Mechanical versus bioprosthetic valves for tricuspid valve replacement in Chinese patients: a systematic review and meta-analysis

Introduction

Tricuspid valve replacement (TVR) is the definitive surgical treatment for severe tricuspid valve disease that is not amenable to repair. Despite advances in perioperative care, TVR continues to be associated with substantially higher operative mortality and major postoperative complications compared with left-sided valve surgery (1,2). A central clinical dilemma in TVR is the choice of prosthesis: mechanical valves provide superior long-term durability but mandate lifelong anticoagulation, which carries inherent risks of thromboembolism and bleeding; bioprosthetic valves largely avoid the need for long-term anticoagulation but are susceptible to structural valve deterioration, particularly in younger individuals, potentially leading to reoperation (3,4).

Several international systematic reviews and meta-analyses have compared outcomes between these two valve types; however, the included populations are predominantly Western, and their conclusions remain inconsistent (5-7). Chinese patients presenting for TVR exhibit a distinct epidemiological profile, characterized by a higher prevalence of rheumatic heart disease and a younger age at operation relative to Western cohorts (8-10), and a relatively younger age at presentation (9,10). Moreover, perioperative management strategies and anticoagulation protocols differ considerably across medical centers in China (11). These factors limit the direct applicability of existing international evidence to the Chinese healthcare context. Although numerous single-center retrospective studies have been published domestically, a high-quality systematic review and meta-analysis that synthesizes these data is currently absent.

The objective of this systematic review and meta-analysis is to comprehensively evaluate and compare the perioperative and long-term outcomes of mechanical versus bioprosthetic valves in Chinese patients undergoing TVR, thereby providing an evidence-based reference to inform clinical decision-making by cardiac surgeons in China. We present this article in accordance with the PRISMA reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1215/rc) (12,13).

Methods

Registration

This systematic review was registered prospectively in the PROSPERO database (registration number CRD420261328669).

Eligibility criteria

Studies were considered eligible if they met the following criteria:

• Study designs: randomized controlled trials, prospective or retrospective cohort studies, and case-control studies comparing mechanical and bioprosthetic TVR. Case reports, narrative reviews, systematic reviews, conference abstracts, and animal experiments were excluded.
• Participants: Chinese patients (including individuals from China’s mainland, Hong Kong, Macau, and Taiwan) of any age who underwent TVR.
• Intervention and comparison: studies that directly compared TVR with a mechanical prosthesis versus a bioprosthesis. For the syntheses, studies were grouped according to the valve type received.
• Outcomes: studies were required to report at least one of the prespecified outcomes (see section “Data items”).
• Language: publications in Chinese or English.

Information sources

The following electronic databases were searched: PubMed, Embase, Cochrane Library, Web of Science, China National Knowledge Infrastructure (CNKI), Wanfang Data, VIP Database, Chinese Biomedical Literature Database (CBM), and the Chinese Medical Journal Database. The reference lists of all included studies were screened manually to identify additional potentially eligible reports.

Search strategy

The search strategy was developed using a combination of controlled vocabulary and free-text terms adapted for each database (see Appendix 1).

Selection process

Two reviewers independently screened the titles and abstracts of all retrieved records against the eligibility criteria. The full texts of potentially relevant reports were then obtained and assessed independently by the same two reviewers to determine final inclusion. Disagreements at either stage were resolved through discussion; when consensus could not be reached, a third reviewer adjudicated. The study selection process, including reasons for exclusion at the full-text stage, is documented in a PRISMA flow diagram (Figure 1). No automation tools were used.

Figure 1 Literature screening flowchart. CBM, Chinese Biomedical Literature Database; CNKI, China National Knowledge Infrastructure.

Data collection process

Data were extracted independently by two reviewers using a standardized, piloted data extraction form developed in WPS. The two sets of extracted data were cross-checked for accuracy, and any discrepancies were resolved by re-examining the source reports and through discussion. When data were missing or unclear, no attempt was made to contact the original study investigators; instead, studies were handled according to the assumptions described in section “Other variables”.

Data items

Outcomes

The following outcomes were sought from each included study. For each outcome domain, all reported time points and definitions were considered, and the most commonly reported measure across studies was selected for analysis.

