Mitral Valve Transcatheter Edge-to-Edge Repair With PASCAL Vs MitraClip: A Systematic Review and Meta-analysis

Background. Transcatheter edge-to-edge repair (TEER) using the MitraClip (Abbott Vascular) system has emerged as a standard treatment for patients with symptomatic severe secondary or inoperable primary mitral regurgitation (MR). The relatively recent approval of the PASCAL Transcatheter Valve Repair System (Edwards Lifesciences) has expanded the options of TEER devices. However, evidence comparing PASCAL with MitraClip systems is still limited.

Methods. We conducted a systematic literature research and meta-analysis in PubMed, Medline, and EMBASE databases for studies comparing PASCAL and MitraClip systems.

Results. Four observational studies and 1 randomized controlled trial, involving 1315 patients total, were eligible for inclusion. All patients exhibited symptomatic (NYHA II-IV) MR grades 3+ or 4+. Baseline characteristics were comparable across all included studies. The clinical outcomes were assessed according to the Mitral Valve Academic Research Consortium consensus. The procedural success rates for the 2 devices were comparable in terms of achieving post-procedural MR grades of less than or equal to 2+ and less than or equal to 1+. Furthermore, most patients improved their clinical status, with no significant differences between patients treated with PASCAL and those treated with MitraClip. In terms of safety, both procedures exhibited low overall mortality rates and occurrence of major adverse events (MAE), without significant difference between the 2 devices. These findings remained consistent in both short- and long-term follow-up assessments.

Conclusions. Our study revealed similar effectiveness and safety profiles between the PASCAL and MitraClip devices in patients experiencing significant symptomatic MR.

Mitral regurgitation (MR) is the second most frequent valvular heart disease in Europe and the first in the United States.^1,2 Left untreated, severe symptomatic MR is associated with significant morbidity and mortality.³ Mitral valve surgery remains the best option for symptomatic severe degenerative MR and selected patients with symptomatic severe functional MR.^4-6 However, many patients are not fit for mitral valve surgery because of high surgical risk related to relevant comorbidities and/or advanced age.⁷ In this setting, transcatheter edge-to-edge repair (TEER) has emerged as an alternative for patients with symptomatic severe functional MR or inoperable degenerative MR.^8,9

The safety and efficacy of the MitraClip device (Abbott Vascular) was primarily tested in multiple randomized controlled trials (RCT), leading to an exponential expansion of its utilization.^10-12 In early 2019, the PASCAL Transcatheter Valve Repair System (Edwards Lifesciences) received Conformité Européenne (CE) mark approval, extending the field of TEER devices. Although both devices address MR by leaflet approximation, the PASCAL system offers distinct technical features, including a central spacer to fill the regurgitation orifice, and a less traumatic grasping feature. However, since the more recent approval, the evidence on PASCAL use is scarce compared to that on MitraClip. Therefore, it is reasonable to question whether these 2 procedures can be considered interchangeable. Accordingly, the present study aimed to systematically compare PASCAL with MitraClip regarding safety and efficacy in patients with significant symptomatic MR.

Search strategy. We performed a systematic literature review through February 2023 according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Guidelines.¹³ We systematically searched pertinent articles on PubMed, Medline, and EMBASE databases. Terms used to detect the desired studies were the following: “MitraClip”, “PASCAL”, “MitraClip versus PASCAL”, and “Mitral Valve Transcatheter Edge-to-Edge Repair”. The search included only peer-reviewed studies on human subjects published in the English language. To include pertinent papers, 3 authors (F.O., I.A.A., A.P.) independently looked for additional citations from the references list of included relevant articles.

Study selection. Three reviewers (F.O., I.A.A., L.T.) independently proceeded with the initial screening process to recognize all citations of potential acceptability. The inclusion criteria were the following:

1. Comparison between PASCAL and MitraClip in patients suffering from 3+/4+ symptomatic (New York Heart Association [NYHA] ≥2) degenerative (primary) or functional (secondary) MR;
2. Studies including patients 18 years old or older;
3. Studies including 100 or more participants;
4. Observational studies (OBs) or RCT design published in a peer-reviewed journal; and
5. Original untranslated studies written in English language only.

Full-text papers of recognized abstracts pertinent to our inclusion criteria were evaluated for eligibility. The excluded papers were case reports/series, letters, conference abstracts, and editorials. The Kappa statistic was utilized to assess the inter-rater reliability of the 2 reviewers.¹⁴

Data extraction. Two authors (A.B., A.F.) extracted inherent data independently using a standardized recording tool to document the study setting and design, year of publication, number of study participants, country of origin, participant clinical characteristics, and study outcomes.

