Bleeding and vascular complications related to invasive cardiovascular procedures are associated with significant morbidity and mortality. Indeed, the most common complications resulting from cardiac catheterisation are vascular-related, including external bleeding at the arterial puncture site, ecchymosis, retroperitoneal haematoma and pseudoaneurysms. As guidelines recommend that the patient should be given anticoagulant medications to lower the risk of developing an arterial blood clot (thrombosis) or of blood clots forming and travelling through the body (embolisation), such therapy increases the risk of post-percutaneous coronary intervention (PCI) bleeding. The association between the use of bleeding avoidance strategies and post-PCI bleeding as a function of a patient’s pre-procedural risk of bleeding is unknown. If use of a vascular closure device is to be considered as a part of the bleeding avoidance strategy, it should be noted that, while the use of an intraprocedural VCD has been shown to significantly reduce haemostasis and ambulation times after diagnostic coronary, angiography and percutaneous coronary intervention, there are conflicting data on whether use of a VCD increases or decreases the risk of access-site bleeding (ASB).1
Bivalirudin Versus Heparin
Among the current agents in the class of direct thrombin inhibitors, bivalirudin (Angiomax®, The Medicines Company, NJ, USA) has seen increased use in cardiovascular medicine over the past decade through its primary indication as an anticoagulant used during PCI. Bivalirudin has been further investigated and used as the anticoagulation strategy in the setting of cardiac and endovascular surgical procedures and is frequently used in the management of patients with heparin-induced thrombocytopenia. In comparison with heparin, bivalirudin exhibits a low immunogenic profile and provides similar or reduced major bleeding rates as well as a predictable degree of anticoagulation that is dose-related. Bivalirudin primarily undergoes dual elimination via proteolytic cleavage and renal elimination, and thus requires dose adjustment in the setting of severe renal dysfunction. Given the body of supportive data, bivalirudin is likely to continue to figure prominently as a reliable and efficient anticoagulation strategy.2
Four randomised controlled trials3 were identified that compared bivalirudin with heparin (with or without glycoprotein GP IIb/IIIa inhibitors [GPI]) in PCI. The incidence of death, myocardial infarction (MI), revascularisation and major bleeding at 48 hours was compared between these two agents overall and in patients with and without diabetes, hypertension, renal insufficiency and advanced age. The trials consisted of 11,638 patients (bivalirudin 5,861; heparin 5,777). There were no significant differences in patient characteristics between the two groups. At 48 hours, the incidence of death, MI, revascularisation and major bleeding was significantly reduced in the bivalirudin group (7.8 versus 1.08 %; p<0.001); individual ischaemic endpoints were significantly reduced for death (0.01 versus 0.02 %; p=0.049) and revascularisation (2.0 versus 2.7 %; p=0.02), with a similar reduction for major bleeding (2.7 versus 5.8 %; p<0.001). In a recent prospective PCI cohort study,4 bivalirudin was associated with reduced major and minor bleeding without a significant increase in hospital costs compared with other anticoagulation regimens. This analysis further supports the superiority of bivalirudin compared with heparin. Bivalirudin therefore provides excellent ischaemic protection with a significant reduction of bleeding complications, even in high-risk subgroups.
Overview of Vascular Closure Devices
VCDs are medical devices used to achieve haemostasis of the small hole in the artery resulting from a cardiovascular procedure or after an endovascular surgery requiring a catheterisation. During such procedures, a small incision is made in the groin area and a hole is created in the femoral artery to gain access to the artery. This hole is referred to as the access site or puncture site. On completion of the procedure, this hole needs to be closed. If appropriate care is taken with the initial arterial puncture, the chances of achieving an excellent arterial closure will be increased. Before the development of VCDs, the main method for closing the femoral artery was the application of manual compression; this involved up to 30 minutes of manual pressure or mechanical clamps applied directly to the patient’s groin, which was very painful. This was followed by up to eight hours of bed rest in the hospital recovery room. VCDs were introduced in the early 1990s in an effort to reduce the time to haemostasis, enable early ambulation and improve patient comfort. Initially, devices focused on technologies involving a suture or a collagen plug. While these technologies are effective at closing the hole, they often leave an intravascular component in the artery, which can cause complications. In addition, these technologies failed to completely address patient pain.
