Comparison of a Single Versus Double Perclose Technique for Percutaneous Transfemoral Transcatheter Aortic Valve Replacement

Background. The optimal strategy for arterial closure in percutaneous transfemoral transcatheter aortic valve replacement (TF-TAVR) remains under debate. Methods. Single-center, prospective, observational study of consecutive patients undergoing TF-TAVR between March 2018 and December 2019 who underwent closure with an upfront single vs double Perclose device. Device success, access-site vascular, and bleeding complications were defined according to the Valvular Academic Research Consortium (VARC)-2 criteria. Inverse-probability of treatment weighting (IPTW) was used to balance baseline characteristics between groups. Results. A total of 241 patients (mean age, 81.4 ± 8.5 years, 47% women) were included, of which 127 underwent an upfront single-Perclose (SP) strategy and 114 underwent an upfront double-Perclose (DP) strategy. Fifty-six percent of patients were treated with a CoreValve (Medtronic). The SP group was less likely to be on dialysis and on aspirin, but were more likely to receive a CoreValve, with larger valve sizes and larger delivery sheaths. Baseline characteristics were well balanced after IPTW adjustment. Device success rate was comparable between groups (96% in the SP group vs 93% in the DP group; P=.39). The SP technique was associated with fewer vascular complications (8.7% in the SP group vs 26.3% in the DP group; P<.01; IPTW relative risk [RR], 0.34; 95% confidence Interval [CI], 0.16-0.71) and bleeding complications (2.4% in the SP group vs 12.3% in the DP group; P<.01; IPTW RR, 0.21; 95% CI, 0.06-0.76) compared with the DP technique. There were no differences in 30-day mortality. Conclusion. An upfront SP technique is equally efficacious and not associated with increased vascular and bleeding complications compared with an upfront DP technique in patients undergoing percutaneous TF-TAVR.

J INVASIVE CARDIOL 2021;33(7):E540-E548.

Key words: large-bore access, Perclose, transcatheter aortic valve replacement

Transcatheter aortic valve replacement (TAVR) is an effective treatment option for symptomatic severe aortic stenosis.^1-6 There are several access routes available to deliver the device, including transfemoral (TF), transapical, direct aortic, and subclavian or axillary, with the TF site being the most commonly used approach.⁷ As the delivery sheaths have become smaller with each subsequent generation (now as small as 14 Fr), a fully percutaneous TF-TAVR has particular benefits, such as obviating the need for general anesthesia, reduced vascular and bleeding complication rates, and fewer adverse outcomes.⁸ Arterial closure is commonly performed using suture-mediated devices, such as the Perclose Proglide or Perclose Prostar devices (Abbott Cardiovascular), with utilization of the “preclose” technique.^9-12

There are several devices and combinations of devices that have been employed for arterial closure. Studies have compared Prostar vs Proglide,^13-15 single vs double Prostar,¹⁶ single Perclose combined with AngioSeal (St. Jude Medical) vs double Perclose,¹⁷ and single vs double Perclose.^18,19 The double-Perclose technique is used in many medical centers performing percutaneous TF-TAVR. Two studies have suggested a single-Perclose (SP) strategy may have similar vascular and bleeding complications, as well as device success rates compared with a double-Perclose (DP) approach.^18,19 However, these studies have been limited to Edwards Sapien valves (Edwards Lifesciences), and have utilized additional AngioSeal if needed (up to 76% of cases using a single Perclose device) to achieve hemostasis.^18,19 Little is known about whether SP and DP techniques are equally safe and efficacious in patients undergoing self-expanding valves (ie, CoreValve [Medtronic]) and without use of an additional AngioSeal to achieve hemostasis. Therefore, we aimed to determine the safety and efficacy of the SP strategy vs the DP strategy in combination with external manual compression in patients undergoing either CoreValve or Edwards Sapien transcatheter heart valve implantation.