Quality assessment. Two authors (F.R.G, D.C.) assessed the risk of bias of all considered RCTs according to the Cochrane Collaboration's tool.¹⁵ In case of discrepancy, a third author (F.A.) rechecked data and made the final evaluation (Figure S1).

We applied the Newcastle-Ottawa Scale (NOS) to assess the included quality assessment of the included observational studies. Two authors (F.R.G., D.C.) evaluated the quality of each study by checking 3 broad categories: (1) selection, (2) comparability, and (3) outcomes. Any individual study could achieve a maximum of 9 stars (4 stars for selection, 2 stars for comparability, and 3 stars for the outcome).¹⁶ Then, we converted an individual study's NOS score into Agency for Healthcare Research and Quality (AHRQ) standards.¹⁷ Thus, studies were sub-classified into 3 class of quality categories (Supplemental Table):

1. Poor quality: less than or equal to 1 star for selection domain OR 0 stars for comparability domain OR less than or equal to 1 star for outcome domain.
2. Fair quality: 2 stars for selection domain AND 1 or 2 stars for comparability domain AND 2 or 3 stars for outcome domain.
3. Good quality: greater than or equal to 3 stars for selection domain AND greater than or equal to 1 star for comparability domain AND greater than or equal to 2 stars in outcome domain.

Data analysis and synthesis. We utilized the Review Manager software (RevMan-5.4.1) to conduct our statistical analyses. The overall odds ratio (OR) or mean difference (MD) and the 95% CI for each analyzed parameter were pooled using a random effects model. Furthermore, we drew forest and funnel plots to visually evaluate the pooling results and identify possible publication biases and the power of the included studies. OR greater than 1 indicated an increased risk of the considered outcome; OR value of 1 indicated no observed association; and OR less than 1 indicated a decreased risk of the considered outcome. A 2-sided P-value less than .05 was considered statistically significant. Furthermore, the studies’ heterogeneity results were calculated using the Higgins I², which measured the percentage of the total variation across the included studies.¹⁸ The values of I² lie between 0% and 100%, and a value of 0% indicates no heterogeneity. We classified heterogeneity as mild (I²<25%), moderate (25≤I²<50%), severe (50≤I²<75%), and very severe (I²≥75%).¹⁹

Literature search. Figure 1 shows the PRISMA flow diagram of the study identification. A total of 1875 records were identified in the preliminary search. Of these, 1750 did not meet inclusion criteria and were therefore excluded. A total of 10 studies were used for data extraction and synthesis for our systematic review. Five articles were finally included in our meta-analysis.

Characteristics of the included studies. Characteristics of the included studies are listed in Table 1 and Table 2. The majority of the included papers were observational-designed (3 retrospective and 1 prospective),^20-23 while one was an RCT.²⁴ The included studies were conducted in Germany and the United States and included more than 20 academic hospitals. The total number of individuals in our meta-analysis was 1315 (659 PASCAL and 656 MitraClip). Most of the patients were in class NYHA III-IV at baseline (81.3% vs 81.9% for PASCAL and MitraClip, respectively) with no heterogeneity between the 2 groups (OR, 1.12; 95% CI, 0.84-1.49; P=.45; I²=0%). All patients had symptomatic 3+ (39.6% vs 37.5%; OR, 1.19; 95% CI, 0.93-1.51; P=.16; I²=0%) or 4+ MR (60.4% vs 62.5%; OR, 0.84; 95% CI, 0.66-1.07; P=.16; I²=0%). Functional MR was the main indication for intervention (53.4% vs 58.7%; OR, 1.00; 95% CI, 0.78-1.28; P=.99; I²=0%), followed by degenerative (46.6% vs 39.5%; OR, 1.04; 95% CI, 0.81-1.34; P=.74; I²=0%) and mixed (overall 1.4%) regurgitation (Figure S2).

The clinical outcomes were reported using the Mitral Valve Academic Research Consortium consensus,²⁵ except for a minimal difference in the procedural success definition in 1 retrospective study.²⁰ Follow-up differed among the studies, ranging from hospital discharge to 18 months. Thus, we divided the outcomes into 3 temporal categories: before hospital discharge, short-term follow-up (30 days to 4 months), and long-term follow-up (6 months up to 18 months).

