Table 1 presents the baseline characteristics of the two groups, demonstrating a well-balanced distribution. There were no significant differences between the Portico and ES3 groups in terms of demographic characteristics, comorbidities and pathological preprocedural rhythms. The baseline left ventricular ejection fraction (LVEF) was 54.6 (11.1) % in the Portico group compared to 52.0 (12.3) % in the ES3 group, showing a statistically significant difference (p = 0.03). However, no significant differences were observed in other baseline echocardiographic parameters, including the mean aortic valve gradient [41.1 (14.4) mmHg vs. 39.9 (14.5) mmHg; p = 0.499] and peak aortic velocity [4.42 (4.29) m/s vs. 4.14 (3.14) m/s; p = 0.382] for the Portico and ES3 groups, respectively.

Immediate post-interventional survival rate was 99.5% (n = 216) in Portico group and 100% (n = 217) in ES3 group (p = 1). There were significant differences between the Portico and ES3 groups in the usage of contrast medium, 127 (51.4) ml compared to 96.5 (46.0) ml (p < 0.001). Furthermore, the Portico group exhibited higher radiation exposure time [13.4 (5.96) min] and dose [44.3 (38.7) Gy·cm²] compared to the ES3 group with radiation exposure time of 12.9 (9.33) min and dose of 43.5 (63.4) Gy·cm² (p = 0.002; p = 0.05), respectively. Lastly, balloon valvuloplasty occurred more frequently in the Portico group, with 134 cases (61.8%), compared to 82 cases (37.8%) in the ES3 group (p < 0.001). There were no significant differences in mild final paravalvular leakage (PVL) between Portico (37.8%, n = 82) and ES3 groups (30.0%, n = 65) (p = 0.105). However, there were significantly higher rates of moderate final PVL in the Portico group (15.2%, n = 33) compared to the ES3 group (2.3%, n = 5) (p < 0.001). Similarly, final relevant transvalvular regurgitation (TVR) was more frequently observed in the Portico group (10.2%, n = 22) compared to the ES3 group (0.9%, n = 2) (p < 0.001). In two cases, patients initially received the Portico valve; however, in one case, the valve was subsequently replaced with the ES3 valve due to high-grade insufficiency, while in the other case, the replacement was performed due to valve dislocation. Furthermore, in two other cases that received the ES3 valve, a conversion to surgical aortic valve replacement was necessary because of technical issues. Post-interventional echocardiography revealed patient-prosthesis mismatch in one patient of the ES3 group. Most valve implantations were performed under general anesthesia in both groups: 157 cases (74.8%) in the Portico group and 162 cases (76.8%) in the ES3 group (p = 0.712). Conscious sedation was used in 53 cases (25.2%) in the Portico group, compared to 46 cases (21.8%) in the ES3 group (p = 0.474). Details for procedural parameters are summarized in Table 2 and illustrated in Figure 2B.

Among the ES3 group, the 29 mm valve size was the most frequently used, representing 39.2% (n = 85) of cases. The 29 mm valve size was also predominantly used for Portico valves, accounting for 40.6% (n = 88) of cases. The distribution of implanted prosthesis sizes is presented in Table 3.

All clinical outcomes at 30-days are detailed in Table 4 and illustrated in Figure 2A. At 30-days, all-cause mortality was 3.7% (n = 8) in Portico group compared to 0.9% (n = 2) in ES3 group (p = 0.11). All reported deaths in the study occurred during the post-interventional hospital stay. Consequently, the in-hospital mortality rate can be considered equivalent to the 30-day mortality rate. There was a significantly higher occurrence of new PPM implantation in the Portico group compared to the ES3 group, with 21.2% (n = 46) and 13.4% (n = 29) respectively (p = 0.042). The VARC-2 early safety composite defining complications were observed significantly more often in Portico compared to ES3 group [9.2% (n = 20) vs. 3.7% (n = 8); p = 0.032]. There were no significant differences between the two groups in the occurrence of bleeding, vascular complications, major access site complication or acute kidney injury. No events of postinterventional myocardial infarction, ventricular perforation or valve-related dysfunction requiring repeat procedures were observed in either valve group.