Patient population. This is a prospective, observational study of consecutive patients undergoing percutaneous TF-TAVR between March 2018 and December 2019 at the regional cardiac cath lab at Kaiser Permanente Los Angeles Medical Center in California. The regional cardiac cath lab is a tertiary referral center providing advance cardiac services for 9 Kaiser Permanente medical centers across the Southern California region. All patients were evaluated by a heart team that included a senior cardiothoracic surgeon and interventional cardiologist who both agreed upon TAVR as the best treatment option. Inclusion criteria were: (1) ≥18 years of age; and (2) planned percutaneous TF-TAVR. Exclusion criteria were: (1) those undergoing alternative access routes such as subclavian access site; (2) planned surgical cut-down; or (3) valve-in-valve in mitral position. A total of 254 patients were evaluated; after applying inclusion and exclusion criteria, there were ultimately 241 patients included in the analysis. This study was approved by the institutional review board of Kaiser Permanente Southern California and the authors conformed to all institutional guidelines to perform the study.

Procedural protocols. Both CoreValve and Edwards Sapien transcatheter heart valves were used. Determination of valve type was made at the discretion of the operator. Five highly experienced operators (>75 cases/year/operator) performed the procedures included in the study. The decision to pursue an upfront SP strategy was done regardless of valve type, size, patient body habitus/body mass index, blood disorders/dyscrasias, or vascular access characteristics, such as femoral artery calcification, size, or tortuosity. If an upfront SP strategy was used, the protocol was as follows: (1) one Perclose device was deployed at the access site in a preclose fashion; (2) upon completion of the procedure, protamine 10 mg per 1000 U of heparin administered were given (heparin dose during the procedure was 100 U/kg with additional given for a goal activated clotting time >250 seconds); (3) six minutes were allowed to pass; (4) the large-bore sheaths were removed; (5) the Perclose knot was tightened; (6) while the access-site wire remained in place, external manual compression was applied for 3-5 minutes; (7) if there was residual bleeding, additional manual compression was applied for up to a total of 10 minutes; and (8) if hemostasis was still not achieved, an additional Perclose device was deployed and/or additional prolonged external manual compression was applied; no AngioSeal closure device was used. The access-site wire was removed once hemostasis could be achieved without need for further additional Perclose devices. The procedural protocol for an upfront DP strategy was as follows: (1) two Perclose devices were deployed at the access site in a preclose fashion; (2) upon completion of the procedure, protamine 10 mg per 1000 U of heparin administered were given, but optional at the discretion of the operator; (3) six minutes were allowed to pass; (4) the large-bore sheaths were removed; (5) the Perclose knots were tightened; (6) while the access-site wire remained in place, external manual compression was applied for 3-5 minutes; (7) if there was residual bleeding, additional manual compression was applied or an additional Perclose device was deployed. The access-site wire was removed once hemostasis could be achieved without need for further additional Perclose devices. Following closure, peripheral angiograms via the contralateral non-access site were performed on all patients to assess for vascular-related complications. There was 1 patient in the upfront SP group who did not receive protamine due to a history of allergic reaction to this medication.

Clinical characteristics & outcomes. All patient demographics, comorbidities, medication use prior to TAVR, echocardiographic data, computed tomography data, procedural characteristics, complications and outcomes, and 30-day all-cause mortality were obtained via manual review of electronic medical records. Access-site related vascular and bleeding outcomes and device success were categorized and defined according to the Valve Academic Research Consortium (VARC)-2 criteria.²⁰ All patients were followed for 30 days post TAVR.  Outcomes were adjudicated by two reviewers (D. Phan and N. Mansukhani).

Statistical analysis. A comparison between SP and DP groups was performed. Independent Student’s t-test (expressed as mean ± standard deviation) and Chi-square test or Fischer’s Exact test were used for continuous and categorical variables, respectively. Inverse probability of treatment weighting (IPTW) was used to balance baseline characteristics between groups. The propensity score was estimated by a logistical regression model that included 23 covariates covering demographics, and clinical and procedural characteristics. All subjects had complete data available for analysis. Differences in outcomes were assessed by relative risk (RR) both with and without IPTW adjustment. A two-sided P-value of <.05 was used as the cut-off for statistical significance. All statistical analyses were performed using R, version 3.6.3 (R Project for Statistical Computing).

