Mid- to Long-Term Clinical and Echocardiographic Effects of Post-procedural Permanent Pacemaker Implantation After Transcatheter Aortic Valve Replacement: A Systematic Review and Meta-Analysis

Aims To date, the prognostic effects of permanent pacemaker implantation (PPI) after transcatheter aortic valve replacement (TAVR) remain controversial. The purpose of this meta-analysis was to investigate the mid- (1 year) to long-term (> 1 year) clinical and echocardiographic effects of post-procedural PPI in patients after TAVR. Methods PubMed, Embase, Web of Science, and Cochrane Library databases were systematically searched from the establishment of databases up to 1 December 2021. Studies comparing clinical and echocardiographic outcomes between patients with and without post-TAVR PPI of ≥ 1-year follow-up were collected for further meta-analysis. Results A total of 39 studies comprising of 83,082 patients were included in this meta-analysis. At mid-term follow-up (1 year), the pooled results demonstrated a higher risk of all-cause mortality in patients with post-procedural PPI than those without following TAVR (relative risk (RR), 1.17; 95% CI, 1.10–1.24; P < 0.00001). No significant differences were observed in cardiovascular mortality (RR, 0.86; 95% CI, 0.71–1.03; P = 0.10) or heart failure rehospitalization (RR, 0.91; 95% CI, 0.58–1.44; P = 0.69) at 1-year follow-up. At long-term follow-up (> 1 year), post-TAVR PPI had negative effects on all-cause mortality (RR, 1.18; 95% CI, 1.09–1.28; P < 0.0001) and heart failure rehospitalization (RR, 1.42; 95% CI, 1.18–1.71; P = 0.0002). There was no difference in long-term cardiovascular mortality between the two groups (RR, 1.15; 95% CI, 0.97–1.36; P = 0.11). Left ventricular ejection fraction (LVEF) was not significantly different at baseline (mean difference, 1.40; 95% CI, –0.13–2.93; P = 0.07), but was significantly lower in the PPI group at 1-year follow-up (mean difference, –3.57; 95% CI, –4.88 to –2.26; P < 0.00001). Conclusion Our meta-analysis provides evidence that post-TAVR PPI has negative clinical and echocardiographic effects on patients at mid- to long-term follow-up. Further studies are urgently needed to explore the cause of these complications and optimize the treatment and management of patients requiring permanent pacing after TAVR. Systematic Review Registration [https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42021289935], identifier [CRD42021289935].

Methods: PubMed, Embase, Web of Science, and Cochrane Library databases were systematically searched from the establishment of databases up to 1 December 2021. Studies comparing clinical and echocardiographic outcomes between patients with and without post-TAVR PPI of ≥ 1-year follow-up were collected for further meta-analysis.
Conclusion: Our meta-analysis provides evidence that post-TAVR PPI has negative clinical and echocardiographic effects on patients at mid-to long-term follow-up. Further

INTRODUCTION
Transcatheter aortic valve replacement (TAVR) has become a well-established therapy for patients with severe aortic stenosis and high risk for surgical aortic valve replacement (1,2). Recent randomized controlled trials provided evidence to extend the application of TAVR to low-risk patients (3,4). Despite technological advances and clinical experience accumulation, atrioventricular node, and infranodal tissues remain easily impaired during the implantation of the valve prosthesis. Conduction abnormalities (e.g., high-degree atrioventricular block and new-onset persistent left bundle branch block) are frequently observed after TAVR, and patients often require permanent pacemaker implantation (PPI) (5). The application of post-TAVR PPI was reported in approximately 2.3-37.7% of patients, and the rates largely vary according to the types and generations of the transcatheter valves (6).
Cardiac pacing is a recommended therapy to reduce the risk of death related to severe bradycardia arrhythmias. However, traditional right ventricular pacing (RVP) can cause electrical and mechanical dyssynchrony (7,8), thus increasing the risk of mortality and heart failure hospitalization (9)(10)(11). Currently, it remains controversial whether the application of PPI could influence the clinical symptoms and survival outcomes after TAVR (12). Previous meta-analyses were limited by a small number of studies or lack of long-term follow-up (13,14). This meta-analysis aims to investigate the mid-to long-term clinical and echocardiographic outcomes of post-procedural PPI in patients after TAVR.