In the multivariable logistic regression model several factors were identified as significant risk predictors for new PPM implantation (Figure 3). These included higher age (OR: 1.07; 95% CI: 1.01–1.15; p = 0.029), preprocedural AVB (OR: 2.38; 95% CI: 1.10–5.00; p = 0.024), the use of the Portico valve prosthesis (OR: 2.17; 95% CI: 1.09–4.41), and longer total duration of procedure (OR: 1.02; 95% CI: 1.01–1.03; p < 0.001).

Immediate post-interventional success was high for both groups indicating a generally high level of safety for TAVR procedure. The 30-day mortality of 3.7% associated with Portico valve prostheses is comparable to earlier reported mortality rates in other trials (11–14). In contrast, the ES3 group showed a relatively low 30-day mortality rate of 0.9%, which aligns with previously reported data (15, 16). Considering the preprocedural calculated EuroScore II for each group, the expected mortality rates were anticipated to be quite similar. However, the observed mortality outcomes differed from these predictions, suggesting the presence of other factors that might have influenced the EuroScore II model.

New PPM implantation emerged as the second most frequent complication after new AVB. Our findings are consistent with previously reported data on new generation valve prostheses, demonstrating a relatively low overall rate of clinical complications except the occurrence of new PPM implantation (17). The rates of new PPM in patients with Portico valve prostheses were quite similar as observed by Mas-Peiro et al. (11). However, we found a significantly lower rate of PPM in patients receiving ES3 (21.2% for Portico vs. 13.4% for ES3). Previously reported rates of new PPM implantation with the ES3 valve show high fluctuation and a decreasing trend over the years. For instance, Murray et al. reported a rate of 25.5% (18), followed by De-Torres et al. reporting a rate of 19.1% in 2016 (19). Most recently, Monizzi et al. described a substantially lower rate of 6.3% in 2022 (16). The rate of 13.1% of PPM implantation in ES3 in our study aligns with the decreasing trend reported in previous studies. In our multivariable regression model, we found that the implantation of Portico valve prostheses remained an independent risk predictor for new PPM with twice the odds compared to patients receiving ES3 valve. These findings are consistent with previous studies, which show that PPM is more often associated with self-expandable prostheses than balloon-expandable prostheses (5, 6, 20).

In the context of VARC-2 early safety composite endpoint, we found a significantly higher rate in Portico valves compared to ES3 valves. Our findings suggest that patients receiving the Portico valve may be at a greater risk of experiencing one or more of these adverse events within the first 30-days following the procedure. This observation is in line with the findings from the prospective PORTICO IDE trial, which also reported higher event rates in the 30-day early composite endpoint for the Portico valve when compared to other commercially available valves (7).

As for TAVR procedure significantly more contrast medium and radiation time was necessary for Portico valve prostheses. Despite the Portico valve group showing higher numbers in all AKIN stages, there were no significant differences observed in terms of postprocedural acute kidney injury when compared to ES3 group. However, the higher usage of contrast dye may be an important aspect when choosing a valve prosthesis for patients with preexisting kidney impairment.

While Mas-Peiro et al. did not report significant differences between the Portico and ES3 valve prostheses in terms of PVL, they did observe higher numbers of PVL in Portico valves (11). In our study, we also observed a higher occurrence of PVL and found a significant increase in moderate PVL and TVR in the Portico valve group. However, with the introduction and increasing usage of the Navitor valve prothesis, the latest iteration of Portico prothesis specifically designed to improve PVL outcomes, lower rates of PVL are likely to be observed in clinical practice. The higher rate of TVR in the Portico group may be attributed to the frequent need for post-dilation, which is sometimes required to ensure proper expansion of the prosthesis.

Notably, our study also affirmed the results reported by Mas-Peiro et al. regarding the favorable procedural outcomes associated with larger valve sizes.