Manual compression has been the gold standard for obtaining haemostasis at the femoral artery access site since being introduced in conjunction with the Seldinger technique in the 1950s.1 However, the limitations of manual compression, including patient discomfort, delayed haemostasis and ambulation and bleeding at the access site despite satisfactory initial haemostasis, have spurred the development of alternative closure methodologies. Vascular closure devices (VCDs) were developed in the 1990s to address these limitations in response to the aggressive anticoagulation strategies initially required for intracoronary stents and use of atherectomy devices requiring large-bore sheaths. The initial devices included the bioabsorbable anchor–collagen sponge arteriotomy sandwich Angio-Seal, the suture-based Perclose, the extravascular collagen VasoSeal and the extravascular thrombin– collagen Duett. The first generations of these devices were a challenge to use and were associated with an unacceptably high rate of closure failure. Through multiple design modifications, in particular of the Perclose and Angio-Seal devices, ease of use improved markedly and closure failure rates fell considerably, with very high rates of successful in-laboratory closure being obtained. Interestingly, practitioners chose the Angio-Seal and Perclose as their VCDs of choice, as reflected by market utilisation data, perhaps because of the perception that the extravascular devices Duett and Vasoseal were not as effective as these other devices. Nonetheless, interest in extravascular closure devices has re-surfaced, and the two latest entrants into the closure market (Mynx and Starclose) are based on extravascular closure methodologies.
Despite these improvements in the devices, utilisation has not paralleled the adoption seen in other devices in interventional cardiology, such as drug-eluting stents. VCDs continue to be used in a minority of patients and utilisation has hovered around 20–40% over the past 10 years. This likely reflects concerns about a perceived lack of data demonstrating significant reductions in complications with VCD use, limited applicability in a number of patients because of anatomical contraindications to their use and increased costs without adequate reimbursement for their use. With an increased emphasis on reducing bleeding after procedures performed from the femoral access site, the use of VCDs is coming under greater scrutiny as an appropriate strategy to achieve this goal.
Evidence of Vascular Closure Devices versus Manual Compression
Concerns over increased rates of vascular complications from single-centre registries in the late 1990s prompted closer scrutiny of outcomes after VCD use.2 Two meta-analyses were published examining rates of vascular complications from VCDs versus manual compression in both randomised trials and registries up to 2001, and did not find clear evidence of a benefit of VCDs.2,3 These studies led to the widespread clinical perception that VCDs did not reduce vascular complications and clearly hampered expanded utilisation of these devices, despite a report from the American College of Cardiology National Cardiovascular Data Registry (ACC/NCDR) in 2001 showing that the same devices were associated with significantly lower rates of vascular complications compared with manual compression.4 Unfortunately, to date, no large randomised clinical trials have been performed or planned with newer devices.
Nonetheless, two recent large single-centre studies from Wake Forest5 and the Brigham and Women’s Hospital6 and a post hoc evaluation from the Acute Catheterization and Urgent Intervention Triage strategY (ACUITY) trial7 observed that the current generation of VCDs reduced rates of vascular complications compared with manual compression by 30–50%. Taken together, these observations provide strong evidence (in the absence of a definitive randomised clinical trial) that newer generations of VCD reduce vascular complications. Whether this will provide compelling reasons for practising physicians to reconsider their choice of vascular closure technique remains to be determined.
Vascular Closure Devices
The Workhorses – Angio-Seal and Perclose
The Angio-Seal closure device (St Jude Medical, St Paul, MN, US) has undergone extensive modification since its commercial availability in 1996. The platform was recently updated with the introduction of automated compaction of the collagen sponge–anchor closure complex (Evolution). The bioabsorbable closure elements remain the same: an intravascular polylactic-glycolic acid (PGA) anchor attached to a bovine collagen sponge located on the surface of the artery by an absorbable suture. In benchtop testing, the compaction forces were more uniform with the Evolution device and automated compaction than with traditional hand compaction of the VIP device.
The clinical utility of the Evolution device has been examined in a prospective registry in 1,003 patients from 10 sites experienced in use of VCDs.8 Device deployment success was 99% and the overall rate of vascular complications was 2%, with only a 0.7% rate of major vascular complications in the subset of percutaneous coronary intervention (PCI) patients. While these data need to be confirmed and compared with manual compression in a larger number of patients, they demonstrate the continued efficacy and safety of the Angio-Seal device.
The suture-based Perclose device has also undergone extensive modifications since its commercial availability in 1994. Currently, the Perclose device is used in three ways: for primary closure of up to 8–9Fr sheaths with the ProGlide platform, for primary closure of up to 10Fr sheaths with the Prostar platform and for pre-closure with both platforms for sheaths ≤24Fr. The introduction of percutaneous aortic valve replacements performed from femoral artery access sites and haemodynamic support devices such as Impella have created a resurgence of interest in suture-based technologies such as Perclose to allow closure of these very-large-bore sheaths. It is clear that there is a substantial need for devices specifically designed to provide closure for these large-bore sheaths.
