Femoral venous closure: A single-centre retrospective analysis in real world all comers with MynxGrip® vascular closure device

Results: 85 patients (mean age 74 years) underwent femoral venous closure with the MynxGrip® device. 51.8% were male. The rate of concomitant anticoagulant or antiplatelet use was 52.9%. Device deployment was 100% successful with full haemostasis in all cases. There were no major vascular complications (bleeding, thrombosis, or infections). There was one case of a minor small venous hematoma which did not require treatment. The mean length of stay was less than 1 day (67.1% patients discharged the same day) and overnight stay only indicated by interventional procedure.


Introduction
Vascular closure devices (VCDs) are routinely used to achieve haemostasis following percutaneous arterial procedures as they have a number of distinct advantages over the standard method of manual pressure [1]. VCDs are known to reduce time to haemostasis, allow earlier ambulation, are less time consuming for clinicians, and improve comfort for patients [2]. A number of VCDs are currently available, and are broadly categorised by their mechanism as either 'active' or 'passive' approximators (1). A recent systematic review of randomised controlled trials (RCTs), and a Cochrane review including data from over 19,000 patients across 52 RCTs, have demonstrated similar ef icacy, safety and complication rates in closure of arteriotomy with VCDs, compared to manual pressure [3,4]. The use of VCDs has been encouraged further by the increasing number of patients who require complex endovascular procedures on the back ground of concomitant signi icant comorbidities whilst also requiring aggressive antiplatelets and anticoagulants.
However, the use of VCDs in closure following large bore venous cannulation is less de ined. Femoral venous access is required during various cardiac procedures, such as right heart catheterisation, insertion of temporary pacing wire with or without other procedures (e.g. complex PCI, Balloon Aortic Valvuloplasty-BAV, Transcatheter Aortic Valve Implantation-TAVI), electrophysiology (EP) studies including atrial/ventricular ablations, leadless pacemaker (Micra) implantation, percutaneous closure of septal defects or patent foramen ovale [5]. Thereafter, venous haemostasis following sheath removal is achieved by the application of manual pressure, for between 10-15 minutes, sometime more, but usually followed by a period of prolonged immobilisation. The use of VCDs for venous closure would therefore be advantageous if it ills the need for comfortable and reliable haemostasis reducing risks of access-site complications (bleeding, haematoma, pseudoaneurysm and infections) whilst allowing safe and early mobilisation as well as discharge [1].
The use of some VCDs for femoral venous closure has been reported, mainly from small retrospective studies (summarised in table 1) [6][7][8][9][10][11][12][13][14][15]. Coto, et al, (2002) studied the use of AngioSeal (St Jude) to close the common femoral vein in 110 patients undergoing cardiac catheterisation or intervention requiring venous access, demonstrating 100% successful deployment and haemostasis without complication [6]. In another retrospective analysis of 26 patients undergoing EP procedures, with a total of 73 femoral vein access sites closed using AngioSeal, Maraj, et al. (2015) reported device success of 98.7%, with one case of minor bruising and no major bleeding or venous complications [7]. Likewise, Shaw, et al. (2004) retrospectively examined use of Perclose devices (Abbott Vascular) for femoral vein closure in 42 patients following cardiac catheterisation [8]. The device success was reported as 88.9%, with 3 cases of small hematoma, 1 accesssite infection and 1 deep vein thrombosis [8]. Subsequent other studies have evaluated an alternative 'pre-closure' technique in which the 6F Perclose device is deployed at the start of the procedure followed by insertion of larger venous sheaths of up to 24 Fr, with tightening of the sutures delayed until the sheath is removed from the vessel, demonstrating similarly higher success rates and only minimal minor bleeding complications [9][10][11][12][13]. Two separate retrospective studies involving use of VASCADE device (Cardiva Medical) in 21 and 102 patients respectively reported 93.8% and 99% haemostasis with a few minor hematomas, one case of deep vein thrombosis and one device failure despite a signi icant proportion of patients receiving some periprocedural antiplatelet or anticoagulation therapy [14,15].
MynxGrip ® (Cardinal Health) is a passive approximator that uses a polyethylene glycol (PEG) sealant which actively adheres to the vessel surrounding the arteriotomy/venotomy site to achieve haemostasis. The sealant readily absorbs blood and subcutaneous luid to rapidly expand and ill the access tract to physically close the site. The lack of an intravascular component theoretically reduces the risk of intravascular embolization. In 2014, the MynxGrip ® device (Cardinal Health) received U.S. Food and Drug Administration (FDA) approval for use in a venous occlusion. However, only limited published data on the safety and ef icacy of the MynxGrip ® (Cardinal Health) in femoral venous closure is available. A preclinical study, in a porcine model, demonstrated the effectiveness of using the MynxGrip ® following 7F sheath use with 30day follow-up evaluation using venography plus doppler ultrasound demonstrating normal low and no evidence of intravascular sealant, thrombosis or distal embolism [16].
Ben-Dor, et al. (2018) in a prospective randomised clinical trial in 208 patients undergoing diagnostic/interventional procedures evaluated use of MynxGrip ® device for venotomy closure, compared with manual pressure and reported 100% success in achieving effective haemostasis, with no signi icant vascular or bleeding complications (in either group) whilst reducing mean time to haemostasis to 0.12 minutes, compared to an average of 7.6 minutes with manual pressure [17]. However, this study did not fully evaluate for concomitant use of necessary antiplatelets/anticoagulants nor their consequences on venous closure with use of MynxGrip ® .