Search Strategy
We performed a systematic literature search in PubMed, Embase, Web of Science, and Cochrane Library from the establishment of databases up to 1 December 2021 by two investigators (Shun Xu and Enrui Zhang) independently. (pacemaker [Title/Abstract])))). The searching strategies for Embase, Web of Science, and Cochrane Library were provided in Supplementary Table 1. We also manually screened reference lists of retrieved reviews, reports, and other relevant publications to identify additional pertinent studies.

Study Design
The protocol of this meta-analysis has been registered in PROSPERO (Registration ID: CRD42021289935). Clinical studies were eligible if they met the following inclusion criteria: (1) studies comparing clinical and echocardiographic outcomes between patients with and without post-procedural PPI after TAVR, including all-cause mortality, cardiovascular mortality, heart failure rehospitalization, and left ventricular ejection fraction (LVEF); (2) studies with a follow-up of ≥ 1 year; (3) studies with full texts published in English in peerreviewed journals. We only included the study containing the most data for multiple publications of the same trial. We excluded review articles, case reports, letters, editorials, articles lacking outcomes of interest, studies without detailed data, and studies with a follow-up of < 1 year. Importantly, we also excluded studies that failed to distinguish patients with PPI before TAVR. Two independent investigators (Shun Xu and Enrui Zhang) assess eligibility by screening and reviewing article titles, abstracts, and full texts. Any disagreement about eligibility was clarified via consulting a third investigator (Jinyu Sun).

Data Extraction and Quality Assessment
Two investigators (Shun Xu and Enrui Zhang) independently extracted data for each eligible study. Any disagreement was resolved through discussion with a third investigator (Jinyu Sun) to reach a consensus. The following characteristics were included: first author, year of publication, inclusion period, number and region of centers, sample size, PPI criteria, patient demographic characteristics, and the following midterm (1 year) to long-term (> 1 year) outcomes, including all-cause mortality, cardiovascular mortality, heart failure rehospitalization, and LVEF.
The quality of studies involved was assessed by two investigators (Shun Xu and Enrui Zhang) independently using the Newcastle-Ottawa Scale (NOS). The NOS tool involved three aspects, and a maximum of 9 stars can be allotted to each study: the selection of cohorts (0-4 stars), the comparability of cohorts (0-2 stars), and the assessment of the outcome (0-3 stars). A NOS score ≥ 6 stars indicated moderate-to-high quality, while a NOS score < 6 stars indicated low quality. Discrepancies were resolved by consulting a third investigator (Jinyu Sun) to reach a consensus.

Statistical Analysis
Continuous variables were presented as mean ± standard deviation, and categorical variables were presented as frequencies or percentages. Relative risk (RR) with corresponding 95% confidence intervals (CIs) for each endpoint was calculated and analyzed for categorical variable outcomes. Continuous data were summarized as a mean difference with 95% CI. P-value < 0.05 was considered statistically significant. The heterogeneity between studies was quantified by I-squared (I 2 ) statistic, with a fixed-effects model adopted when the I 2 -value was < 50% and a random-effects model applied otherwise. Review Manager version 5.3 was used for all the statistical analyses. The meta-analysis was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (15). Figure 1 shows the flow chart of the study selection. A total of 9,852 records were initially identified from the databases according to the searching strategies, including 1,842 from PubMed, 4,782 from Embase, 3,053 from Web of Science, and 175 from Cochrane Library. After title and abstract screening, a total of 4,321 duplicates and 5,461 irrelevant records were excluded, the remaining 70 full-text articles to be reviewed for eligibility. Of those, 22 studies were excluded for having no outcomes of interest or without provided data. Two studies were excluded due to failing to distinguish patients with PPI before TAVR. One study was excluded because the follow-up duration was less than 1 year. Six case reports were also excluded. Finally, 39 studies containing 83,082 patients were included for further analysis (16-54) ( Table 1).

Study Selection and Quality Assessment
All included studies had moderate-to-high quality while none had less than 6 points according to NOS: two with 9 points, nineteen with 8 points, six with 7 points, and 12 with 6 points. The details of the quality assessment are shown in Table 2.

