radial or contralateral femoral artery using a 6-Fr Glidesheath (Terumo). Intraprocedural unfractionated heparin was administered in all study procedures, and the use of protamine to reverse anticoagulation at closure was left to the discretion of the operator. Closure of the primary TF access site was subsequently achieved with either 1 or 2 PPs, and a single ASe was added if deemed necessary by the operator. In patients with secondary contralateral TF access, an additional ASe was used for closure per operator discretion.

Statistical analysis

Descriptive statistics analysis (means and SDs for continuous variables and frequency tables for categorical variables) was conducted for the baseline demographic and health characteristics. The chi-square test and Student’s t test were used to test for differences between the 2 groups (those who received 1 PP + 1 ASe vs > 1 PP + 1 ASe). The differences in the clinical outcomes, including all vascular complications, bleeding, and occlusions, were examined between the 2 groups using the chi-square test. The statistical analysis was performed using SAS version 9.4 (SAS). All tests were conducted using an α = 0.05 as the probability for a type I error.

A total of 267 consecutive patients underwent TF TAVR at the George Washington University from May 2015 through December 2019. The 2 groups compared were patients who received 1 PP + 1 ASe (n = 145) vs > 1 PP + 1 ASe (n = 122). Baseline patient demographics and characteristics are demonstrated in Table 1. The mean age of the patients was 80 ± 9 years, 70.0% were male, the mean Society of Thoracic Surgeons (STS) score was 4.6 ± 3.8, and 18.0% had a history of peripheral vascular disease (PAD). It should be noted that patients in the > 1 PP + 1 ASe group had higher baseline rates of PAD compared with the 1 PP + 1 ASe group (25.6% vs 12.4%, P = .01), and a higher STS score (5.5% vs 3.8%, P < .01). Also, the patients in both groups had similar baseline BMI, hemoglobin levels, platelet counts, and international normalized ratio levels.

Patients who received 1 PP + 1 ASe were associated with a lower rate of vascular complications compared with those who received > 1 PP + 1 ASe (5.5% vs 13.1%; P = .03; Table 2). Bleeding complications were seen in 4.8% of the patients in the 1 PP + 1 ASe cohort compared with 8.2% of the patients in the > 1 PP + 1 ASe cohort (NS; P = .26; Table 2). Occlusion or stenosis complications were seen in 0.7% of patients in the 1 PP + 1 ASe cohort compared with 4.9% in the > 1 PP + 1 ASe cohort (NS; P = .77; Table 2).  BMI did not correlate with a statistically significant difference in vascular complications between the 2 groups (28.5 vs 30.0 NS; P = .13).

Several different vascular closure strategies have been adopted for large bore arteriotomy after TF TAVR and to close large access sheath sites after other interventions such as percutaneous ventricular assist devices. These approaches include surgical cut-down with arterial access obtained under direct visualization and the use of various percutaneous vascular closure devices (VCDs). Studies have shown the advantages of percutaneous VCDs over surgical closure, including decreased complication rate, patient morbidity, and procedural time.^22,23While several closure strategies are available, underlying patient characteristics, anatomy, operator expertise, and financial aspects are pivotal in choosing the optimal approach.

The VCDs most commonly utilized are either suture-based, delivered in “pre-closure” fashion, or collagen-mediated.¹⁷ The Prostar XL or ProGlide (Abbott) are 2 commonly used suture-based closure devices often placed using a pre-closure technique.¹⁸In 1 study looking at TAVR closure, ProGlide was associated with lower rates of major vascular complications when compared with Prostar XL.¹⁹ Collagen-based devices include the Angio-Seal (AS) device for smaller arteriotomies, which is among the most widely used VCDs, and the novel percutaneous MANTA VCD (Essential Medical), which uses technology dedicated to the closure of large bore arteriotomies.¹⁷ One study by Gheorghe et al evaluated vascular and bleeding complications in TAVR patients comparing the MANTA VCD with the Prostar XL VCD and found that the MANTA had a greater successful closure rate compared with the Prostar XL, and was associated with similar major vascular and bleeding complications.²⁰ Biancari et al found no significant difference between the MANTA and ProGlide in terms of bleeding or major vascular complications in TAVR patients.¹⁶

The Perclose ProGlide suture-mediated closure device has been widely utilized in a dual pre-close strategy as well as adjunctively with the Angio-Seal (AS) device to augment the dual PP pre-close approach in achieving improved hemostasis. However, vascular complications are reported in 4% to 9% of cases that utilize a VCD.^11-15,24,25 The potential advantage of achieving hemostasis with the use of VCD can often be offset by adverse events such as stenosis, limb ischemia, pseudoaneurysm, thrombosis, and infection.^12,26