We then performed a sensitivity analysis in a similar fashion as described above for the following subgroups: (1) only patients who received protamine (excluded those who did not receive protamine); (2) patients undergoing TAVR with a self-expanding valve; and (3) patients undergoing TAVR with a balloon-expandable valve. Furthermore, we compared outcomes stratified by valve size and sheath size.

Demographic and clinical characteristics. A total of 241 patients (mean age, 81.4 ± 8.5 years; 47% women) were included in the study (Table 1A and 1B). Of these, 127 (52.7%) underwent an upfront SP approach and 114 (47.3%) underwent an upfront DP approach. Compared with a DP approach, those treated with SP were less likely to be on dialysis (0.8% in the SP group vs 6.1% in the DP group; P=.02) and to be taking aspirin (59.1% in the SP group vs 72.8% in the DP group; P=.03), but more likely to have been administered protamine (99.2% in the SP group vs 80.7% in the DP group; P<0.01), to have a self-expanding valve used (78% in the SP group vs 32.5% in the DP group; P<.01), to receive a larger valve size (mean size, 27.7 ± 3.7 mm in the SP group vs 26 ± 3.3 mm in the DP group [P<.01]; 34 mm valve, 18.1% in the SP group vs 7.9% in the DP group; 29 mm valve, 30.7% in the SP group vs 17.5% in the DP group), and larger delivery sheath size (mean sheath size, 15.3 ± 1.3 Fr in the SP group vs 14.6 ± 1.0 Fr in the DP group  [P<.01]; 16 Fr sheath, 55.1% in the SP group vs 25.4% in the DP group; 14 Fr sheath, 40.2% in the SP group vs 73.7% in the DP group). There were no significant differences in access-site femoral artery calcification, femoral artery diameter, P2Y₁₂ inhibitor or anticoagulation use prior to TAVR, or history of peripheral vascular disease between groups. After IPTW adjustment, baseline and procedural characteristics between groups were well balanced (all P>.05).

Access-site related outcomes (Figure 1). There were fewer vascular complications (8.7% vs 26.3%; P<.01), minor vascular complications (7.1% vs 20.2%; P<.01), any bleeding complications (2.4% vs 12.3%; P<.01), access-site hematomas (0.8% vs 8.8%; P<.01), and femoral artery dissections (2.4% vs 9.6%; P=.02) in the SP group vs the DP group, respectively. There were no differences in device success rate (96.1% vs 93%; P=.39) or 30-day mortality rate (3.1% vs 2.6% P>.99) in the SP group vs the DP group, respectively. There were no differences in major vascular complications, minor or major bleeding complications, retroperitoneal bleeds, or pseudoaneurysms. After IPTW balancing of baseline and procedural characteristics, the SP technique remained associated with fewer vascular complications (Table 2) (IPTW-RR, 0.34; 95% confidence interval [CI], 0.16-0.71), minor vascular complications (IPTW-RR, 0.32; 95% CI, 0.14-0.75), any bleeding complications (IPTW-RR, 0.21; 95% CI, 0.06-0.76), hematomas (IPTW-RR, 0.1; 95% CI, 0.01-0.79), and femoral artery dissections (IPTW-RR, 0.18; 95% CI, 0.05-0.67).

Three patients in the upfront SP group and 6 patients in the upfront DP group required deployment of an additional Perclose device. All retroperitoneal bleeds (n = 5) and hematomas (n = 11) were treated conservatively (not requiring invasive procedures or surgeries). There were 2 pseudoaneurysms, of which 1 was treated with invasive thrombin injection. All femoral artery dissections (n = 14) were treated either conservatively without further intervention (n = 1), with balloon angioplasty (n = 9), or with stent placement (n = 4). The 1 patient in the upfront SP group who did not receive protamine due to a history of an allergic reaction did not have any vascular or bleeding complications.

