MANTA Versus Suture-based Closure Devices Following Transcatheter Aortic Valve Replacement: An Updated Meta-analysis

Background Vascular access closure is essential in large-bore arteriotomy procedures, such as transcatheter aortic valve replacement. Suture-based devices are frequently used for vascular access closure. MANTA (Teleflex) is a collagen plug–based device used to achieve hemostasis with evolving efficacy and safety data. This study aimed to evaluate plug-based versus suture-based closure devices following large-bore arteriotomy procedures. Methods We conducted a systematic review searching PubMed, Cochrane Library, and ClinicalTrials.gov (inception through November 2021) for studies evaluating plug-based versus suture-based closure devices following large-bore arteriotomy procedures. We performed a meta-analysis comparing the length of stay, device failure, mortality, bleeding, and vascular complications between these 2 types of devices. Results Eleven studies (2 randomized controlled trials and 9 observational studies) with a total of 3123 patients were included in this analysis. Compared with suture-based devices, plug-based devices were associated with a significant decrease in the length of stay (standardized mean difference: −0.14; 95% CI, −0.25 to −0.03) and vascular closure device failure (odds ratio, 0.63; 95% CI, 0.44-0.91) following the procedure. There were no significant differences in all-cause mortality, major or minor bleeding, and major or minor vascular complications between plug-based and suture-based closure devices. Conclusions Plug-based vascular closure devices were associated with a shorter length of stay and lower risk of device failure following large-bore arteriotomy procedures without differences in mortality, bleeding, or vascular complications than suture-based closure devices.


Introduction
Vascular access closure is essential in large-bore arteriotomy procedures, such as transcatheter aortic valve replacement (TAVR).5][6][7] Vascular complications lead to worse procedural outcomes, and successful closure of large-bore access sites is essential. 8Suture-based devices have been the standard method of large-bore closure.ProGlide and Prostar XL are the 2 widely used suture-based vascular closure devices in clinical practice. 9A novel collagen plug-based device, MANTA (Teleflex), has become available to address this issue, allowing for postclosure.Plug-based closure devices have demonstrated safety and efficacy in single-arm studies, 10 but comparative studies versus suture-based closure devices have yielded conflicting results. 11,12There is no contemporary meta-analysis that has aggregated recently published randomized controlled trials (RCTs) and observational studies comparing these 2 vascular closure modalities.][15][16][17][18][19][20]

Materials and methods
We conducted a meta-analysis based on the Cochrane Collaboration guidelines and according to the Preferred Reporting Items for Systematic Reviews and Meta-analysis methodology. 21,22The protocol of this study was submitted to the International Prospective Register of Systematic Reviews on November 22, 2021; the registration is pending.This study was deemed exempt by the institutional review board of the University of Kentucky, as it was a meta-analysis of published studies that included deidentified patient information.

Data sources and searches
We performed a comprehensive literature search using 3 electronic databases, PubMed, Cochrane library, and ClinicalTrials.gov,from their inception to November 2021.Our search was restricted to English publications.Our search strategy included broad search terms: "plug-based," "MANTA," "suture-based," "Proglide," "Prostar XL," "vascular closure device," "transcatheter aortic valve replacement," and "TAVR" (Supplemental Table 1).Although plug-based closure may also be used after large-bore vascular access for mechanical circulatory support or endovascular aneurysm repair, 23 we identified no studies comparing vascular closure devices after these procedures.

Study selection
The meta-analysis included the following: (1) all observational studies and RCTs comparing plug-based versus suture-based closure devices following large-bore arteriotomy; and (2) studies with periprocedural outcomes reported, including vascular complications, bleeding, closure device failure, hematoma, pseudoaneurysm, vascular occlusion, and length of stay following the procedure.There were no criteria for the duration of follow-up or sample size of the included studies.Single-arm studies, editorials, and case reports were excluded.The search process and the selection of included studies were performed independently by 2 reviewers (A.A-A.and M.M.).The details of the selection process are outlined in Figure 1.
Data abstraction, outcomes, and quality assessment Two reviewers (W.A. and A.A-A.) independently abstracted data from the included studies using prespecified data collection forms, and discrepancies were resolved by a third investigator (M.M.).
We evaluated the quality of the included studies using the Newcastle-Ottawa Scale for observational studies 24 and the revised Cochrane risk-of-bias tool for the RCTs. 21Two authors (M.M. and A.A-A.) assessed each study independently for bias, and all discrepancies were resolved by consensus (Supplemental Table 2).Publication bias was assessed using funnel plots for all outcomes (Supplemental Figure 1).We did not perform the Egger test to evaluate for publication bias, given the number of the studies with reported results in all outcomes was <10. 25 Leave-one-out analyses were conducted and sorted to identify each trial's contribution to overall heterogeneity (Supplemental Figure 2).
Outcomes of interest were the postprocedural length of stay, vascular closure device failure, major and minor vascular complications, major and minor bleeding, need for additional intervention, mortality, hematoma, pseudoaneurysm, vascular occlusion, dissection, and need for blood transfusion.Major/minor vascular and bleeding complications were based on the Valve Academic Research Consortium definitions. 26

