A systematic literature search of PubMed and EMBASE database was conducted from the inception of these two databases (EMBASE from 1974 and PubMed from 1966) to 16 April 2022, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. The following terms were used for searching: (TAVR OR TAVI OR “transcatheter aortic valve”) AND (“cusp overlap” OR “cusp overlapping”). The title and abstract were screened by two independent investigators (YC and GZ), and the study eligibility and the quality were assessed. The disagreements were solved by consulting a third investigator (LW), following which a consensus was reached between three investigators. The study protocol was registered in INPLASY (INPLASY202240092).

The research meets the following criteria was included: (1) the study made a comparison between the cusp-overlap technique and the standard technique; (2) self-expanding valves were used; (3) at least 10 patients were enrolled in the study; (4) published in the English language; and (5) conference presentation and abstract fulfilling the above criteria. Case reports or single-arm studies were excluded. If the same population was reported in different studies, only the largest and the most recent studies were included.

As our aim was to evaluate the efficacy and safety of the COP technique, the primary outcome of interest was post-operative (including in-hospital and 30-day) PPI. Secondary outcomes of interest were second valve implantation, prosthesis pop-out, intra-procedural radiation doses, fluoroscopic time, prosthesis implantation depth, length of hospital stay, and post-operative (including in-hospital and 30-day) LBBB, mortality, stroke, mean aortic valve gradient, and moderate or severe paravalvular leakage. Baseline and procedural characteristics were also gathered based on pre-defined extraction lists, including age, sex, surgical risk (Society of Thoracic Surgeon score or European System for Cardiac Operative Risk Evaluation score), pre-existing atrial fibrillation, LBBB, right bundle branch block, access of procedure, pre-dilatation, and post-dilatation. YC and GZ independently assessed the accuracy of the data and reached a consensus when discrepancies occurred.

The results of the meta-analysis were summarized as odds ratio (OR), risk ratio, and 95% confidence interval (CI). Data presented as median (interquartile range) in the included studies were converted to mean ± SD, as described in previous studies (11, 12). The Cochrane Q-statistic (χ²) and Higgins’ and Thompson’s I² statistics were assessed to test heterogeneity. A random-effect model was used if there was significant heterogeneity, which was defined as I² > 50% or p ≤ 0.01. Otherwise, a fixed-effect model was used. A subgroup analysis was performed for PPI: data based on original article or conference presentation. Besides, a sensitivity analysis was also performed for PPI by removing each study from the pooled analysis. A two-sided p < 0.05 was considered to be statistically significant. The statistical analyses were performed with stata software version 14.0 (StataCorp LP, College Station, United States) using metan command.

Since all studies included in this meta-analysis were observational cohort studies, the quality of studies was assessed using Newcastle-Ottawa Scale. Publication bias for PPI was evaluated by the Begg test/Egger test. A funnel plot was also presented to visually evaluate publication bias.

The flowchart of the literature search is presented in Figure 2. A total of 11 studies, including 3,647 patients, were finally included in this meta-analysis (13–23). All of the included studies were non-randomized cohort studies. Three studies were original articles, while the others were conference presentations or abstracts. The Newcastle-Ottawa Scale score ranged from 6 to 9 points, revealing a relatively low risk of bias (Supplementary Table 1). Across the 11 selected reports, 1,453 and 2,194 patients underwent self-expanding TAVR using COP or TCC technique, respectively. Detail of the studies’ baseline characteristics (including age, sex, surgical risk, pre-existing arrhythmia, etc.) and procedural characteristics are shown in Table 1 and Supplementary Table 2.

All the included studies have reported the incidence of PPI after TAVR. The incidence of post-operative PPI ranged from 2.4 to 26.1% in the COP group, while the PPI rate ranged between 11.9 and 27.9% in the TCC group. In pooled analysis, the cumulative PPI incidence was 9.3% (95% CI: 6.9–11.7%, random-effects model, I² = 57.6%, heterogeneity Chi-square p = 0.009) and 18.9% (95% CI: 15.5–22.3%, random-effects model, I² = 72.2%, heterogeneity Chi-square p < 0.001) in COP group and TCC group, respectively. The application of the COP technique was associated with a significant PPI risk reduction (I² = 40.3% and heterogeneity Chi-square p = 0.070, random-effects OR: 0.49, 95% CI: 0.36–0.66, p < 0.001; Figure 3).

The forest plot and details of the analyses of the secondary outcomes are shown in Figures 4,5 and Tables 2,3. Using COP technique, a higher implantation depth was achieved [standardized mean difference (SMD) = −0.324, 95% CI: (−0.469, −0.180)]. There were no significant differences between the two groups in second valve implantation, prosthesis pop-out, fluoroscopic time, post-operative LBBB, mortality, stroke, moderate/severe paravalvular leakage, mean aortic valve gradient, and length of hospital stay. However, radiation doses were higher in COP group [SMD = 0.394, 95% CI: (0.216, 0.572), p < 0.001].

The funnel plot is presented in Figure 6, with no obvious asymmetry. Besides, both Egger’s test (p = 0.764) and Begg’s test (p = 0.436) showed that no significant publication bias existed.

In subgroup analyses, PPI risk was still significantly lower in COP group when only pooled data of original articles (fixed-effect OR: 0.41, 95% CI: 0.27–0.63, p < 0.001; I² = 0% and heterogeneity Chi-square p = 0.715) or conference abstracts (random-effect OR: 0.53, 95% CI: 0.35–0.81, p = 0.003; I² = 56.2% and heterogeneity Chi-square p = 0.025, Figure 7) were used. Furthermore, the sensitivity analysis was performed by sequentially omitting each study from the pooled analysis using a random-effects model (Figure 8). The effect size remained stable when any of the studies were omitted. The application of the COP technique was associated with a lower risk of PPI in all models.