Publication bias. The funnel plot in Figure S3 highlighted no significant asymmetry, limiting the overall risk of publication bias.

Before hospital discharge. Figure 2 shows the comparison between PASCAL and MitraClip in terms of pre-hospital discharge outcomes. The procedural success was comparable (97.0% vs 97.9%; OR, 0.60; 95% CI, 0.30-1.20; P=.15, I²=0%) among the 2 procedures. No statistically significant difference in post-procedural MR less than or equal to 2+ (98.1% vs 100%; OR, 1.14; 95% CI, 0.72-1.81; P=.57; I²=0%) as well as MR less than or equal to 1+ (72.8% vs 64.9%; OR, 1.16; 95% CI, 0.76-1.20; P=.49; I²=58%) was assessed. Mean anterograde transmitral valve gradient was also comparable (MD -0.30 mmHg; 95% CI, -0.80/+0.20; P=.24; I²=64%). Furthermore, the overall mortality was similar (0.18% vs 1.24%; OR, 0.35; 95% CI, 0.06-1.91; P=.22; I²=0%).

Short-term follow-up. At short-term follow-up (Figure 3), there were comparable rates of overall mortality incidence (1.9% vs 3.3%; OR, 0.67; 95% CI, 0.32-1.39; P=.28; I²=0%), major adverse events (4.7% vs 8.4%; OR, 0.68; 95% CI, 0.38-1.19; P=.18; I²=0%), cerebrovascular events (0.3% vs 1.1%; OR, 0.44; 95% CI, 0.09-1.19, P=.31; I²=5%), and severe bleeding (2.5% vs 1.1%; OR, 1.91; 95% CI, 0.69-5.33; P=.21; I²=0%) between PASCAL and MitraClip. The necessity for reintervention was also similar between the 2 devices (0.8% vs 1.1%; OR, 0.87; 95% CI, 0.26-2.93; P=.83; I²=0%). No statistically significant difference was observed in post-procedural MR less than or equal to 2+ (93.2% vs 90.9%; OR, 1.21; 95% CI, 0.68-2.18; P=.52; I²=17%) or in MR less than or equal to 1+ (68.5% vs 55.6%; OR, 1.31; 95% CI, 0.74-2.31; P=.36, I²=70%). Moreover, the NYHA I-II was comparable (65.1% vs 63.2%; OR, 0.85; 95% CI, 0.45-1.61; P=.49; I²=67%).

Long-term follow-up. At long-term follow-up (Figure 4), the overall mortality incidence (8.2% vs 12.3%; OR, 0.83; 95% CI, 0.43-1.61; P=.58, I²=0%), major adverse events (12.7% vs 29.0%; OR, 0.58; 95% CI, 0.31-1.10; P=.10; I²=0%), and heart failure re-hospitalization (5.1% vs 18.6%; OR, 0.42; 95% CI, 0.17-1.00; P=.05; I²=0%) were comparable between PASCAL and MitraClip. No statistically significant difference in post-procedural MR less than or equal to 2+ (96.6% vs 96.3%; OR, 0.89; 95% CI, 0.26-3.05; P=.85; I²=0%) or MR less than or equal to 1+ (80.9% vs 68.8%; OR, 1.61; 95% CI, 0.96-2.72; P=.07, I²=0%) was assessed. The NYHA I-II was comparable (81.1% vs 80.0%; OR, 0.55; 95% CI, 0.24-1.22; P=.14; I²=0%).

The aim of our study was to assess the short- and long-term outcomes between the PASCAL and MitraClip devices in patients with significant (grades 3+/4+) symptomatic MR undergoing TEER. Figure 5 summarizes the main findings.

Considering the more recent approval of PASCAL devices compared to MitraClip (2019 and 2008, respectively), some concerns might arise in terms of technical outcome discrepancies, such as increased implantation time or procedural success. However, our pooled analysis highlighted comparable procedural technical success rates between the 2 systems (97.0% vs 97.9%; OR, 0.60; 95% CI, 0.30-1.20; P=.15) with no heterogeneity (I²=0%). Nevertheless, the results were consistent at the short-term follow-up, where mitral valve interventions were comparable between the 2 groups (0.8% vs 1.1%; OR, 0.87; 95% CI, 0.26-2.93; P=.83, I²=0%). Although it was impossible to quantitatively assess procedural time differences between the 2 devices, 4 of the 5 included studies did not find any statistically significant difference.^20-23 At the same time, the remaining study showed only that Mitraclip was a few minutes quicker than Pascal.²⁴ One possible limitation of our findings is that the studies were conducted in experienced centers and, therefore, procedural time and success rates might differ in less experienced centers. Nevertheless, we believe that any learning curve issues will likely be overcome in the long term.