Data synthesis and analysis
The standardized mean difference (SMD) for the length of stay was calculated from the means and SDs in the included studies.When unavailable from the selected studies, means and SDs were calculated as described by Wan et al. 27 A random-effects model was used to combine odds ratios (ORs) with 95% CIs.We used the Mantel-Haenszel method to estimate τ 2 .The I 2 statistic was used to measure heterogeneity among the included studies (<25% considered low heterogeneity and >50% considered significant heterogeneity). 28eave-one-out sensitivity analysis was performed to detect any outliners.We performed a sensitivity analysis limited to studies that compared MANTA with ProGlide.Analyses were performed using R Studio version 3.6.3.
A total of 3123 patients were included in our analysis: 1365 in the MANTA group and 1758 in the suture-based devices group.In weighted aggregate, the study population had a mean age of 80.4 AE 6.7 years in the MANTA group and 80.7 AE 7.1 years in the suture-based group.There were 46.4% women in the MANTA group and 50.0%women in the suture-based group.The baseline characteristics of the included patients are detailed in Table 2.

Sensitivity analyses and heterogeneity evaluation
Omitting the study by van Wiechen et al, 13 with the highest events and highest weight (33%), made the difference in closure device failure nonsignificant (OR, 0.83; 95% CI, 0.54-1.29;I 2 ¼ 0%) (Supplemental Figure 3C).Heterogeneity was not substantial among specific vascular complications (Supplemental Figure 3A-M).Three leave-one-out analyses produced I 2 of >50%: 1. Hematoma: The study by Abdel-Wahab et al 16 was the study with the highest contribution to heterogeneity, and when excluded, I 2 decreased from 75% to 46% without substantial changes to the results (OR, 0.68; 95% CI, 0.34-1.33;I 2 ¼ 46%) (Supplemental Figure 3I).2. Pseudoaneurysm: The study by Abdel-Wahab et al 16 was the study with the highest contribution to heterogeneity.When excluded, the I 2 decreased from 52% to 29% without substantial changes to the results (OR 1.02; 95% CI 0.23-14.56;I 2 ¼ 29%) (Supplemental Figure 3L).

Requiring blood transfusion:
The study by Moriyama et al 20 was the study with the highest contribution to heterogeneity.When excluded, the I 2 decreased from 58% to 41% without substantial changes to the results (OR, 1.20; 95% CI, 0.64-2.25;I 2 ¼ 41%) (Supplemental Figure 3M).Sensitivity analysis, including only studies that used ProGlide as a suture-based device, showed a significantly lower risk of closure device failure in the MANTA group than in the ProGlide group (OR, 0.64; 95% CI, 0.43-0.94;P ¼ 0.02; I 2 ¼ 8%).There were no significant differences between the groups in other outcomes (Supplemental Figure 4).

Quality assessment
The quality of the included studies was assessed using the Newcastle-Ottawa Scale for cohort studies 24 and the revised Cochrane risk-of-bias tool for RCTs, 21 as shown in Supplemental Table 2.For the Newcastle-Ottawa Scale, each asterisk counts as 1 point.The maximum points are 2 for comparability and 1 for all other categories (Supplemental Table 2).A score of <5 is considered low quality, 5 to 6 is medium quality, and 7 to 9 is high quality.All the included observational studies scored high in quality assessment.The overall risk of bias for the included RCTs was low.