Although TAVR was thought to have similar or better mortality and stroke outcomes than surgical aortic valve replacement (SAVR), the risk of PPI after TAVR was more than twofold greater than SAVR (24–26). Self-expanding valve, which is widely used worldwide, was proven to be associated with a higher risk of PPI compared with a balloon-expandable valve. Even if in the last ten years great advances have been made in devices and techniques, there were still 17.4% of patients who underwent PPI within 30 days after self-expanding TAVR in the Evolut Low-Risk Trial (2), while the incidence of 30-day PPI was only 6.6% in patients who underwent balloon-expandable TAVR in PARTNER 3 study (1).

Since the conduction system such as the bundle of His and left bundle branch are located in the membranous septum just beneath the annulus plane, it is widely acknowledged that a higher prosthesis implantation depth could reduce PPI risk. Nevertheless, it is challenging for operators to precisely release prosthesis within a margin of millimeters.

Cusp-overlap projection technique, which was firstly reported at a conference in 2016, was introduced systematically in 2018 (27, 28). COP technique rapidly became the focus of research all over the world. Nine studies published between 2020 and 2022 have compared COP with standard TCC technique. Recently, a well-designed propensity score matching study also reported a significantly lower PPI risk after TAVR (relative risk: 0.54) (21). As a current topic of interest, COP is considered as a great procedural improvement and has even been recommended as the standard implantation technique for self-expanding TAVR (29). However, the evidence on COP is still relatively scarce. Therefore, we conducted this meta-analysis, finding that COP reduced over 45% risk of post-operative PPI (risk ratio: 0.544, 95% CI: 0.418–0.708, p < 0.001, random effects model, Supplementary Figure 1), proving the efficacy of COP technique in self-expanding TAVR.

Normally, a standard TCC view is suitable for the balloon-expandable valve. The balloon-expandable prosthesis is more easily located in the center of the aortic annulus and could often be released perpendicular to the annulus. However, the releasing process of the self-expanding valve was different. Since the self-expanding valve had an obvious longer stent frame, the prosthesis would first contact NCC. Accordingly, the NCC side served as a support point, and the prosthesis naturally progressed toward to the outer curvature of the aortic root (29). Finally the prosthesis contacted the LCC side, after which the valve releasing was finished. Therefore, the deployment of self-expanding was usually asymmetric, and the visibility of the NCC side hinge point was found to be important. In the COP technique, besides the three coronary cusps in the same plane, the LCC and RCC overlapped during fluoroscopy (Figure 1), which could facilitate better visualization of NCC and remove the parallax of a delivery system. Moreover, the conduction system and membranous septum, which are just located between the NCC and RCC, could also be better visualized. Besides, the left ventricular outflow tract was elongated, and the implantation depth could be evaluated more accurately. Considering the above, a lower PPI incidence was achieved using the COP technique.

Since the aim of the COP technique was to achieve higher prosthesis implantation, the risk of prosthesis pop-out, paravalvular leakage, second valve implantation, and whether this technique would lead to additional peri-procedural risk were the main issues. In the present study, the risk of paravalvular leakage, second valve implantation, and clinical events such as mortality and stroke were similar between COP and standard TCC groups. However, although only 2 out of 363 patients who received the COP technique suffered prosthesis pop-out in the included studies, Maier et al. suggested that COP was associated with higher rates of device reposition compared to the standard technique (57.8 vs. 16.2%, p < 0.001) in their center (30). Besides, it should be noticed that the radiation doses were higher in the COP group. Even if this might be related to the learning curve of a new technique, the difficulty of achieving higher prosthesis implantation and the potential additional risk of COP needs to be further studied.

Higher implantation of a self-expanding prosthesis might lead to coronary occlusion. Prevention techniques such as chimney stent protection and BASILICA (Bioprosthetic or native Aortic Scallop Intentional Laceration to prevent Iatrogenic Coronary Artery obstruction during TAVR) can be considered for patients at high risk of coronary occlusion. Besides, coronary access availability is another concern when using the COP technique to implant prosthesis higher (31). The method used to achieve commissure alignment during TAVR is another new technique, intended to make coronary ostia more easily available (32–34). In the SAVR procedure, surgeons can directly see coronary ostia and align the bioprosthesis with native valve commissures. However, this is quite challenging during the TAVR procedure. Recently, the COP technique was thought to be a feasible solution to achieve commissure alignment (34, 35). Using COP technique, the left side of fluoroscopy was aligned to the midpoint of NCC, and the commissure could be better visualized. However, the relevant studies are scarce and further evaluation of commissure alignment using COP versus TCC is needed.

Ultimately, COP could significantly reduce the risk of PPI. The incidence of PPI after self-expanding TAVR was only 6.4% in a low-volume center where the COP technique was used (23). The preliminary data from Optimize PRO study (NCT04091048), which evaluated the COP technique in Evolut PRO and PRO+, also revealed a favorable PPI rate (8.8%) (10). Besides, the incidence of PPI could be reduced by membranous septum guided prosthesis implantation, which could also be suitable when using the COP technique (6, 36). Theoretically, the combination of these two techniques might further reduce PPI risk. It was worth anticipating that the risk of PPI would be obviously reduced with the application of the COP technique, and the incidence of PPI after self-expanding TAVR could be as low as balloon-expandable valve implantation and even comparable to SAVR in the future.