Studies have shown that the presence of residual MR following transcatheter MV repair negatively impacts prognosis, and that reducing MR to grade less than or equal to 1+ should be the primary goal.^26,27 Failure to achieve this goal has been linked to poor clinical outcomes, emphasizing the importance of achieving MR less than or equal to 1+.^26,27 Although PASCAL and MitraClip both target the mitral leaflets to reduce MR, the devices present differences in technical design. One of the main differences is related to the central spacer present in the PASCAL system, which serves to fill the regurgitation orifice and reduce the tension of the leaflets, possibly offering additional benefits in reducing MR and limiting the potential transmitral mean gradient increase. Our pooled analysis revealed excellent and consistent results in terms of postprocedural MR less than or equal to 2+ and MR less than or equal to 1+ and mean anterograde transmitral valve gradient without any statistically significant difference between PASCAL and MitraClip.

However, heterogeneity was severe. Looking in depth, the Mauri et al study included the largest number of patients (with a significant effect size).²² In this study, a PASCAL-based strategy resulted in more prevalent residual MR less than or equal to 1+ and reduced transmitral mean gradient greater than or equal to 5 mmHg compared to the MitraClip group. As previously shown, these conclusions were not confirmed when we pooled these data with other studies. Upon closer examination of the included studies, Mauri et al included patients from 2010 to 2018. Hence, these patients were treated with older generations of MitraClip only (NT, NTR, or XTR systems). In contrast, in the only included RCT, more than 60% of the patients were treated with the modern G4 system (NT, NTW, XT, XTW).²⁴ These evolved devices allow independent leaflet grasping (like PASCAL) and better clip manipulation, limiting the design differences between PASCAL and MitraClip. Nevertheless, the preliminary results of the G4 MitraClip were positively evaluated in the EXPAND G4 Study, where 91% of the included patients had residual MR <=1 at 30-day follow-up.²⁸ Indeed, we determined that the severe heterogeneity might be explained mainly by different generations of device types rather than substantial procedural outcomes or patient selection. We therefore performed a sensitivity analysis (Figure S4) excluding the Mauri et al study. Not only did the results not change, but the severe heterogeneity previously evidenced became null.

In terms of safety, PASCAL and MitraClip were associated with excellent results. Specifically, the overall mortality and MAE were low in both procedures, with comparable results between the 2 devices at short- and long-term follow-ups. Most patients improved their clinical status tremendously, with more than 4 out of 5 patients classified as NYHA I-II during long-term follow-up. However, no statistically significant differences were assessed between the PASCAL and MitraClip, demonstrating potential interchangeability of the 2 TEER systems.

Study limitations. Our study presents some limitations. First, most of the evidence was derived from observational studies. In addition, there was an important patient loss at long-term follow-up, potentially affecting the outcomes. It was not possible to have a subgroup analysis for degenerative MR and functional MR due to lack of data.

Our study demonstrated comparable efficacy and safety of PASCAL and MitraClip systems in patients with significant symptomatic MR undergoing TEER. The availability of 2 devices may widen TEER utilization, while competition between companies may lead to improved quality and more options for patients with less cost.

From the ¹Department of Molecular Medicine, Division of Cardiology, University of Pavia, Pavia, Italy; ²Department of Cardiology, Leiden University Medical Center, Leiden, the Netherlands; ³Cardiology Unit, Ospedale Maggiore, Lodi, Italy; ⁴Interventional Cardiology, Policlinico San Donato, Milan, Italy; ⁵Division of Cardiology, Fondazione IRCCS Policlinico San Matteo, Pavia, Italy.

Disclosures: The Department of Cardiology of the Leiden University Medical Center received unrestricted research grants from Abbott Vascular, Bayer, Biotronik, Boston Scientific, Edwards Lifesciences, GE Healthcare and Medtronic. Jose M. Montero received a research grant from Shockwave Medical. The remaining authors report no financial relationships or conflicts of interest regarding the content herein.

Address for correspondence: Federico Oliveri, MD, University of Pavia: Universita degli Studi di Pavia, Pavia, Italy. Email: federico.md.oliveri@gmail.com