• Primary outcomes: early mortality (defined as 30-day or in-hospital mortality), long-term survival (all-cause mortality over the entire follow-up period), and reoperation rate (any surgical re-intervention on the tricuspid prosthesis).
• Secondary outcomes: structural valve deterioration (as defined by each study), thromboembolic events (including valve thrombosis and systemic embolism), bleeding events (major bleeding or clinically relevant non-major bleeding attributable to anticoagulation), and permanent pacemaker implantation rate.

Other variables

Additional data items extracted comprised: first author, publication year, study hospital, operative time range, sample size (overall and by valve group), sex distribution, age, follow-up duration, and key baseline comorbidities (e.g., atrial fibrillation, prior cardiac surgery). When survival data were reported only as Kaplan-Meier curves without accompanying hazard ratios (HRs), the lnHR and its standard error were estimated using the methods described by Tierney et al. under the assumption of constant censoring (14,15). Missing or unclear information was treated as not reported, and the corresponding study was excluded from the specific analysis.

Study risk of bias assessment

Risk of bias for each included cohort study was assessed using the Newcastle-Ottawa Scale (NOS). Two reviewers independently rated each study across the domains of selection, comparability, and outcome. Disagreements were resolved through consensus. A NOS score of ≥7 was considered indicative of low risk of bias. All studies, irrespective of their risk of bias, were included in the primary syntheses. No automation tools were employed.

Effect measures

For dichotomous outcomes (early mortality, reoperation rate, structural valve deterioration, thromboembolic events, bleeding events, and pacemaker implantation), the effect measure was the risk ratio (RR) with 95% confidence interval (CI). For time-to-event data (long-term survival), the HR with 95% CI was used. When an HR was not reported directly, it was estimated from published Kaplan-Meier curves using the spreadsheet method of Tierney et al. (14,15).

Synthesis methods

Eligibility for synthesis

Studies were included in a meta-analysis only if they provided sufficient data for the outcome of interest. Studies judged to have overlapping patient populations from the same center were identified by comparing authors, institutions, and recruitment periods; only the most comprehensive or most recent report was retained for each outcome to avoid duplication of participants.

Data preparation

All analyses relied on published aggregate data. For survival outcomes, published Kaplan-Meier curves were digitized and survival probabilities at multiple time points were extracted to calculate lnHR and its variance. Studies for which lnHR could not be reliably estimated were excluded from that specific meta-analysis.

Tabulation and visual display

Study characteristics are presented in a summary table (Table 1). Results of individual studies and meta-analyses are visually displayed in forest plots generated with Review Manager 5.4.1.

Synthesis methodology

Meta-analysis was performed using Review Manager 5.4.1 (The Cochrane Collaboration). A random-effects model (DerSimonian and Laird method) was chosen a priori to account for anticipated clinical and methodological heterogeneity across studies. Statistical heterogeneity was quantified using the Chi-squared test (significance level α=0.10) and the I² statistic, with I²>50% indicating substantial heterogeneity. For outcomes where I² was <50% and the number of studies was small, a fixed-effect model was also considered; the primary results presented are from the random-effects model. When meta-analysis was not feasible (e.g., too few studies or insufficient data), results were summarized narratively.

Exploration of heterogeneity

Potential sources of heterogeneity were explored by inspecting clinical characteristics of the included studies (e.g., inclusion of redo-surgery patients only, variation in follow-up duration). When substantial heterogeneity was identified, sensitivity analyses excluding individual studies were performed to assess the influence of particular studies on the pooled estimate.

Sensitivity analyses

A pre-planned sensitivity analysis was performed by sequentially omitting each study to evaluate the robustness of the overall findings. If a single study was found to be a primary source of heterogeneity (e.g., I² decreasing substantially upon its removal), it was highlighted in the interpretation.

Reporting bias assessment

Publication bias was planned to be assessed using funnel plots and Egger’s regression test if 10 or more studies were available for an outcome. Because all meta-analyses included fewer than 10 studies, formal assessment was not undertaken. The potential for publication bias is therefore acknowledged, and both the Results and Discussion sections reflect this limitation.

Certainty assessment

No formal assessment of the certainty of the body of evidence (e.g., using the GRADE framework) was performed. Instead, limitations of the evidence are discussed narratively, with emphasis on study design, risk of bias, inconsistency, and imprecision.