Mid-Term (1 Year) Clinical Effects of Post-procedural Permanent Pacemaker Implantation After Transcatheter Aortic Valve Replacement
The risk of mid-term all-cause mortality was pooled from 27 studies that included 49,579 patients, and 7,235 patients were implanted with permanent pacemakers after TAVR. There were 1,197 of 7,235 (16.54%) cases of all-cause mortality in the PPI group while 6,285 of 42,344 (14.84%) cases in the no PPI group. The pooled results demonstrated that patients with PPI had a higher risk of death than those without PPI following TAVR (RR, 1.17; 95% CI, 1.10-1.24; P < 0.00001; I 2 = 22%; Figure 2A). After pooling the results from nine studies, no significant difference in mid-term cardiovascular death was observed (RR, 0.86; 95% CI, 0.71-1.03; P = 0.10; I 2 = 0%; Figure 2B). The risk of 1-year   heart failure rehospitalization was assessed in five studies using a random-effects model. As shown in Figure 2C, no significant difference was observed in heart failure rehospitalization (RR, 0.91; 95% CI, 0.58-1.44; P = 0.69; I 2 = 83%).

Long-Term (> 1 Year) Clinical Effects of Post-procedural Permanent Pacemaker Implantation After Transcatheter Aortic Valve Replacement
Long-term mortality between patients with and without PPI after TAVR was reported in 18 studies enrolling 39,172 patients with a mean follow-up period of 2.59 years. A random-effects model was applied, and patients with PPI after TAVR had a higher risk of all-cause mortality than those without PPI after TAVR (RR, 1.18; 95% CI, 1.09-1.28; P < 0.0001; I 2 = 57%; Figure 3A). However, there was no statistical difference in long-term risk of cardiovascular mortality between the two groups (RR, 1.15; 95% CI, 0.97-1.36; P = 0.11; I 2 = 59%; Figure 3B) after a mean followup of 2.12 years. Seven studies demonstrated a deleterious effect of PPI on heart failure rehospitalization after a mean follow-up of 2.16 years (RR, 1.42; 95% CI, 1.18-1.71; P = 0.0002; I 2 = 76%; Figure 3C).

Echocardiographic Effects of Post-procedural Permanent Pacemaker Implantation After Transcatheter Aortic Valve Replacement
Two studies reported LVEF both at baseline and 1-year followup. Figure 4A shows no significant difference in LVEF between the two groups at baseline (mean difference, 1.40; 95% CI, -0.13 to 2.93; P = 0.07; I 2 = 0%). LVEF at 1-year follow-up after TAVR was assessed using a fixed-effect model, and the overall value of LVEF was significantly greater in the no PPI group than in the PPI group (mean difference, -3.57; 95% CI, -4.88 to -2.26; P < 0.00001; I 2 = 0%; Figure 4B).

