Review
An overview of vascular closure devices: What every radiologist should know

https://doi.org/10.1016/j.ejrad.2008.09.023Get rights and content

Abstract

Haemostatic devices can be categorised according to their mechanism of action into three main types; namely pressure devices, topical haemostatic pads and vascular closure devices (VCD). Of these three categories, it is the development of VCDs that revolutionised management of endovascular procedures. Currently available VCDs fall into three major classes, those that use a collagen plug, those that use clips and those that perform suture closure at the arteriotomy site. This article provides a comprehensive review of the all three classes with examples of commercially available devices.

Introduction

It is remarkable that almost half of century since the introduction of Seldinger’s technique, manual compression remains the gold standard for achieving haemostasis after gaining vascular access for diagnostic and interventional arterial catheterisation. Increasing time pressure, issues of patient satisfaction and the use of large diameter catheters for endovascular intervention however have recently necessitated alternative ways of establishing post endovascular procedure haemostasis.

From the early 1990s, various devices have been used to achieve haemostasis post arterial puncture. These haemostatic devices can be categorised according to their mechanism of action into three main types; namely pressure devices (Femostop®, RADI medical systems), topical haemostatic pads and vascular closure devices (VCD). Of these three categories, it is the development of VCDs that revolutionised management of endovascular procedures.

Ideally, a VCD should be user friendly, provide rapid, reliable haemostasis regardless of anticoagulation status and also allow repeated access. The complication rates of such devices should be no greater than that of manual compression. Currently available VCDs fall into three major classes, those that use a collagen plug, those that use clips and those that perform suture closure at the arteriotomy site (Table 1). This article provides a comprehensive review of the all three classes with examples of commercially available devices.

Primary vascular haemostasis after arterial puncture is facilitated by blood contact with the exposed arterial wall smooth muscle cells and collagen. This in turn causes platelets adherence, activation and aggregation resulting in clot formation. Bovine collagen used in these devices augment haemostasis by increasing the availability of collagen at the arterial wall defect [1]. Furthermore the swelling of the collagen mass which occurs after deployment completes haemostasis by mechanically sealing the vessel and tissue tract. Bovine collagen is eventually degraded by macrophages and is reabsorbed within 4–6 weeks time.

The main disadvantage of collagen based products is that immediate re-puncture is not advised due to potential increased risk of local infection and the required time period for collagen biodegradation. In addition, localised proliferative reaction resulting in scar formation caused by the bovine collagen hinders future open surgical access [2]. However, in the last few years, this been refuted by a study suggesting that re-accessing is safe in post Angio-Seal® deployment [3].

In general, collagen plug devices are used in patients who are unlikely to require repeated access either immediately or within 90 days after the initial procedure, though in practice if re-access is required, ultrasound can be used to guide re-puncture remote to such devices. Examples of commercially available collagen based vascular devices include: Angio-Seal®, VasoSeal® and Duett Pro® sealing device.

The Angio-Seal® device consists of a rectangular 1 mm × 2 mm × 10 mm co-polymer footplate attached to a 18 mg bovine collagen plug by an absorbable Dexon traction suture. It produces a sandwich closure of the arteriotomy site with a foot-plate that sits against the inner arterial wall intraluminally, and a collagen plug that is tied to lie immediately outside the arteriotomy.

Because of its simple design and easy deployment, the Angio-Seal® makes up more than half of the VCD market in the USA.

The newer versions of Angio-Seal® (Angio-Seal STS Platform®) ensure a smooth transition from sheath tip to dilator, and have repositioned blood inlet holes to facilitate arterial location. Compared with the previous models, this has been shown provide faster and easier deployment along with a similar success rate and incidence of major complication [4]. Angio-Seal® has also been used in other vascular sites such as the aorta and carotid arteries [5], [6]. It has also been used to seal inadvertent subclavian artery puncture during central venous catheter placement [7].

This device should not be used in patients due to have a femoral cut down, as there is an inadvertent risk of cutting through the sutures and causing embolisation of the footplate.