Results

Study selection

The electronic search yielded 2,007 records. After removing duplicates, 1,524 records were screened based on title and abstract, of which 29 full-text articles were assessed for eligibility. Ultimately, 8 studies met all inclusion criteria and were included in the systematic review. No studies were identified that met most but not all inclusion criteria (i.e., ‘near-misses’) and merited specific citation; the majority of excluded full-texts were conference abstracts, reviews, studies with overlapping populations already represented in the final set, or investigations that did not report outcomes separately for mechanical and bioprosthetic valve groups. The study selection process is detailed in the PRISMA flow diagram (Figure 1).

Study characteristics

The eight included studies (2,10,16-21) were all retrospective cohort studies conducted in Chinese hospitals between 1993 and 2023. Together, they enrolled 1,504 patients undergoing TVR: 537 received a mechanical prosthesis and 967 received a bioprosthesis. Patient ages ranged from the early 40s to the mid-60s across cohorts, with a consistent predominance of rheumatic etiology. Follow-up duration extended from several months to 24 years in the largest studies. Detailed characteristics of each study, including sample size, sex distribution, age, and follow-up, are presented in Table 1.

Risk of bias in studies

Risk of bias was assessed with the NOS. Scores ranged from 6 to 9, indicating generally moderate-to-high methodological quality. Common limitations included a lack of comparability adjustment for important confounders and retrospective outcome assessment. Individual study ratings are available in Table 2.

Results of individual studies

For each outcome, the number of events per group and the corresponding effect estimate (RR or HR) derived from the individual studies are displayed in the forest plots (Figure 2). The results varied: several larger studies showed a trend toward lower early mortality with mechanical valves, whereas others reported similar mortality between groups; long-term survival data were heterogeneous, with unadjusted analyses of one study suggesting a survival advantage for mechanical valves that disappeared after propensity-score matching.

Figure 2 Forest plots for meta-analysis. (A) Early mortality comparison. (B) Long-term survival comparison. (C) Reoperation comparison. (D) Structural valve deterioration. (E) Thromboembolic events. (F) Bleeding events. (G) Pacemaker implantation. Bio, bioprosthetic valve; CI, confidence interval; IV, inverse-variance; M-H, Mantel-Haenszel; Mec, mechanical valve; SE, standard error.

Results of syntheses

Early mortality

Five studies (10,17,18,20,21) contributed to the meta-analysis of early mortality. These were conducted across diverse centers and had NOS scores of 6–9. The pooled RR for early mortality (mechanical vs. bioprosthetic) was 0.70 (95% CI: 0.48–1.02, P=0.06), with low statistical heterogeneity (I²=26%, P=0.25) (Figure 2A). There was no suggestion of excessive influence from any single study; removal of studies one at a time did not meaningfully change the summary estimate. In the included literature, Ke et al. was the only group with a P<0.05, but there was no significant difference between the two groups after propensity score matching (PSM) in their literature.

Long-term survival

Four studies (10,16,20,21) were included in the long-term survival analysis after exclusion of studies with overlapping populations or from which a HR could not be reliably extracted. The pooled HR was 1.02 (95% CI: 0.57–1.84, P=0.94), indicating no significant difference between valve types. Substantial heterogeneity was observed (I²=62%, P=0.03) (Figure 2B). Exploration of heterogeneity identified the study by Ke et al. [2025] (10) as the primary source; after its exclusion, the I² value decreased to 0%. This finding was confirmed by sensitivity analysis: when Ke et al. was omitted, the direction and precision of the pooled estimate remained consistent, demonstrating the robustness of the principal conclusion. A sensitivity analysis restricted to studies with long follow-up times [i.e., the two studies with the longest follow-up: Liu et al. (20), and Ke et al. (10)]. The pooled HR remained similar (HR 1.10, 95% CI: 0.43–2.79, P=0.84), indicating that the conclusion of no significant difference is robust to exclusion of shorter‑follow‑up studies.

Reoperation rate

Four studies (10,17,20,21) reported reoperation events and were pooled. The combined RR was 1.12 (95% CI: 0.61–2.05, P=0.72), with no evidence of statistical heterogeneity (I²=0%, P=0.55) (Figure 2C). Sensitivity analyses did not alter this null finding.

Structural valve deterioration

Four studies (2,17,19,20) provided data on structural valve deterioration. The summary RR was 1.08 (95% CI: 0.34–3.44, P=0.89), but significant heterogeneity was present (I²=78%, P=0.003) (Figure 2D). Because of the small number of studies, formal meta-regression or subgroup analyses to explore sources of heterogeneity could not be reliably performed; therefore, this pooled estimate should be interpreted with caution.