More recent methods to close the hole involve the use of novel materials that dissolve over a short period of time, such as polyethylene glycol.4 These technologies incorporate a more gentle deployment of the material to the outside of the artery and avoid the use of intravascular components, leaving nothing behind in the artery and consequently improving patient comfort.5
In general, VCDs can be categorised as either active closure devices (ACDs) (including immediate closure with suture devices, clips and collagen plug devices; see Table 1), passive closure devices (including devices that help with compression, such as clamps/tamping devices, assisted or enhanced coagulation and sealants; see Table 2) or hybrid devices (see Table 3). These VCDs may have a gradient of risk for vascular complications based on their respective mechanisms of action. Although small studies have identified differing rates of vascular complications for specific devices, there are no definitive data across a wide spectrum of patients.
Randomised Clinical Studies in Active Closure Devices Collagen-mediated Devices
The Vasoseal® device (Datascope Corp, Montvale, NJ, US) was one of the first ACDs developed. It was originally studied in a cohort of 100 patients undergoing diagnostic or interventional catheterisation procedures and randomised to the device or to conventional pressure dressing.6 Use of the Vasoseal resulted in a shorter compression time (4 versus 42 minutes; p<0.001) and faster ambulation (6 versus 22 hours; p<0.001) and demonstrated a trend towards a reduction in bleeding rates.6–8 The efficacy of the Vasoseal was subsequently confirmed in other randomised studies,9,10 although other trials showed small increases in access-site complications or failed to demonstrate superiority over mechanical compression during PCI.11,12 This device is no longer marketed because of findings from subsequent observational studies and inconsistency in haemostasis performance.
Angio-Seal™ (St Jude Medical, St Paul, MN, US) is among the most widely used ACDs. One of its earliest and largest randomised studies was a multicentre trial of 435 patients undergoing cardiac catheterisation and angioplasty procedures who were randomised (1:1) to the device or to manual compression.13 Angio-Seal achieved faster haemostasis (2.5 versus 15.3 minutes; p<0.0001) and lower rates of bleeding, haematoma or any complication.
A more contemporary randomised study was conducted by Chevalier and colleagues,14 who enrolled 612 moderate- to high-risk patients undergoing PCI and demonstrated faster haemostasis (5 versus 52 minutes p<0.001), reduced bed-rest time (438 versus 952 minutes; p<0.001) and reduced complication rates (6 versus 18 %; p<0.001) with Angio-Seal compared with manual compression. Although these effects of Angio-Seal were confirmed by others,15 one study still favoured mechanical compression with the FemoStop® device (St Jude Medical, St Paul, MN, US) over Angio-Seal.16
Suture-mediated Devices
Perclose® (Abbott Vascular, Redwood City, CA, US) was the first suture-mediated ACD. One of its early studies, conducted by Gerckens et al.,17 randomised 600 patients to the device or to manual compression and demonstrated faster haemostasis (8 versus 13 minutes; p<0.0001) and faster ambulation (5 versus 18 hours; p<0.0001) with Perclose use. In this particular study, Perclose was associated with comparable rates of vascular complications to those observed with manual compression in all patient populations but with lower rates in patients undergoing diagnostic procedures. The safety and efficacy of Perclose were subsequently supported by the results of other randomised studies.18,19
Clip-/Staple-mediated Devices
The StarClose™ Vascular Closure System (Abbott Vascular, Redwood City, CA, US) was the first clip-mediated ACD and uses a flexible nitinol clip to complete a circumferential, extravascular arteriotomy closing. Its pivotal study was the Clip closure in percutaneous procedures (CLIP) study, a prospective, randomised (2:1) multicentre trial that compared the safety and efficacy of StarClose with that of manual compression. A total of 596 subjects were enrolled, of whom 208 underwent diagnostic angiography only. In the latter population subset, no differences in the rates of major and minor vascular complications were observed between the two strategies.20 StarClose also reduced the mean time to haemostasis (15.5 versus 1.5 minutes; p<0.001) and time to ambulation (269 versus 163 minutes; p<0.001) and was deployed successfully in 94 % of patients. The remaining CLIP study findings were subsequently demonstrated in the interventional population subset (n=275), in which no differences in the rates of major vascular complications or infections and a trend towards a lower rate of minor complications were observed with StarClose.21
An ultrasound substudy from CLIP also confirmed these findings with StarClose.22 In a recent randomised trial of patients undergoing diagnostic and interventional procedures, StarClose and Angio-Seal were found to have similar rates of haemostasis (69 %), with some increased bruising with Angio-Seal.23
Evidence for Passive Closure Devices
Passive closure devices include devices such as clamps for assisted compression and newer technologies, such as coagulant patches and sealants. Clinical studies examining these devices are limited.
