Multicenter retrospective registry performed at 12 institutions. The study protocol was approved by the institutional ethics committees at each participating center and complied with the Declaration of Helsinki. All patients provided written informed consent for the TAVI procedure and inclusion in the registry.

Real-world consecutive symptomatic patients with severe aortic stenosis and BAV morphology were recruited. All patients were treated with the implantation of 1 of the following TAVI devices: the balloon-expandable Myval (Meril Life Sciences Pvt Ltd, India), the balloon-expandable S3U (Edwards Lifesciences, United States) or the EP+(Medtronic, United States). The THV type and size were left to the discretion of the treating physician according to patient's anatomical singularities. In all participating institutions, all devices have been previously used by the main operators and a retrospective analysis of outcomes did not show inter-institutional significant differences in main outcomes.

The Myval THV system, Conformité-Européenne approved since April 2019, is an Indian balloon-expandable TAVI prosthesis6 with a nickel-cobalt alloy frame covered internally and externally in the ventricular portion with polyethylene terephthalate to minimize the potential for paravalvular aortic regurgitation (AR) and 3 leaflets composed of bovine pericardium tissue. It is available in conventional (20, 23, 26, and 29mm), intermediate (21.5, 24.5, and 27.5mm) and extra-large sizes (30.5 and 32mm), all compatible with the 14-Fr sheath. S3U and the EP+have been extensively described previously.7,8 Briefly, the PRO+device is a SE THV with an external porcine pericardial wrap and is available in 23-mm, 26-mm, 29-mm and 34-mm sizes. The Ultra is a BE THV with bovine pericardial leaflets mounted on a cobalt-chromium alloy frame; compared with its predecessor, SAPIEN 3 (Edwards LifeSciences, United States), it adds a textured polyethylene terephthalate outer skirt, which has an approximately 40% increased height that aims to improve annular sealing. It is available in 20-mm, 23-mm, 26-mm, and 29-mm sizes.

Transthoracic echocardiograms were obtained at baseline and at 30 days of follow-up, and the measured parameters followed the recommendations of the European and the American guidelines.9 The hemodynamic performance of each THV was assessed by the degree of AR, residual transvalvular gradient, estimated aortic valve area (AVA), and severe prosthesis-patient mismatch (defined as an indexed AVA ≤ 0.65cm2/m2 [< 0.55cm2/m2 in obese patients]).10 BAV morphology was determined at baseline echocardiography.

Multidetector computed tomography scans were performed according to the guidelines of the Society of Cardiovascular Computed Tomography.11 The following parameters were calculated: aortic annulus (maximal, minimal, and mean diameters, area, and perimeter), eccentricity index, and degree of calcification (graded moderate-severe if>689 AU). BAV morphology was confirmed by MCT scans and classified according to the Sievers and Schmidtke system.12 The extent of calcification to the outflow tract was defined as any calcium detected 2-mm below the annulus.

The primary endpoint was 30-day device success, defined by the Valve Academic Research Consortium-3 consensus10 as a composite endpoint including technical success, freedom from mortality, surgery or intervention related to the device or to a major vascular or access-related or cardiac structural complication, and the intended performance of the valve (mean gradient<20mmHg and less than moderate AR). Secondary endpoints were the composite and individual components of early safety at 30 days.10 Other prespecified secondary endpoints were hemodynamic performance at 30-days as assessed by transthoracic echocardiography.

All baseline, procedural and 30-day follow-up data from the study population were retrospectively collected in a dedicated database at each participating institution. Categorical variables are presented as frequencies and percentages, and comparisons between groups were performed using the chi-square test or Fisher's exact test when necessary. Continuous variables are expressed as mean±standard deviation or median [25th-75th interquartile range]. The normal distribution of continuous variables was tested using Kolmogorov-Smirnov test, and graphically tested using the Q-Q plot. Comparisons between groups were performed using the ANOVA test or Kruskal Wallis test according to variable distribution. The Bonferroni post hoc test was used for multiple comparisons. All tests were 2-sided at the .05 significance level.

TriMatch analysis was carried out; propensity scores with 3 separate logistic regression models for each device and 3 distances were calculated between propensity scores for each possible matched triplet using the 3 models. Given those distances, the matched triplets with minimum standardized distance were retained. We established that a participant could be repeated a maximum of 3 times and that all pairs were unique. The variables included in the model are reflected in figure 1 of the supplementary data.