Aims
The present study aimed to retrospectively evaluate the safety, ef icacy and outcomes following femoral venous closure using 5-6F MynxGrip ® in real-world patients attending the Catheter Laboratory whilst on concomitant antiplatelets and anticoagulants, with results compared to those without such therapies.

Method
A retrospective analysis of electronic clinical records was conducted to identify patients in whom the MynxGrip ® device was used for femoral venous closure following cardiac procedures at our institution. Anonymised data collated included baseline patient demographics (age, sex, current smoking, body mass index -BMI), comorbidities (hypertension, hypercholesterolemia/statin use, diabetes mellitus, previous myocardial infarction, previous coronary artery bypass graft surgery, previous stroke/transient ischaemic attack, peripheral vascular disease, estimated glomerular iltration rate-eGFR) and current use of antiplatelets and/or anticoagulants.
For clarity, the following clinical de initions were used: Current smoking was de ined as smoking within the last 3 months. Hypertension was de ined as either current use of antihypertensive drug treatment, or a previously recorded blood pressure ≥ 160 mmHg systolic or ≥ 95 mmHg diastolic. Diabetes mellitus was de ined as the need for oral antidiabetic agents, a previous fasting glucose ≥ 11.1 mmol/L or a diagnostic oral glucose tolerance test. Hypercholesterolaemia was de ined as a total serum cholesterol ≥ 5 mmol/L or current statin use.
Procedural data collected included the indication for the procedure and venous sheath size (5/6F) used. MynxGrip ® device ef icacy was de ined as the ability of device to achieve 'full haemostasis' within a minute following deployment as per manufacturer's instructions. MynxGrip ® device failure was de ined as 'any requirement for conversion to manual pressure to achieve haemostasis'. MynxGrip ® device safety was assessed by including any bleeding from the venotomy site, haematoma formation, haemoglobin drop requiring transfusion, local or systemic infections, deep vein thrombosis or re-admission with any access-site related complications within 30 days of discharge. The discharge date and length of hospital stay was also recorded.

Statistical analysis
GraphPad Prism was used for data analysis and appropriate t-tests were applied according to categorical or continuous data. For all analyses, the criterion for statistical signi icance was set at p < .05.

Results
A total of 85 patients were identi ied as having undergone femoral venous closure with the MynxGrip ® device. In 64 patients (75.3%) femoral venous access was required for temporary pacing wire insertion during balloon aortic valvuloplasty (BAV), and a further 21 patients (24.7%) underwent EP study with or without ablation. Baseline characteristics of the cohort are as shown in Table 2 Success of device deployment was 100%. Full haemostasis with haemostasis time < 1 minute after deployment was 100%. No cases required reversion to manual pressure to achieve haemostasis. There were no serious complications; bleeding/requirement for transfusion, pseudoaneurysm, access-site infections, or deep-vein thrombosis (DVT). There was a single case of a small venous hematoma which did not require any treatment and did not affect the length of stay. 67.1% patients discharged the same day. 77/85 patients (90.5%) were discharged the same or next day with 8 patients (9.4%) had a length of stay greater than 1 day, which was cardiac procedure related rather than any femoral venous access complications/issues. There were no readmissions within 30 days. Lastly, comparison between patients on any antiplatelets/anticoagulants (55 patients) with those without any such medication (30 patients) did not show any differences in success of device, complications, length of stay or 30-day follow up.