The FemoStop femoral compression system has been studied primarily as part of the control arm for clinical studies on haemostasis after endovascular procedures.24 Single-centre observational studies25,26 and a small randomised study27 found similar or reduced rates of vascular complications with FemoStop compared with manual pressure in patients undergoing interventional procedures. By contrast, a randomised study of 212 patients found FemoStop use to have a higher rate of vascular complications than manual compression.28
The Boomerang™ assisted compression system (Cardiva Medical, Inc., Mountain View, CA, US) has been studied in 96 consecutive patients undergoing diagnostic cardiac catheterisation, with successful deployment in 99 % of patients without any major complications in the study cohort.29 The D-Stat dry patch™ (Vascular Solutions, Minneapolis, MN, US) was compared with manual compression in a multicentre randomised trial of 376 patients undergoing diagnostic or peripheral angiography.30 Although time to ambulation was slightly shorter (392 versus 415 minutes; p=0.02), there were no differences in time to discharge or vascular complications. In a large, single-centr observational study, a retrospective analysis found the D-Stat patch with facilitated manual compression to be associated with earlier time to ambulation and similar vascular complication rates as standard manual compression.31 However, in a randomised trial (n=852) comparing D-Stat with Angio-Seal and Perclose in patients undergoing diagnostic or interventional procedures, the D-Stat was associated with a statistically higher rate of vascular complications than the ACDs (7.1 versus 1.9 %; p<0.0001).32
Other patch and sealant systems have even less available data. The Mynx™ water-soluble sealant system (AccessClosure Inc., Mountain View, CA, US) was studied in a multicentre single-arm study of 190 consecutive patients undergoing diagnostic and interventional procedures (5F to 7F).33 The device was found to have a major complication rate of 0.5 % in this low- to intermediate-risk population. A randomised study using the procoagulant pads Chito-Seal (Abbott Vascular, Redwood City, CA, US) and the Clo-Sur P.A.D.™ (Medtronic Vascular, Santa Rosa, CA, US) compared with manual compression for patients undergoing PCI found only a slightly reduced time to haemostasis with procoagulant pad use without an effect on overall bed-rest time or vascular complications. The Syvek Patch® (Marine Polymer Technologies, Inc., Danvers, MA, US) has been examined in a single-centre study of 1,000 consecutive patients undergoing diagnostic and interventional catheterisation and electrophysiological procedures, with minimal clinical complications.34
Overall, the evidence for passive closure devices varies significantly across device types, with the majority of studies characterised as observational studies or small randomised controlled trials in limited-risk populations.
The literature search yielded 31 prospective, randomised studies including 7,528 patients who were randomised to VCDs or manual/mechanical compression after diagnostic angiography and/or endovascular procedures. Most of these studies have excluded patients at high risk of puncture-site complication.35 Meta-analysis showed similar results in the study groups in terms of groin haematoma, bleeding, pseudoaneurysm and blood transfusion. Lower-limb ischaemia and other arterial ischaemic complications (arterial stenosis/device entrapment in the artery: 0.3 versus 0 %; p=0.07) as well as need for surgery for vascular complications (0.7 versus 0.4 %; p=0.10) were more frequent with VCDs. The incidence of groin infection was significantly more frequent with VCDs (0.6 versus 0.2 %; p=0.02).36 The use of VCDs was uniformly associated with a significantly shorter time to haemostasis. Such differences in complication rates were more evident in patients undergoing PCI, whereas use of VCD was associated with similar rates of adverse events among patients undergoing diagnostic coronary angiography.37 A large number of studies have been performed to evaluate the safety and efficacy of several types of VCDs, the results of which have encouraged their widespread use. However, a few years ago, meta-analyses on this topic failed to show a real advantage with the use of these devices.38
Of particular concern is the perceived increased risk of developing severe complications at the puncture site. Despite this, a significant number of studies have been carried out and, more recently, several prospective, randomised studies have been performed to evaluate the superiority of one device over another without any control group.39–41 Further studies are needed to obtain more conclusive results, particularly in patients at high risk of femoral-puncture-related complications (patients at the highest risk of vascular complications include but are not limited to those with known peripheral arterial disease, advanced age, female sex, liver disease, coagulopathy, immunosuppression, following valve replacement and renal dysfunction). High-risk clinical indications include emergent procedures such as primary percutaneous intervention for acute MI, prolonged multivessel intervention or procedures that require larger sheath sizes (>8F). In fact, the real challenge is in endovascular procedures currently addressed by the Preclose technique.
Bivalirudin Use – Are Closure Devices Necessary?