DISCUSSION
The results of this meta-analysis can be summarized as follows: (1) patients with post-procedural PPI show a higher risk of all-cause mortality at mid-term follow-up after TAVR; (2) post-TAVR PPI is associated with an increased risk of all-cause mortality and heart failure rehospitalization at long-term followup; and (3) post-procedural PPI adversely affect LVEF recovery on patients undergoing TAVR. Weferling et al. Twenty years after the first procedure in 2002 (55), TAVR has become a first-line treatment for patients with symptomatic severe aortic stenosis regardless of the estimated surgical risk (1)(2)(3)(4). Although TAVR technology has matured significantly over the years, conduction abnormalities remain one of the major complications to be resolved. Currently, there is insufficient evidence to support that the newer-generation devices could reduce the rate of post-procedural PPI (56,57). The underlying pathophysiological mechanisms compose of direct trauma, hemorrhage, inflammation, and ischemic injury of the conduction system during the expansion of the valve prosthesis (5). With accumulating TAVR cases, it is important to investigate the mid-to long-term clinical and echocardiographic outcomes of post-procedural PPI after TAVR.
Numerous studies have confirmed that RVP can negatively impact left ventricular function and increase the risk of the occurrence of atrial fibrillation (10,(58)(59)(60). The detrimental effects of RVP may elevate the risk of mortality and heart failure rehospitalization. As shown in our study, the pooled results revealed that patients undergoing PPI after TAVR had a higher risk of death at both mid-and long-term followup. They were also more likely to be hospitalized for heart failure during long-term follow-up. Similarly, a recent study containing the largest sample size reported that PPI after TAVR was independently associated with higher mortality and heart failure rehospitalization rate during follow-up, which was based on the entire France nationwide-level population (20).
We observed no significant difference in cardiovascular mortality and 1-year risk of heart failure rehospitalization between the two groups in our meta-analysis. The potential protective effects of PPI with respect to lethal bradyarrhythmias may counterbalance the negative effects of ventricular pacing. After the improvement of aortic stenosis, hemodynamic improvement of left ventricular function may compensate for the potential deleterious effects of ventricular pacing in such patients. In addition, implanting biventricular pacemakers in patients after TAVR may partially offset adverse effects linked to RVP.
Inconsistent with our results, few previous meta-analyses showed significant impacts of PPI after TAVR on clinical outcomes (13,61,62), except for a study by Faroux et al. (14), which was the first meta-analysis to reveal a significantly higher risk of all-cause death and heart failure rehospitalization in patients with PPI post-TAVR at 1-year follow-up. There are several explanations underlying the conflicting results in different studies. The small number of samples and short followup time may account for the distinct results. The occurrence and severity of pacing-induced cardiomyopathy are associated with ventricular pacing burden and duration, especially in patients with long-term pacing percentage ≥ 40% (11,63,64). Studies on TAVR have shown that new-onset conduction disturbances after TAVR may recover during follow-up, and about half of the patients requiring post-TAVR PPI are not pacing-dependent eventually (65)(66)(67). This may also partly explain why there was no significant difference in 1-year heart failure rehospitalization rates between the two groups.
Conduction disturbances occur commonly after TAVR, and an expert consensus algorithm was provided for managing post-TAVR conduction disturbances, but the optimal management of this complication is still unknown (68,69). Schoechlin et al. (18) compared patients' outcomes between different PPI implantation indications and revealed that the restrictive PPI strategy they adopted reduces the PPI rate significantly and is safe after a follow-up of 3 years. In consideration of the mid-to long-term negative effects demonstrated in our metaanalysis, we recommended adopting a relatively restrictive PPI strategy after TAVR, but the detailed indications and management need to be further explored. Furthermore, His-Purkinje system pacing (HPSP) allows for electrical stimulation signaling through the physiological conduction system, which  has the potential to prevent pacing-induced dyssynchrony, heart failure hospitalization, and mortality (70)(71)(72)(73). Previous studies have confirmed the feasibility and safety of HPSP in patients after TAVR. De Pooter et al. (74) found that the valve prosthesis can serve as an anatomical landmark for the implantation of the His-bundle lead. A multicenter study by Vijayaraman et al. (75) revealed that left bundle branch pacing had a higher success rate than His-bundle pacing after TAVR, with more ideal pacing parameters. Eleven patients with reduced left ventricular function who underwent HPSP successfully in this study showed significant LVEF improvement from 35 to 42% during followup. However, there is no systematic large-scale study evaluating the clinical and echocardiographic effects of HPSP in patients undergoing TAVR. Therefore, further studies are needed to focus on this area.

Limitations
Several limitations of our meta-analysis should be acknowledged. First, most studies included in our meta-analysis were retrospective observational studies. Thus, prospective, multicenter, randomized comparative studies are urgently needed. Second, TAVR technology has developed over time, and the types of valve prostheses are different. Patients included in the prior studies might have different PPI inclusion criteria compared with later ones so the heterogeneity among studies was relatively high in our study. Third, we had inadequate numbers of studies reporting ventricular pacing percentage to assess any significance of pacing-induced cardiomyopathy. We also do not have enough information to study other complications of PPI, such as infection, pneumothorax, and pocket hematoma, which may result in significant clinical consequences outside of mortality. Last but not least, our study is a meta-analysis, and we lack access to individual patient data which may provide more information.

CONCLUSION
The present meta-analysis provides evidence that post-TAVR PPI has negative clinical and echocardiographic effects at midto long-term follow-up. This study highlights the importance of identifying patients at high risk of developing conduction disturbances and requiring PPI after TAVR. Cardiologists should optimize treatment strategies and management of these patients.
TAVR technology should also improve to reduce the incidence of such complications.

DATA AVAILABILITY STATEMENT
The original contributions presented in this study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

AUTHOR CONTRIBUTIONS
SX, EZ, JS, and JZ contributed to the conception and designed the study. SX, EZ, and JS extracted the data and evaluated the quality. SX and EZ analyzed the data and wrote the manuscript. ZQ, FZ, YW, XH, and JZ critically reviewed and revised the manuscript. All authors have read and approved the final version of the manuscript.