Discussion
The main inding of this retrospective study is that femoral venous closure with the MynxGrip ® device is effective and safe in real-world cohort of patients undergoing diagnostic or interventional catheterisation procedures, even with concomitant use of aggressive antiplatelets and anticoagulants. The only documented complication was a small venous hematoma of no clinical consequence.
Our indings are consistent with that of a number of other retrospective published abstracts [19,20], and a prospective randomised clinical study funded by the manufacturer of MynxGrip ® , Cardinal Health, comparing the use of this device with manual pressure that similarly reported 100% success rate for device deployment and haemostasis without complications [17]. Previous published data however did not assess for concomitant use of antiplatelet and/or anticoagulants. Accordingly, our independent all-comers real world data with 5/6F sheaths that includes patients on various combinations of antiplatelets/anticoagulants dictated by clinical needs establishes the ef icacy and safety of the use of MynxGrip ® for femoral venous closure in such high-risk patients also.
Admittedly, although the MynxGrip ® device has widely been proved to be safe, there have been isolated case reports of serious vascular complications, including one case of popliteal artery embolization [22]. It is therefore plausible that a larger study would be needed to establish if such distal embolization might also occur following venous closure as well de ine any clinical consequences arising thereof (i.e. any Deep Vein Thrombosis/Pulmonary Embolism).
Generally, bleeding and local vascular complications remain amongst the most common complications of any cardiac catheterisation procedure. In previous literature complications after femoral venous access closure by manual compression, following catheter ablation procedures, are reported at a low rate of 1.4% [24]. In addition, studies of VCDs in arterial closure have generally failed to demonstrate any reduction in complications, compared with manual pressure [3]. Therefore, given that complication rates associated with venous access are generally low, it is likely that any reduction in venous access related morbidity would require a study with very large number of patients to discern any statistically signi icant differences, compared to manual compression alone or the use of an alternate device. In addition, the use of MynxGrip ® outside of femoral venous closure is yet to be documented. The widespread use of large bore venous access in seriously ill patients (such as in Intensive Care Units), nearly always closed by manual compression, might allow alternate applications for the use of MynxGrip ® and may allow such a study in future.

Conclusion
This retrospective analysis supports the existing published data that the MynxGrip ® device is effective and safe for use in femoral venous closure, and we now report similar bene its in patients using aggressive antiplatelets/anticoagulants. Along with previous reports that the MynxGrip ® device is more comfortable for patients, its greater use in venous closure leading to improved patient comfort, shorter time to haemostasis and reduced length of stay in all cases would offer another major advantage. Future applications of the MynxGrip ® device may even evolve to closure of other large bore venous access sites.

Limitations of study
As complications following venous access are relatively low, particularly when compared to that of arterial access complications, the small number of patients studied in this report may therefore not necessarily identify true complication rates which ideally would require a study with a larger number of patients to identify. The retrospective nature of this analysis meant that some data on patient outcomes including comfort and time to ambulation was not fully extracted, and therefore fully identi ied. Furthermore, the use of the MynxGrip ® for venous closure was based on the clinical decision made by the practitioner, which may theoretically introduce an element of selection bias, with higher-risk patients more likely to experience complications having haemostasis managed with just manual pressure (i.e. excluded). Lastly, the lack of randomisation or a control group consisting of patients in whom femoral venous closure was achieved with manual pressure alone or the use of another device means that no de initive conclusions can be drawn from this data on the superiority of venous closure using the MynxGrip ® .

Ethics declaration
None required. However, in conformity with our institutional guidelines only anonymised data regarding the use of MynxGrip ® was retrospectively extracted from computerised records and used for work in this manuscript.