Discussion
We conducted a systematic review and meta-analysis including 11 studies comparing outcomes between plug-based and suture-based devices for large-bore arteriotomy closure.Compared with suture-based devices, plug-based devices were associated with a significant decrease in the length of stay following the procedure and a decrease in vascular closure device failure.There were no significant differences in mortality, bleeding, or vascular complications.
Since the 1990s, vascular closure devices have been used to achieve hemostasis following large-bore arteriotomy.The primary aims of these devices are to decrease the time to stop bleeding associated with arterial puncture, improve patient comfort, and facilitate early ambulation.Devices are categorized based on their mechanisms as passive and active approximators.Passive approximators deploy a plug without physically closing the puncture site, whereas active approximators close the puncture site using a suture or clip. 29here are 2 widely used suture-based closure devices, ProGlide and Prostar XL.The ProGlide device is more frequently used in clinical practice than Prostar XL, which is unavailable in the United States.The Prostar XL device has 4 needles at the ends of 2 polyester sutures, and misfiring of the sutures can lead to device failure and bleeding.The ProGlide device has 1 needle at the end of a polypropylene monofilament suture, which makes it easier to be replaced by a new device in case of misfiring.The suture-based devices need to be in place at the beginning of the procedure before dilating the vessel.On the other hand, the MANTA device can be deployed after the procedure, making it more helpful in case an emergent large-bore vascular access is needed. 30The MANTA device is composed of collagen along with an anchoring mechanism that allows the closure of punctures up to 25 French.Its components are resorbed within 6 months, except for the suture lock, which can help guide future access away from the prior arteriotomy site. 10There are many limitations associated with plug-based closure devices.First, vessel reaccess may be difficult, especially if any vascular complications occur and require surgical cutdown or contralateral balloon occlusion.Second, optimal plug opposition in severely calcified vessels can be challenging as the device toggle may interact with calcium.However, using suture-based devices is also challenging in severely calcified vessels because of the risk of breakage of these devices. 19,30Third, the MANTA device is more expensive than the currently available suture-based devices. 17ur meta-analysis showed a shorter length of stay in the hospital when plug-based closure devices were used, and this result persisted in the leave-one-out sensitivity analysis.However, the difference could be clinically not meaningful.We calculated the SMD between the groups, which is a measure of the distance between the groups (equals the mean difference divided by the SD).The SMD between the groups was À0.14 and, in general, it is considered a small effect since it is <0.5. 31When we limited our analysis to studies that compared MANTA to ProGlide, there was no significant difference in the length of hospital stay between the groups.Our findings confirm the results of a meta-analysis by Megaly et al 30 that includes 5 observational studies (3 studies using ProGlide and 2 studies using Prostar XL).The 2 RCTs included in our study (Abdel-Wahab et al 16 and van Wiechen et al 13 ) showed comparable hospital lengths of stay between the groups, which is different from the results of most of the included observational studies that showed shorter length of stay among the MANTA group.This could be due to a selection bias associated with the included observational studies.Additionally, the hospital length of stay seems to be driven by the results of the 3 studies that were conducted earlier (2017-2019), 2 of the studies used Prostar XL, which could be the reason.However, a post hoc analysis of the BRAVO-3 trial (Bivalirudin Versus Heparin Anticoagulation in Transcatheter Aortic Valve Replacement) showed that ProGlide was associated with a lower rate of major and minor vascular complications than Prostar XL. 32 It was noted that rates of moderate to severe calcification of the access sites were slightly lower for the MANTA group (29.9% vs 32.3%) compared with the suture-based group.As such, the length of stay could be at least partially related to differences in patient complexity in addition to the frequency of device failure.
To our knowledge, the present study is the first to show significantly lower rates of vascular closure device failure with plug-based devices than with suture-based devices.It is noteworthy that most included studies reported that most of the operators had more experience with suture-based devices at the time of their studies; thus, the lower device failure in the plug-based group was likely related to the direct effect of the device. 16,18,19There were 2% to 4% major vascular complications in selected patients and up to 9% in unselected patients with calcified arteries. 33The first single-arm study included 50 patients, and only 1 patient developed a major vascular and bleeding complication. 10Another single-arm study of 73 patients by Halim et al reported a 13.7% rate of vascular complications (all minor) and a 6.8% incidence of bleeding complications, with only 1.4% classified as major. 34The Pivotal Clinical Study to Evaluate the Safety and Effectiveness of the MANTA Percutaneous Vascular Closure Device trial by Wood et al was also a single-arm multicenter study including 341 patients who reported 4.2% major vascular complications and 2.3% major bleeding events. 35everal unresolved issues remain regarding the relative safety and efficacy of plug-based and suture-based closure devices.For example, we found that despite a higher rate of device failure in the suture-based arm, bleeding events and vascular complications were not statistically different.It was unclear how most of the studies dealt with device failure, as only a few of the included studies reported what alternative devices were used and/or what procedures were performed in case of device failure, which may affect the hospital length of stay.The ongoing MANTA Versus Suture-based Closure After Transcatheter Aortic Valve Implantation trial (NCT03811119) will help further understand the safety and efficacy of the MANTA device.There is also an ongoing need to evaluate plug-based closure devices in procedures other than TAVR, such as mechanical circulatory support and/or endovascular aneurysm repair.
Our study has several limitations.First, because of the availability of only 2 RCTs, most studies in our meta-analysis were observational and were prone to selection bias and treatment bias. 36Second, several outcomes exhibited heterogeneity on leave-one-out sensitivity analyses.Third, means and SDs were calculated for the length-of-stay outcome for a few included studies from the published medians and interquartile ranges, and this calculation assumes a normal distribution of the data, which might be inaccurate.

Conclusions
Plug-based vascular closure with MANTA was associated with a shorter length of stay and lower chance of device failure following large-bore arteriotomy procedures without significant differences in mortality, bleeding, or vascular complications, compared with suturebased closure devices.Further RCTs are necessary to evaluate the comparative safety and efficacy of plug-based closure devices across the spectrum of patient subgroups, clinical settings, and procedure types.

Figure. 1 .
Figure. 1.Details of the search results.A total of 125 records were identified, 27 full-text articles were assessed for eligibility, and 11 articles were included in the study.

Figure. 2 .
Figure. 2. Forest plots of outcomes.(A) Forest plot of length of stay after the procedure.(B) Forest plot of vascular closure device failure.(C) Forest plot of all-cause mortality.(D) Forest plot of major bleeding.(E) Forest plot of minor bleeding.(F) Forest plot of major vascular complications.(G) Forest plot of minor vascular complications.
MANTA ® is associated with shorter hospital stay compared with suture-based closure devices following TAVR Central Illustration.MANTA is associated with shorter hospital stay compared with suture-based closure devices following transcatheter aortic valve replacement.

Table 1 .
Characteristics of the included studies.

Table 2 .
Baseline characteristics of the patients in the included studies.