The VasoSeal® device is another example of a collagen based vascular closure device. Unlike the Angio-seal®, these devices have no intraluminal components. This particular feature allows it to be used in patients with peripheral vascular disease without obtaining an angiogram. The collagen plug is mainly deposited along the arterial puncture tract. There are several types of VasoSeal® product available, including the VasoSeal ES® and Vasoseal Elite®.

Early VasoSeal® devices (VasoSeal VHD®) were made up of a needle measuring kit, an 11F blunt dilator on an 11.5F sheath and an 11F cartridge loaded with 80–100 mg of purified bovine collagen. Measurement of the skin-to-vessel distance with the needle measuring kit enabled selection of the appropriate size of the device, which in turn determined whether one or two collagen plugs were to be used. Once the device is deployed manual compression for few minutes is also recommended to achieve adequate haemostasis. Cumbersome step of measuring skin to vessel distance along with the requirement of large introducer sheath and manual compression at the final stage made the initial version of VasoSeal® less popular.

Newer versions of VasoSeal® such as the VasoSeal ES®, VasoSeal Elite and VasoSeal-low however do not require the initial skin-to-vessel distance measurement or manual compression. These devices also reduce the collagen required by 40% and use more acceptable 4–8.5F delivery system. Furthermore use of a rapidly expanding newer version of collagen plug results in quick and effective haemostasis without manual compression.

Duett Pro®sealing device, unlike the Angio-Seal® and VasoSeal® can be directly inserted though any introducer sheath used for endovascular procedures. The Duett Pro®device consists of a 3F balloon-positioning catheter and a pro-coagulant mixture of bovine collagen (250 mg) and thrombin (10,000 units) (Fig. 5). The balloon is used as a temporary intraluminal anchor, and used as shown in Fig. 6. Conversion of fibrinogen into fibrin by the action of thrombin in the presence of collagen results in an accelerated coagulation cascade.

Due to the absence of any intra-luminal biodegradable material, vascular re-access if required is probably safer. The major potential complication of this device is that of accidental injection of the pro-coagulant mixture into the artery [8]. This has resulted in the introduction of a new VCD called the Matrix VSG (AccessClosure Inc., Mountain View, CA). Matrix VSG uses a water soluble material, polyethylene glycol, as the pro-coagulant. This material is easily dispersible if injected accidentally into the vascular lumen.

These devices close the arteriotomy site by deploying usually a pair of needles accurately, to pass a suture through the artery wall. The knot is then tied and slid to the arteriotomy site through the tissue tract. Closure is secure almost immediately, and re-access is easier, as there is no collagen plug or intravascular anchor to consider. Current examples include Perclose® (Abbott Vascular), X-Site® (Datascope) and SuperStitch® (Sutura).

This was one of the first suture-mediated closure devices available and takes up to 40% of the VCD market in the USA. This device uses a single suture (for 5–8F arteriotomies) for closure. There is no residual intraluminal component. The initial version of Perclose® called Prostar required a large tissue tract and a relatively complex series of manoeuvres, to launch the needles, capture the sutures and tie knots.

In 2002, Perclose AT® was introduced to simplify the knot tying step. It has a pre-tied knot that obviates the need for manual knot tying, allowing rapid suture deployment. The newer recent Proglide model uses a polypropylene monofilament suture in place of the polyster braided suture. The monofilament suture increases knotted tensile strength and allows easier knot delivery due to its slippery nature. In addition, less inflammatory response has been noted compared with the previously used braided suture. One of the main attractive features of the Perclose® device is that it can be used to close larger arteriotomy holes up to 26F in diameter (e.g., during placement of aortic stent grafts and intra aortic balloon pump insertion), by using the ‘preclose’ technique [9]. The main criticism of this device is the deployment steps are complex.

This is a relatively new suture-based device that is intended to close 6F arteriotomy holes.

The X-site device is composed of a sheath that is mounted to the distal end of a hub and an integrated needle pusher/cutter. The hub incorporates a slit that allows needles attached to a single strand of suture to be manually advanced across a target tissue region and then back into the device. As the device is withdrawn, a suture loop across the arteriotomy site is formed. A clinch knot is prepared and then pushed in using the knot pusher. Unlike Perclose®, the X-site has no intra arterial footplate which may decrease the possibility of any luminal trauma.