All analyses were performed using R software, version 3.6.1 (R Project for Statistical Computing).

Between January 2018 and January 2022, we included 360 consecutive patients with BAV and symptomatic aortic stenosis who underwent TAVI in the registry (figure 2 of the supplementary data). The global study population was divided into 3 groups depending on the implanted THV: 122 patients (33.9%) Myval THV, 129 patients (35.8%) S3U THV, and 109 patients (30.3%) EP+THV.

The mean age of the study population was 76.6±7.6 years and 101 patients (28.1%) were women. Most patients had low or intermediate surgical risk, with median Society of Thoracic Surgeons Score and EuroSCORE II of 3.6% [1.9-4.9] and 2.6% [1.7-4.2], respectively. A total of 38.3% of the patients had known coronary artery disease and 4.7% had prior coronary artery bypass grafting. Mean left ventricular ejection fraction was 54.7±13.3%, and there were no significant differences according to the THV. Aortic valve area at baseline was lower in the Myval group than in the S3U and EP+groups and the mean aortic gradient (52.3±14.2 mmHg) was also higher (46.7±14.2 mmHg in the S3U group and 48.8±18.4 mmHg in the EP+group, P=.017). Further clinical, echocardiographic, and imaging characteristics are summarized in table 1. Baseline differences were corrected after adjustment (see table 1 of the supplementary data and figure 2 of the supplementary data).

Median aortic annulus area was 501 [379-587] mm2 with severe calcification of the leaflets in most patients (a median of 4080 UA) and calcification extending toward the left ventricular outflow tract in 15.6% of the patients, with a similar proportion in each device group. Most patients had bicuspid type 1 morphology (91%) but 7.5% had type 0 morphology. There were no differences between the groups.

Balloon predilation was more frequently used before Myval than before S3U (71.2% vs 31.8%, P=.001), while postdilation was more often required following EP+(42.9% vs 16.5% with Myval and 13.2% with S3U, P<.001 for both). Prosthesis sizes are summarized in table 2 and in table 2 of the supplementary data; intermediate Myval sizes were used in 26.2% of the cases and extra-large sizes in 4.9% of Myval cases and in 32.8% of the EP+cases. The need for a second prosthesis was more frequent in all cases in the EP+group due to THV embolization (4.7% vs 0.8%; P=.033 for EP+vs Myval), but there were no other severe procedural complications such as coronary artery occlusion, annulus rupture, aortic dissection, and procedural death in our study population.

After adjustment, postdilation continued to be more often required following EP+(table 2). There were no differences in valve sizes or in the rate of periprocedural complications. One patient in the EP+group required open surgery due to left ventricular perforation. There was a trend to better outcomes regarding the composite endpoint of technical success with the balloon-expandable devices (100% for both) than with EP+(93.8%; P=.063).

The outcomes of the unmatched cohort are reported in table 3 of the supplementary data. In the adjusted analysis, the primary composite endpoint of device success at 30 days was significantly higher in the Myval group compared with the S3U group (100% vs 87.5%; P=.002), mainly due to a mean aortic gradient>20mmHg or peak aortic velocity>3m/s (table 4 of the supplementary data), and also compared with EP+(81.3%; P<.001) due to a higher rate of>moderate AR (figure 1; figure 3 of the supplementary data). The secondary safety endpoint was also better for Myval than for S3U (85% vs 70%; P=.031), and for EP+(67.5%; P=.022). Pacemaker implantation was required in 46 patients (12.8%) with no significant differences between devices (10.9%, 9.9%, and 18.3% for S3U, Myval and EP+, respectively).

Figure 1.

Primary and secondary endpoints at 30-days of follow-up in the matched population.

Device success was described according to VARC-3 criteria as a composite endpoint including technical success, freedom from mortality, surgery or intervention related to the device or to a major vascular or access-related or cardiac structural complication and the intended performance of the valve (mean gradient<20mmHg and less than moderate AR).

Early safety was described according to VARC-3 criteria as a composite endpoint including freedom from all-cause mortality, all stroke, VARC type 2 to 4 bleeding, major vascular, access-related or cardiac structural complication, AKI stage 3 to 4, more than moderate AR, new PPI, and surgery or intervention related to the device.

(0.15MB).