Thromboembolic events

Five studies (2,10,17,20,21) were combined. The mechanical valve group showed a statistically significant higher risk of thromboembolic events (RR 2.09, 95% CI: 1.06–4.13, P=0.03), with mild, non-significant heterogeneity (I²=31%, P=0.21) (Figure 2E). Sensitivity analyses confirmed the robustness of this association.

Bleeding events

Two studies (10,20) reported anticoagulation-related bleeding events. The pooled RR demonstrated a markedly increased risk in the mechanical valve group (RR 4.89, 95% CI: 3.12–7.72, P<0.00001), with no heterogeneity (I²=0%, P=0.90) (Figure 2F). The small number of studies precluded sensitivity analysis, but both studies individually showed a highly significant association.

Pacemaker implantation

Two studies (10,20) were included. No significant difference was observed between mechanical and bioprosthetic valves (RR 0.82, 95% CI: 0.44–1.53, P=0.54; I²=0%, P=0.71) (Figure 2G). The small number of studies precluded sensitivity analysis, but both studies individually showed a highly significant association.

Reporting biases

Because fewer than ten studies were available for any single outcome, formal assessment of publication bias via funnel plot asymmetry or Egger’s test was not performed. The possibility that small studies with negative or null findings may be underrepresented cannot be excluded, and the results, particularly those derived from only two studies, should be interpreted with this limitation in mind.

Certainty of evidence

A structured assessment of the certainty of the body of evidence (e.g., using the GRADE framework) was not undertaken. However, all synthesized evidence originates from retrospective observational studies, which inherently carry a risk of selection bias, and several analyses are based on a small number of events or studies, leading to imprecision. These factors are addressed further in the Discussion.

Discussion

General interpretation of the results in the context of other evidence

This systematic review and meta-analysis provides the most current, population-specific synthesis comparing mechanical and bioprosthetic valves for TVR in Chinese patients. The principal findings indicate that valve type does not significantly influence early mortality, long-term survival, or the need for reoperation. In the overall analysis, there was a marginal trend toward lower early mortality with mechanical valves (RR 0.70, P=0.06). Although Ke et al. (10) was the only group to report a P value below 0.05, their study found no significant difference between the two groups after propensity score matching. This strongly suggests that the original finding was driven by selection bias, whereby higher-risk patients were preferentially allocated to bioprostheses. These results align with international meta-analyses by Liu et al. (5), Cheng et al. (6), and Abdul Qadeer et al. (7), which similarly failed to demonstrate a survival or reoperation benefit of one prosthesis over the other. The absence of a survival advantage may reflect the fact that patients requiring TVR frequently present with advanced right ventricular dysfunction and multiorgan impairment, and the prognosis is often dominated by irreversible myocardial remodeling and comorbidities rather than by the type of prosthesis implanted (1,2,10).

In contrast, the risk of thromboembolic events was significantly more than twofold higher in recipients of mechanical valves, and anticoagulant-related bleeding events were nearly five times more frequent. These observations are consistent with international reports and likely reflect the unique low-pressure, low-flow hemodynamic environment of the right heart, which predisposes to prosthetic valve thrombosis (5). While valve choice does not appear to influence the incidence of permanent pacemaker implantation, the presence of a mechanical valve may complicate transvenous lead placement for cardiac implantable electronic devices (22). Furthermore, the substantial burden of lifelong anticoagulation—including stringent monitoring requirements and the risk of hemorrhage—was clearly demonstrated in the Chinese studies, reinforcing the notion that mechanical prostheses in the tricuspid position carry a particularly unfavorable safety profile compared with their use in left-sided positions. Notably, although bioprosthetic valves are inherently subject to structural valve deterioration, no significant difference in reoperation rates was observed, possibly due to limited follow-up in several studies and the reluctance to reoperate on high-risk patients. The emergence of transcatheter valve-in-valve procedures may further mitigate concerns about structural degeneration of bioprostheses in the future (23).

Limitations of the evidence included in the review

All eight studies were retrospective observational cohort studies, making them susceptible to selection bias and confounding by indication. Baseline characteristics, including age, rheumatic etiology, and concomitant procedures, differed appreciably among studies, and not all reports employed multivariable adjustment or propensity matching. The definitions of outcomes—especially structural valve deterioration and bleeding events—varied across studies, contributing to clinical heterogeneity. Follow-up duration was limited in some cohorts, which may lead to underestimation of long-term structural valve deterioration and reoperation. Finally, several outcome analyses were based on only two or four studies, limiting the precision and generalizability of the pooled estimates.