This device uses a monofilament polypropylene suture to achieve access site closure. It is used with conventional access sheaths and is advanced without a guidewire. The device comes in 6, 8 and 12F sizes. The integral tip design allows use in antegrade procedures easily.

In this category are the EVS Vascular Closure System® (Medtronic) and StarClose® (Abbott Vascular). Both are metal based extra-luminal devices which remain in situ post deployment. These are fairly new to the market compared to the rest of the devices.

The Angiolink EVS® obtains extra-luminal percutaneous mechanical closure without residue along the tissue tract. It consists of an introducer/dilator that contains a biocompatible titanium staple and a trigger-activated deployment system. A ‘purse-string’ suture closure concept, where a pair of vessel stabilisers are first deployed to oppose the vessel wall, prior to deployment of a staple is used in establishing haemostasis. Staple penetration is limited to the adventitial and medial layers resulting in a sole extra-luminal closure with no residual intraluminal components.

Multicentre trials has proven the efficacy of this device [24], [25], with no increase in major complications compared to manual compression. Deployment technique was simple and it allowed a short learning curve. In addition, there is minimal biologic response to the staple due to its inert properties.

The StarClose® closure device delivers an extra-vascular nitinol clip to close up to 8F arteriotomy holes. The device consists of a clip applier, 6F exchange sheath, 6F dilator and a J-tip guidewire. An implantable clip is mounted on to the clip applier. The device has been demonstrated to be safe with 3Tesla MR equipment. The safety of repuncture at anytime through any part of the clip, and subsequent closure of this repuncture with another Starclose® device has not been fully established. A caveat of this device is that steep punctures tend to cause shredding of the central shaft, causing the device to get stuck in the patient.

In general, the deployment is simple and quick, with an average time of 48 s. Deployment success rate was about 96%, and there were no major complications quoted in the latest prospective study. Again, the device has the advantage of being inert, and induces less biologic response compared to collagen plugs [26].

This device has comparable safety and efficacy compared to AngioSeal® and manual compression. It however, requires additional manual compression to achieve haemostasis after successful deployment.

According to initial studies in the late 90s, vascular complications were twice as common compared with manual compression [30]. Frequency of surgical repair of arterial damage, formation of haematoma and drop in haematocrit were noted to be higher in the VCD cases [30]. Operator unfamiliarity contributed mainly to the initial increase in complication rates. Meta-analysis findings are continued to be flawed by heterogeneous patient groups, lack of inclusion of physicians’ learning curves, absence of good quality randomised controlled trials and inclusion of early forms of VCDs [31]. Findings such as increased haematoma and pseudoaneurysm formation with the use of VCD as well as worsening outcome with the use of Vasoseal revealed by published meta-analyses must therefore be interpreted with caution [31].

Improvements in device design as well increase in experience in use VCDs meant that more recently reported data showing improved results, with studies analysing individual VCDs demonstrating reduced or absent vascular complications between the device and manual compression arms [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33]. With the use of newer devices, complications related VCD use are generally similar to those observed in patients managed with manual compression with the exception of device embolization [33]. Pseudoaneurysm remains the most frequent access site related complication, followed by bleeding, arteriovenous fistula, arterial thrombosis and infection [33].

Recently several studies have been conducted to evaluate the safety and efficacy of VCD in the setting using anticoagulation. Use vascular closure devices such as Agioseal, Vasoseal, Perclose and Duett showed equal or less complication risk in patients on antiplatelet therapy compared with manual compression [16], [34], [35]. Similar finding was also seen in patients who on Warfarin or Abciximab [35], [36]. However certain rare complications such as retroperitoneal haemorrhage and severe access site infection may be more common with the use of VCD in patients on anticoagulation [37].

An accurate analysis of complication rates between different VCDs is however yet to be performed. Such analysis is difficult due to high levels of heterogeneity between studies, but in general there seem to be no significant differences in complication rates between different types of VCD in the setting of diagnostic and interventional procedures [9], [10], [31], [32].