New Devices – Mynx and StarClose
The Mynx Vascular Closure Device (AccessClosure, Inc., Mountain View, CA, US) is a relatively new device that utilises extravascular polyethylene glycol (PEG) sealant in the tissue tract above the arteriotomy site to achieve haemostasis. The absence of an intra-arterial component and the ability to use this device with the existing procedural sheath have made it an appealing option. Although this device will require larger studies, it may offer an advantage in patients with peripheral vascular disease due to its lack of an intraluminal component.
The StarClose SE (Abbott Laboratories, Abbott Park, IL, US) utilises nitinol clip technology to create purse-string-type closure on top of the arteriotomy site without an intra-arterial component. In its current form, the device requires dilation of the tissue tract to allow introduction of the 11Fr delivery shaft to the arteriotomy site. Because of its extravascular closure methodology, it has been used in anatomical sites traditionally not felt to be suitable for closure, such as the femoral bifurcation, and in the presence of mild to moderate atherosclerotic disease at the access site.9,10 Further studies will need to confirm the safety and efficacy of this expanded use of the StarClose device.
Emerging and Novel Devices
New and novel VCDs continue to be developed. In particular, there are three devices that merit mention. The ExoSeal® device (Cordis, Miami, FL, US) is an extravascular passive closure device that delivers a PGA ‘felt-like’ plug to the arteriotomy site, anchored by the neurovascular bundle sheath.11 The PGA plug undergoes hydrolysis in the body and is degraded over three months. A pivotal US Food and Drug Administration (FDA) trial has been performed that demonstrated satisfactory efficacy and safety, and the product awaits FDA approval (see Table 1). The Femoral Introducer Sheath and Hemostasis (FISH; Morris Innovative, Bloomington, IN, US) incorporates a sleeve of small intestinal submucosa into the introducer sheath, which acts as a haemostatic plug at the arteriotomy site during sheath removal. Only limited clinical data are available with this device. The Arstasis device introduces a novel concept of access-site closure by creating a long, shallow access tract within the arterial wall, which creates haemostasis when the sheath is removed with limited manual compression.12 Clinical studies are currently under way to determine whether this novel concept can be translated into effective haemostasis in routine practice.
Device Strategies in 2010
The decision to use a closure device and which closure device to use is influenced by several factors, including the experience and confidence of the operator in device use. First, regardless of the desire to use a closure device, appropriate anatomy at the access site is essential for optimal success with VCDs. This includes access within the common femoral artery (CFA), vessel diameter at least >4mm (device-dependent) and the absence of significant local atherosclerotic disease. Second, the sheath size and procedure type used for the procedure often dictate whether a VCD is desirable. In our laboratory, diagnostic catheterisation is routinely performed with 5Fr catheters without heparin and ambulation is allowed within two hours, so we elect not to use VCDs routinely in these patients. For any patient who receives anticoagulation for a procedure such as intravascular ultrasound (IVUS) or fractional flow reserve (FFR), or undergoes PCI or other intervention, we will routinely opt for VCD over manual compression. In these patients, the choice of VCD is individualised by each operator.
In our laboratory, the workhorses Angio-Seal and Perclose are preferred for use because of their excellent track record. For routine interventional cases, Mynx and Starclose would be acceptable alternatives. For procedures performed with sheaths >10Fr, we will pre-close with the Perclose ProGlide device (one or two devices). When substantial obesity and scar tissue are detectable at the access site, we will use Angio-Seal preferentially, as successful device deployment can be obtained routinely. For very thin patients, VCDs are problematic as there is little room for extravascular collagen or sealant, and Perclose is preferred. Starclose would be an alternative. In patients in whom the access site is below the CFA bifurcation or atherosclerotic disease is present, an extravascular sealant such as Mynx would be preferred, although there are reports of successful Starclose use in these patients.9,10 Finally, in those patients with a high stick (i.e. above the inferior epigastric artery and into the external iliac artery), VCD use is problematic.13 We have migrated to a strategy of manual compression when the anticoagulation has worn off in these patients, although optimal closure management of these patients remains to be determined.
Conclusions
In 2010, VCDs provide efficacious and safe haemostasis of femoral artery access sites. While the perception is that VCDs do not reduce the rate of vascular complications compared with manual compression, recent large, single-centre studies and a post hoc analysis of the ACUITY trial strongly suggest that current devices are safer than simple manual compression. VCDs remain underutilised for access closure, and expanding their use remains a clinical challenge. Continued development of VCDs that reduce complications, especially bleeding, remains a high priority and an opportunity for new devices.