Marso et al.42 analysed bleeding rates after PCI procedures associated with the use of manual compression, VCDs, bivalirudin or both strategies (VCD plus bivalirudin) in patients across a spectrum of pre-procedural bleeding risk. The analysis included data from 1,522,935 patients undergoing PCI procedures performed at 955 US hospitals participating in the National Cardiovascular Data Registry (NCDR) CathPCI Registry from January 2004 to September 2008.
Overall, bleeding occurred in 30,429 patients (2 %). Manual compression was used in 35 % of patients, VCDs in 24 %, bivalirudin in 23 % and VCD plus bivalirudin in 18 %. Bleeding events were reported in 2.8 % of patients who received manual compression compared with 2.1 % of those receiving VCDs, 1.6 % of those receiving bivalirudin only and 0.9 % of those receiving both strategies. According to the NCDR CathPCI bleeding risk model, bleeding risk was classified as low (less than 1 %) in 31 % of patients, intermediate (1–3 %) in 49 % and high (greater than 3 %) in 20 % of patients. Observed rates of bleeding in these categories were 0.72, 1.73 and 4.69 %, respectively.
In high-risk patients, use of both strategies was associated with lower bleeding rates (manual compression 6.1 %; VCDs 4.6 %; bivalirudin 3.8 %; VCD plus bivalirudin 2.3 %). Use of both strategies was selected least often in high-risk patients (14.4 % compared with 21.0 % in low-risk patients). The results of this study suggest the need for additional research to better understand why higher-risk patients are least likely to receive bleeding avoidance strategies but also to determine the need to test interventions to overcome the risk-treatment paradox, such as enabling physicians to purposefully direct bleeding avoidance strategies to patients by providing pre-procedural estimates of post-PCI bleeding.
The Acute catheterization and urgent intervention triage strategy (ACUITY)43 demonstrated that bivalirudin monotherapy significantly reduces major bleeding compared with heparin (unfractionated or enoxaparin) or bivalirudin plus a GPI in acute coronary syndromes (ACS). This study sought to determine whether VCD use impacts major ASB in patients with ACS undergoing early invasive management by the femoral approach. Between 23 August 2003 and 5 December 2005, 13,819 patients with ACS were enrolled at 450 academic and community-based centres in 17 countries and randomised to heparin (unfractionated or enoxaparin) plus GPI, bivalirudin plus GPI or bivalirudin monotherapy. Access site information was collected only for the first coronary angiography procedure. Patients undergoing deferred PCI, in whom a different access site was potentially used, were therefore excluded from the analysis (n=28), as were patients with brachial access (n=90), radial access (n=798) or those whose records lacked access-site (n=914) or VCD use (n=368) information. After excluding these patients, the study population consisted of 11,621 patients who underwent coronary angiography with or without PCI by the femoral approach. Of the 11,621 patients in the study population, 4,307 (37.1 %) received a VCD (2,971 AngioSeal, 1,113 Perclose, 109 VasoSeal, 33 Duett and 81 other or unknown) and 7,314 (62.9 %) did not. Patients who received a VCD were generally a lowerrisk population compared with those who did not; they were younger in age and had lower rates of diabetes, hypertension, anaemia, baseline renal insufficiency prior MI and prior coronary artery bypass graft. Furthermore, patients who received a VCD were less likely to have elevated biomarkers at baseline. The ACUITY trial demonstrated that, compared with antithrombin regimens containing a GPI, bivalirudin monotherapy significantly reduces major and minor bleeding complications, including ASB, without increasing ischaemic complications in patients with moderate- and high-risk ACS managed with an early invasive strategy. Furthermore, the rates of major femoral ASB were significantly lower with VCD use compared with manual compression and with bivalirudin monotherapy compared with GPI-containing antithrombin regimens, regardless of VCD usage. Major ASB rates were lowest (<1 %) in patients who received both a VCD and bivalirudin monotherapy. These results suggest that the combined use of bivalirudin and a VCD may reduce major ASB in patients with ACS managed with an early invasive strategy from the femoral approach. As this trial examined only major ASB and did not consider other vascular complications such as pseudoaneurysm, arteriovenous fistula and limb ischaemia, it is important to consider the relative occurrence of these with and without the use of VCD when determining whether to use a VCD.
Conclusion and Outlook
Use of a VCD plus bivalirudin is associated with significantly lower bleeding rates. However, this strategy has been less often used among higher-risk patients. We recommend that prospective clinical studies are undertaken to determine the potential disadvantages of using VCDs and bivalirudin in combination in high-risk patients.