Comparison of patients with vs without vascular/bleeding complications in the SP group. To evaluate for patient characteristics that may be associated with complications, we compared those with any vascular/bleeding events (n = 12) vs no vascular/bleeding events (n = 115) in those receiving an upfront SP device (Table 3). There were no statistically significant differences in baseline characteristics between groups. However, those with any vascular/bleeding complications had a non-statistical trend (all P>.05) toward being Asian race (8.3% vs 4.3%) or Hispanic race (25% vs 15.7%), taking less aspirin prior to TAVR (42% vs 61%), taking more P2Y₁₂ inhibitors prior to TAVR (42% vs 21%), smaller femoral artery diameter (6.5 ± 1.0 mm vs 7.3 ± 1.5 mm), moderate femoral artery calcification (67% vs 42%), valve-in-valve implantation (25% vs 8%), fewer self-expanding valves (CoreValve) used (58% vs 80%), and smaller valve sizes implanted (mean valve size, 25.7 ± 4.1 mm vs 27.9 ± 3.6 mm; 20 mm valve, 16.7% vs 0%; 23 mm valve, 25% vs 19.3%) compared with patients who had no vascular/bleeding complications, respectively. There were no noticeable differences in history of PVD and delivery sheath size.

Sensitivity analysis. We then performed the same analysis in different patient subgroups. In only patients who received protamine (n = 218) (Supplemental Table S1A and S1B and Supplemental Table S2), an upfront SP strategy was associated with fewer vascular complications (IPTW-RR, 0.34; 95% CI, 0.16-0.70) and bleeding complications (IPTW-RR, 0.22; 95% CI, 0.06-0.85), with no differences in 30-day mortality and device failure. Similarly, in those undergoing TAVR with a self-expanding valve (n = 136) (Supplemental Table S3A and S3B and Supplemental Table S4), an upfront SP strategy was associated with fewer vascular complications (IPTW-RR, 0.27; 95% CI, 0.09-0.83) and bleeding complications (IPTW-RR, 0.07; 95% CI, 0.01-0.62), with no differences in 30-day mortality and device failure. In patients undergoing TAVR with a balloon-expandable valve (n = 105) (Supplemental Table S5A and S5B and Supplemental Table S6), there were no differences in vascular complications (IPTW-RR, 0.4; 95% CI, 0.15-1.09) and bleeding complications (IPTW-RR, 0.67; 95% CI, 0.13-3.47), 30-day mortality, and device failure between strategies. When stratified by valve size (Supplemental Table S7) and sheath size (Supplemental Table S8), there were no statistically significant differences between an upfront SP vs DP strategy with respect to outcomes, except in those who had a 14 Fr sheath used where an upfront SP was associated with fewer vascular complications (RR, 0.32; 95% CI, 0.13-0.77) and minor vascular complications (RR, 0.26; 95% CI, 0.08-0.84).

The optimal approach for arterial closure after TAVR remains under debate. This study evaluated patients undergoing either CoreValve (56%) or Edwards Sapien (44%) valve implantation with use of either an upfront SP or DP strategy coupled with external manual compression if needed to obtain hemostasis (in contrast with prior studies using an additional AngioSeal device). We found that those receiving the SP strategy did not have increased vascular or bleeding complications. Both strategies were equally efficacious, with similar device success rates and 30-day mortality rates.

Two prior studies have compared an upfront SP vs DP strategy in patients undergoing TF-TAVR in Edwards Sapien (Sapien-XT and Sapien S3) valves.^18,19 In 2017, an analysis from the Japanese OCEAN-TAVI (Optimized CathEter vAlvular iNtervention) multicenter registry of 279 patients (99 SP vs 180 DP) found technical success (95% SP vs 92% DP), access-site related vascular (5% SP vs 8% DP) and bleeding complications (1% SP vs 3% DP) were similar between the 2 groups.¹⁸ Likewise, in 2019, an analysis of 740 patients (487 SP vs 253 DP) found vascular complication rates were similar between groups.¹⁹ The latter study used additional AngioSeal as a closure device if needed. Both studies performed propensity-score matching to control for differences in baseline characteristics with similar results. We used IPTW over propensity-score matching to control for differences in baseline characteristics while preserving our original sample of patients and size.