In the matched analysis, predischarge echocardiography showed better hemodynamic performance of Myval vs S3U in terms of aortic valve area (1.89±0.37 cm2 vs 1.45±0.18 cm2; P<.001) and mean aortic gradient (9.9±4.4 mmHg vs 13.1±4.8 mmHg; P<.001). Both parameters were comparable between Myval and EP+(figure 2). However, the rate of moderate or more residual AR was significantly higher for EP+(13%) than for any of the balloon-expandable devices (Myval: 1.9%, S3U: 0%; P=.005 (figure 3). Other prespecified secondary endpoints are listed in table 2.

Figure 2.

Thirty-day aortic valve area and gradients according to device type in the global and matched populations. AVA, aortic valve area.

(0.78MB).

Figure 3.

Thirty-day aortic regurgitation according to device type in the overall and matched populations.

(0.25MB).

According to recent European guidelines for the management of valvular heart disease, TAVI is the recommended treatment option for a wide range of patients with symptomatic trileaflet aortic valve stenosis. Conversely, there is no specific recommendation regarding TAVI for patients with BAV stenosis, in whom the default strategy remains surgical aortic valve replacement. The main reason for this is the widely heterogeneous anatomies of this valvulo-arterial disorder: the asymmetric distribution of calcification, the elliptical aortic annulus, the atypical location of coronary artery ostia, the variable phenotypes, and the frequent association with aortic dilation. Indeed, pivotal trials have traditionally excluded patients with BAV stenosis and there is no ongoing trial comparing TAVI to surgical aortic valve replacement.13

Nevertheless, recent studies have reported more favorable outcomes with TAVI than with surgical aortic valve replacement in patients with BAV stenosis, with reduced in-hospital mortality (0.7% vs 1.8%; OR, 0.35; 95%CI, 0.13-0.93; P=.035) and a similar rate of major adverse cardiovascular events at 30 days and 6 months, although TAVI was associated with higher odds of postprocedure need for permanent pacemaker at 30 days.2 Other studies support the feasibility of TAVI in BAV vs trileaflet aortic valve with both balloon-expandable14 and SE15 THV. More recently, the Low-risk bicuspid study16 and the Low-risk TAVI trial17 have reported good clinical outcomes of TAVI in patients with BAV and low surgical risk. All these promising results have allowed an exponential increase in the range of aortic stenosis patients that can be treated with TAVI devices.

Aside from the recent iterations of the most commonly used balloon-expandable and SE THV, the second-generation balloon-expandable Myval THV recently became commercially available with favorable initial clinical outcomes in patients with trileaflet aortic valve stenosis regardless of their surgical risk.3 Moreover, the MATCHBALL study has shown excellent early hemodynamic performance compared with the balloon-expandable SAPIEN 3 THV,16 and Elkoumy et al.5 has recently reported the first case series of Myval THV in BAV stenosis. According to Delgado-Arana et al.,13 intermediate sizes are chosen by ∼40% of operators, addressing the unmet need for a more calibrated THV choice; these additional sizes were only used in around one fourth of our study population but potentially led to a significantly lower residual gradient than S3U and a lower rate of residual AR than EP+, which had an impact on 30-day device success, as defined by Valve Academic Research Consortium-3, but could also affect the long-term durability of this therapy.

Another major concern after TAVI is the need for permanent pacemaker implantation, with known differences between balloon-expandable and SE-THV in trileaflet aortic valve stenosis. The cusp overlap technique is not recommended in patients with BAV, but the lower rate of pacemaker need in patients with SE THV in our cohort compared with historical reports (24.2%-18.3%) suggests that higher implants might also be performed in patients with BAV and that this could minimize the differences with balloon-expandable devices.18

Yoon et al.12 reported more procedural complications with TAVI in BAV than in trileaflet aortic valve stenosis (14.7% vs 8.6%; P=.002) with early-generation devices; these differences were not statistically significant when new-generation devices were used and, according to our finding, this can also be inferred for the new Myval THV (figure 4). Another consideration is that, although previous research suggests similar outcomes in patients with BAV morphology, some potential differences in predictive parameters for paravalvular AR, such as intercommissural distance, merit further investigation.17,18 A definitive prospective randomized clinical trial comparing second-generation THVs is needed to confirm our results and establish the decisive paradigm shift in the treatment of BAV stenosis.