While a more standardized approach exists in the current valvular heart disease guidelines and within TAVR-performing institutions in terms of patient pre-TAVR assessment, there is more diversity and less guidance when it comes to technical procedural aspects, including vascular closure strategies.⁵ In addition, different approaches may change over time depending on operator expertise with the procedure. Early in our program experience, the default strategy for vascular closure of the TF TAVR delivery site included the placement of 2 PP plus 1 ASe. Later, after approximately one-third of the total cases analyzed in this study were performed, we modified our approach and adopted a placement strategy of 1 PP plus 1 ASe; this change allowed for a comparison of these 2 approaches. Currently, our standard closure strategy for the TAVR access site is to use 1 PP plus 1 ASe unless the patient has significant obesity. It has been shown in a previous study that obesity is an independent risk factor in achieving hemostasis when using VCDs thought to be related to difficulty in deliverability or kinking of the VCD.²⁷ For any morbidly obese patient at our institution with a large pannus at the access site, it would remain at the discretion of the implanter to use a second PP upfront if they believe it will assure effective hemostasis after closure and reduce bleeding in case the first PP does not deploy.

The baseline demographics between the 2 groups in our study were largely similar, however, it is notable that there was a higher incidence of baseline history of PAD as well as increased STS score in the group using > 1 PP + 1 ASe (Table 1). While we postulate that the total number of VCDs used contributed to the differences in vascular complications, we cannot fully exclude the possibility that some of these underlying confounding variables may also have contributed towards the difference. Indeed, the group who received > 1 PP + 1 ASe may have had worse underlying prognostic pathologies and, subsequently, were more likely to develop access site complications independent of the type of VCD utilized, specifically related to an increase in the presence of PAD. Furthermore, there was no difference in baseline BMI between the 2 groups (Table 1). While we tend to place a second PP upfront in morbidly obese patients, BMI was not a cause of any statistically significant difference in vascular complications between the 2 groups (Table 2).

It is also important to highlight that, at our institution, access is now routinely obtained using ultrasound (US) guidance. Several studies have shown a reduction in access site complications when US is used in TAVR and other large bore arteriotomy procedures. The use of US allows for direct identification of vascular anatomy and calcium-free puncture sites, as well as visualization of the needle puncture.^28-30

The main results of our study suggest that, for transfemoral TAVR, a VCD strategy reducing the overall number of VCD may minimize the rate of vascular access complications. We found that the use of 1 PP plus 1 ASe was associated with a significant reduction in total vascular complications compared with the use of more VCDs and may further limit complications related to occlusion or stenosis with no increased risk of bleeding.

In general, patients undergoing TAVR tend be older with underlying co-morbidities that may include the presence of PAD as well as higher degrees of iliofemoral calcification, increased vessel stiffness and tortuosity, and smaller caliber femoral arteries. Because of our early institutional experience adopting a standardized strategy of using 2 PP plus 1 ASe, we noticed an increased rate of complications related to vascular stenosis and limb ischemia. Given the underlying characteristics of the peripheral vasculature in this patient population, we modified our closure strategy to adopt the use of fewer VCDs in an attempt to offset these observed complications. As demonstrated by our results, we found that a strategy using 1 PP plus 1 ASe resulted in better outcomes without compromising adequate hemostasis. While no single strategy will be best for every patient, it is important to consider the patient-specific anatomic characteristics when choosing the appropriate VCD strategy.

Limitations

We conducted a single institution retrospective study comparing 2 vascular closure strategies, which has the potential for bias given its design. Operator preference, patient anatomy, use of procedural anticoagulation with heparin with a therapeutic range of activated clotting time, and various usage of protamine at the end of the procedure all have the potential to be confounding factors. Another limitation of this study is that it did not examine VCD selection or complications associated with access sheath size, which may offer further insights into the results.

Our study suggests that, for TF TAVR, a VCD strategy using 1 PP plus 1 ASe is associated with a significant reduction in total vascular complications compared with the use of more VCDs. Prospective randomized controlled trials may be useful in better addressing an optimal VCD strategy.

Raza Yunus, MD¹; Talal Alzahrani, MD, MPH¹; Roy Taoutel, MD¹; Jonathan Reiner, MD¹; Ramesh Mazhari, MD¹; Farzad Najam, MD²; Gregory Trachiotis, MD²; Joseph Krepp, MD¹; Andrew Choi, MD¹; Shawn Sarin, MD³; Christian D. Nagy, MD¹

From the ¹Division of Cardiology, Medical Faculty Associates, The George Washington University, Washington, DC; ²Department of Cardiothoracic Surgery, The George Washington University, Washington, DC; ³Department of Interventional Radiology, Medical Faculty Associates, The George Washington University, Washington, DC.

Disclosures: The authors report no financial relationships or conflicts of interest regarding the content herein.

Acknowledgments: The authors extend a special thank you to Eduard Shaykhinurov for his data stewardship.

This abstract was presented as a poster at Cardiovascular Research Technologies (CRT) in 2021.

Address for correspondence: Christian D. Nagy, MD, Medical Faculty Associates, The George Washington University, 2150 Pennsylvania Ave NW, Suite 4-417, Washington, DC 20037, USA. Email: cnagy@mfa.gwu.edu; X: @roytaoutel