Our study found use of an upfront SP strategy to be associated with fewer adverse outcomes compared to an upfront DP strategy with respect to VARC-2 defined access-site related vascular complications (9% in the SP group vs 26% in the DP group; P<.01; IPTW RR, 0.34; 95% CI, 0.16-0.71) and bleeding complications (2.4% in the SP group vs 12.3% in the DP group; P<.01; IPTW RR, 0.21; 95% CI, 0.06-0.76). These results were driven by increased rates of hematomas (1% vs 9%; P<.01) and femoral artery dissections (2% vs 10%; P=.02) in the DP group. The differences in hematoma rates may be related to the use of prolonged external manual compression in the upfront SP group in the event of continued access-site bleeding. The differences in femoral artery dissections could be explained by the decreased number of Perclose devices, which has its own intrinsic risk for femoral artery dissection/stenosis given the mechanism of suture deployment. Our event rates were higher compared with the two prior studies discussed earlier.^18,19 This could possibly be explained by the inclusion of CoreValves in our study (CoreValve was excluded from the prior two studies), resulting in larger delivery sheath sizes. In addition, our study’s cohort consisted of a large proportion of patients with moderate or severe femoral artery calcification (49% in both groups). Lastly, an additional Perclose device was applied only if there was failure to obtain hemostasis after extensive efforts using external manual compression. The low usage of additional devices could partially explain the benefit seen with an upfront SP strategy, as vascular closure devices have their own inherent risks.

Reduced adverse outcomes with a single Perclose device may be explained by the intrinsic vascular complication risks associated with closure of large-bore access with suture-mediated devices.^21,22 Introducing an additional (second or even third) device would theoretically increase the risk of any vascular complication, such as artery dissection, perforation, or stenosis. A prior study suggests that the use of a single Perclose device combined with a collagen plug may be an acceptable alternative to reduce the need for an additional suture-mediated device.²³ Similar to how arterial closure was performed in our study, the use of a single Perclose device combined with external manual compression may also be an acceptable strategy. Further large studies and randomized controlled trials would help elucidate the optimal arterial closure strategy in patients undergoing TF-TAVR.

Study limitations. This is an observational study of consecutive patients with inherent limitations. We attempted to reduce confounding by adjusting baseline characteristics with IPTW. In addition, the decision for an upfront SP strategy was made irrespective of patient clinical characteristics. The limited sample size and single-center experience necessitate additional data to confirm our findings. Use of a CoreValve vs Edwards Sapien valve as well as SP vs DP arterial closure were determined at the discretion of each operator, which may have led to selection bias. Strengths of the study include: (1) high-risk patients with regard to transfemoral access (with high proportion of femoral artery calcification); and (2) high-volume center (all operators with experience >75 cases per year). In addition, this study included patients undergoing CoreValve implantation, which is lacking in the current literature on this topic.

The SP technique is not associated with increased vascular and bleeding complications compared with a DP technique in patients undergoing percutaneous TF-TAVR with either a CoreValve or Edwards Sapien valve. Both techniques are equally efficacious. Larger studies, including randomized trials, are warranted to confirm these findings.

From the ¹Department of Cardiology, Kaiser Permanente Los Angeles Medical Center, Los Angeles, California; and ²Regional Cardiac Catheterization Lab, Kaiser Permanente, Los Angeles, California.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Aharonian is a consultant for Medtronic. The remaining authors report no conflicts of interest regarding the content herein.

Manuscript accepted September 15, 2020.

Address for correspondence: Naing Moore, MD, Regional Cardiac Catheterization Lab, Kaiser Permanente, Los Angeles Medical Center, 4867 Sunset Boulevard, Cardiac Cath Lab, Room 3757, Los Angeles, CA 90027. Email: Naing.A.Moore@kp.org