Results
Patient characteristics, including baseline CD and anatomical characteristics, are shown in Table 1 . The mean age was 76.8 ± 6.7 years old, 53% were female, and STS PROM was 3.2%. As ECG findings, 8.9% of patients had RBBB, 8.3% LBBB, and 24.9%first-degree atrio-ventricular block. The most prevalent anatomical type was the type 1 (158/169) 93.5%, the second common type was the type 0 (10/169) 5.9%, and one patient had the type 2 BAV (1/169) 0.6%. The mean MS length was 3.0 ± 2.0 mm. Table 1 also presented patient characteristics comparison between the patient without (n = 132) and with CD (n = 37) after TAVR. There were significant differences in the prevalence of atrial fibrillation (without CD 25.8% vs. with CD 43.2%, p = 0.0394), prior stroke (7.6% vs. 21.6%, p = 0.0295), severe leaflet calcification (74.2% vs. 94.6%, p = 0.0060), and severe LVOT calcification (2.3% vs. 18.9%, p = 0.0011), but not in RBBB (8.3% vs. 10.8%, p = 0.6396), first degree AVB (22.7% vs. 32.4%, p = 0.2273), and MS length (3.1 ± 2.0 mm vs. 2.5 ± 2.0 mm, p = 0.1147).
Procedural characteristics and complications are shown in Table 2 . There were significant differences in the implantation depth (3.4 ± 1.7 mm vs. 5.5 ± 3.0 mm, p < 0.0001), implantation depth - MS (0.3 ± 2.5 mm vs. 3.0 ± 3.9 mm, p < 0.0001), and patients with the implantation depth > MS length (55.3% vs. 78.4%, p = 0.0112). No significant difference was observed in the incidence of procedure-related complications.
Table 3 presents clinical and hemodynamic outcomes. In the whole bicuspid patient cohort, new PPI was required in 8.3% (14/169), and the incidence of new-onset LBBB or new PPI was 21.9% (37/169). The most common indication for PPI, was a complete AVB, 7 of 14 (50%), one patient (7.1%) had Mobitz type 2 AVB, two patients (14.3%) had tri-fascicular block and the remaining four patients (28.6%) had other indications for PPI. The duration from the TAVR to the PPI is shown inFigure 3 . The other clinical and hemodynamic outcomes were not significantly different (Table 3 ). Figure 4 shows that severe LVOT calcification and implantation depth – MS length were the independent predictors of new-onset CD in the multivariate logistic regression model (severe LVOT calcification: OR 5.83, CI 1.08 – 31.5, p = 0.0407; implantation depth -MS length: OR 1.31 per 1 mm, CI 1.13 – 1.52, p < 0.0001). The results of univariate analysis were shown in Supplemental table 1 .
Supplemental table 2 presents multivariate analysis for new PPI after TAVR, which showed baseline RBBB, severe LVOT calcification, and implantation depth – MS length were the predictors of new PPI.Supplemental table 3 represents the comparisons of CD and hemodynamic outcomes between the two groups based on implantation depth and MS length (implantation depth > MS length, and implantation depth ≤ MS length groups). The incidence of new-onset CD was greater in implantation depth > MS group, while the other hemodynamic outcomes were not significantly different.Figure 5 plots the distribution of implantation depth and MS length, and the red color indicates the new-onset CD.
To compare the anatomical characteristics and outcomes between TAVR for BAV and TAV, PS matching was performed to balance the patient characteristics (Table 4 ), which resulted in well-balanced except for estimated glomerular filtration rate (eGFR) and aortic valve area (AVA). There were significant differences in annulus size (543.7 ± 96.0 mm2 vs. 505.5 ± 100.8 mm2, p = 0.005), but not in severe leaflet calcification (78.9% vs. 71.0%, p = 0.0994), severe LVOT calcification (6.0% vs. 7.2%, p = 0.6590), and MS length (3.0 ± 2.0 mm vs. 3.0 ± 2.1 p = 0.9038). As to procedural characteristics, matched BAV group underwent more pre-dilatation (61.5% vs. 28.9%, p < 0.0001), implanted the THV deeper (implantation depth: 3.9 ± 2.2 mm vs. 2.9 ± 1.0 mm, p < 0.0001), and had similar rate of procedural complications (Table 5 ). New CD was significantly greater in the patients with BAV than TAV (22.3% vs. 13.9%, respectively, p = 0.0458), and the other clinical and hemodynamic outcomes were not significantly different between the two groups (Supplemental table 4 ).