The pattern of complications may however vary between different class of device. For example, stenosis or occlusion at the level of the puncture site and femoral neuralgic syndrome caused by irritation of the anterior cutanoeus nerve have been described more frequently with the use of suture mediated closure devices [38]. Late femoral artery thrombosis at the site of the anchor has been reported with the use of the Angio-Seal® device, and embolisation of the collagen plug has been reported with the Angio-Seal® and VasoSeal® devices [36], [37]. There have been reports of early arterial thrombosis following use of the Duett® device [39], [40]. Also infection has been reported with both the suture and plug mediated devices [37], [41], [42], [43], [45].

The majority of trials comparing manual compression with VCD show that use of VCDs decrease the time to haemostasis and the time to ambulation, and thereby, increase patient comfort and improve catheterisation efficiency [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33]. Newer vascular closure devices have shown to reduce both major and minor vascular complication rates significantly compared to manual compression [4], [44]. Minor complications such as access site bleeding and haematoma formation occurs more frequently than greater vascular complications with arterial catheterisation [46]. The cost of minor complication remains relatively high due to increased length of hospitalization, blood transfusion and routine evaluation of such patient with vascular ultrasound [47]. The relative cost of the minor complication which occurs with greater frequency than major vascular complications can be reduced significantly with the use of vascular closure devices [47]. This along with earlier patient discharge and reduced length of hospital stay make the use of vascular closure devices is a cost effective option [11], [12], [44], [47], [48], [49], [50].

Collagen and suture based closure devices have been the mainstay of the approach to arterial closure to date. Most of the newer devices continue to use these basic concepts to seal the arteriotomy. Other methods of establishing haemostasis include the use of high intensity ultrasound energy, intraluminal nitinol braided mesh and thermal energy.

The boomerang closure device system is a novel technique for achieving haemostasis. An 18G wire with a temporary nitinol braided mesh disc is inserted through the existing working sheath. The tip of the device is deployed inside the artery, opening a low profile conformable disc. The sheath is removed and the disc is pulled for 10–15 min towards the vessel wall. This creates a site specific compression between the arteriotomy and tissue tract resulting in immediate haemostasis. The device is then completely removed, leaving no residual foreign materials at the puncture site. Finally, occlusive finger pressure is applied to close the remaining needle puncture site. Median compression times range from 5 to 7 min to achieve haemostasis. This device is applicable to 4–10F arteriotomies. Deployment success rates and haemostasis are quoted at 95–99%, with no embolic phenomena. Time to ambulation ranged from 2 to 4 h [51].

This is a non-invasive way of sealing femoral artery access punctures using high intensity ultrasound energy. The device does not enter the subcutaneous tract and no implantable foreign material is used. Targeted high ultrasound energy produces heat mediated alteration in the medial and advential collagen resulting in accelerated haemostasis. Arteriotomy closure with this device is independent of the patient’s coagulation status and puncture size. This product is still undergoing phase I clinical trials, and is yet to be launched commercially.

This device consists of 2 balloons fixed on a three lumen catheter. The distal balloon provides haemostasis through inflation at the inner arterial puncture site. Secondary haemostasis is achieved by applying direct thermal energy through a heating element to the vessel wall. The second balloon facilitates the heating induced sealing effect by creating a void at that region. Thermal energy at cellular level promotes cross linkage between protein strands resulting in a physio-chemical sealing effect. There are no available clinical data as of yet.

Section snippets

Conclusion

On average 30–50% of all catheter based procedures performed worldwide use a VCD to establish haemostais [32]. These devices have been demonstrated to reduce time to haemostasis, complications and facilitate ambulation and decrease length of hospital stay [12], [20], [22], [23], [44], [47]. Evidence suggests that complication rates are not significantly increased compared with manual compression. Emerging newer techniques continue to reduce the risks of device related complications and this is

References (51)

  • M.D. Gonze et al.

    Complications associated with percutaneous closure devices

    Am J Surg

    (1999)
  • H. Whitton Hollis et al.

    Femoral endarteritis associated with percutaneous suture closure: new technology, challenging complications

    J Vasc Surg

    (2003)
  • U.S. Boston et al.

    Infectious and ischemic complications from percutaneous closure devices used after vascular access

    Ann Vasc Surg

    (2003)
  • S.S. Upponi et al.

    Angioseal versus manual compression for haemostasis following peripheral vascular diagnostic and interventional procedures—a randomized controlled trial

    Eur J Radiol

    (2007)
  • F.S. Resnic et al.

    A cost-minimization analysis of the angio-seal vascular closure device following percutaneous coronary intervention

    Am J Cardiol

    (2007)
  • R.G. Carere et al.

    Suture closure of femoral arterial puncture sites after coronary angioplasty followed by same-day discharge

    Am Heart J

    (2000)
  • J.J. Bos et al.

    Use of a collagen hemostatic closure device to achieve hemostasis after arterial puncture: a cost-effectiveness analysis

    J Vasc Interv Radiol

    (1996)
  • W.M. Abbott et al.

    The effectiveness and mechanism of collagen-induced topical hemostasis

    Surgery

    (1975 Dec)
  • M. Chvapil et al.

    Experimental experiences with the collagen sponge as hemostaticum and tampon

    J Biomed Mater Res

    (1968)
  • R.J. Applegate et al.

    Restick following initial AngioSeal® use

    Catheter Cardiovasc Interv

    (2003)
  • Z. Lasic et al.

    Comparison of safety and efficacy between first and second generation of angio-seal closure devices in interventional patients

    J Invasive Cardiol

    (2004)
  • M.K. Hatfield et al.

    Angio-seal device used for hemostasis in the descending aorta

    Am J Roentgenol

    (2004)
  • R. Blanc et al.

    Hemostatic closure device after carotid puncture for stent and coil placement in an intracranial aneurysm: technical note

    Am J Neuroradiol

    (2002)
  • M.R. Mooney et al.

    Immediate sealing of arterial puncture sites after cardiac catheterization and coronary interventions: initial U.S. feasibility trial using the Duett vascular closure device

    Catheter Cardiovasc Interv

    (2000)
  • A. Cremonesi et al.

    Femoral arterial hemostasis using the angio-seal feminine system after coronary and vascular percutaneous angioplasty and stenting

    J Invasive Cardiol

    (1998)
  • Cited by (34)

    • Applications of Plastics in Cardiovascular Devices

      2017, Plastics in Medical Devices for Cardiovascular Applications
    • Vascular Closure Devices: Technical Tips, Complications, and Management

      2015, Techniques in Vascular and Interventional Radiology
      Citation Excerpt :

      Despite this, VCDs are only used in an estimated 30%-50% of catheter-based procedures performed worldwide to obtain hemostasis. This may in part be owing to the complexity of the devices, operator unfamiliarity and learning curve, cost, uncertain efficacy, and more severe complications.2,7 These variables may, however, relate to operator technical skill and experience with these devices.4

    • A new vascular coupling device: Assessment of MRI issues at 3-tesla

      2014, Magnetic Resonance Imaging
      Citation Excerpt :

      Vascular grafting is challenging and particularly the suturing of the anastomosis is time-consuming, which consequently subjects the patient to prolonged anaesthesia and may require cardiopulmonary support [3,4]. Depending on the location of the abnormality as well as the experience of the surgeon, the quality of the anastomosis may be suboptimal and the integrity of the graft may decrease significantly over time, resulting in compromised vasculature [5–7]. As the number of surgical procedures increases, there is a need to develop more robust and efficient techniques, as well as to create user-friendly implants that address potentially unacceptable time demands and can improve patient outcomes.

    • Quality improvement guidelines for vascular access and closure device use

      2014, Journal of Vascular and Interventional Radiology
      Citation Excerpt :

      The device tether is fixed in place externally, and the patient is kept in the supine position for 30 minutes. After this time, the device is removed in its entirety, with no residual retained component, and light manual compression is then provided for 5–7 minutes (44). The device can be used for 5–7-F arteriotomies.

    View all citing articles on Scopus
    View full text