Limitations of the review processes

Several limitations of the review methodology should be acknowledged. First, the search was restricted to Chinese and English publications, which may have introduced language bias and excluded relevant studies reported in other languages. Second, no attempt was made to contact study authors for missing or unpublished data; HRs for long-term survival were estimated from published Kaplan-Meier curves using established methods, which rely on assumptions about censoring patterns (14,15). Third, the decision to exclude studies with overlapping populations or those from which effect estimates could not be reliably extracted may have introduced selection bias. Fourth, because fewer than ten studies were available for each outcome, formal assessment of publication bias could not be performed, and selective reporting cannot be excluded. Finally, a formal assessment of the certainty of evidence (e.g., using GRADE) was not conducted, and the narrative discussion of limitations does not replace a structured certainty rating. Furthermore, although we used HRs and performed sensitivity analyses to address variable follow‑up durations, these methods rely on the proportional hazards assumption, which we were unable to formally test due to lack of individual patient data. Residual confounding by follow‑up duration cannot be entirely excluded.

Implications for practice, policy, and future research

For clinical practice, these findings underscore that the choice between a mechanical and a bioprosthetic valve in the tricuspid position should be individualized, guided primarily by the patient’s anticipated adherence to anticoagulation, bleeding risk, life expectancy, and personal values. In Chinese patients with reliable access to high-quality anticoagulation management and good life expectancy, mechanical valves may remain a reasonable option; however, the significantly elevated thromboembolic and hemorrhagic risks must be clearly communicated. For the large subset of patients with poor anticoagulation adherence, high bleeding risk, or limited life expectancy, bioprosthetic valves likely represent a safer alternative. From a policy perspective, strengthening the infrastructure for anticoagulation monitoring and patient education in China may influence valve selection strategies and postoperative outcomes. Future research should prioritize prospective multicenter registries with extended follow-up, rigorous propensity-matched or risk-adjusted analyses, and standardized definitions of valve-related complications. The impact of evolving transcatheter techniques on the long-term management of bioprosthetic deterioration also warrants dedicated investigation in the Chinese population.

Conclusions

In Chinese patients undergoing TVR, overall early and long‑term survival as well as reoperation rates do not differ significantly between mechanical and bioprosthetic valves when analyzed across heterogeneous populations. However, this finding does not imply that valve type is unimportant for individual patients. Mechanical valves significantly increase the risk of postoperative thromboembolic and bleeding events. Therefore, clinical decision‑making should be highly individualized, taking into account patient age, life expectancy, anticipated adherence to anticoagulation therapy, and personal preferences. For Chinese patients with good life expectancy, reliable anticoagulation monitoring access, and high adherence, mechanical valves remain a viable option. Conversely, for patients with high bleeding risk, limited life expectancy, or inability to adhere to consistent anticoagulation, bioprosthetic valves may be a safer choice.

Acknowledgments

The authors thank Jing Wang for statistical guidance on handling variable follow‑up durations in the long‑term survival analysis.

Footnote

Reporting Checklist: The authors have completed the PRISMA reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1215/rc

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1215/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1215/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Filsoufi F, Anyanwu AC, Salzberg SP, et al. Long-term outcomes of tricuspid valve replacement in the current era. Ann Thorac Surg 2005;80:845-50. [Crossref] [PubMed]
2. Mao B, Sun L, Zhang J, et al. Perioperative factors associated with short- and long-term outcomes after tricuspid valve replacement. Interact Cardiovasc Thorac Surg 2016;23:845-50. [Crossref] [PubMed]
3. Rizzoli G, Vendramin I, Nesseris G, et al. Biological or mechanical prostheses in tricuspid position? A meta-analysis of intra-institutional results. Ann Thorac Surg 2004;77:1607-14.
4. Péterffy A, Szentkirályi I. Mechanical valves in tricuspid position: cause of thrombosis and prevention. Eur J Cardiothorac Surg 2001;19:735-6. [Crossref] [PubMed]
5. Liu P, Qiao WH, Sun FQ, et al. Should a Mechanical or Biological Prosthesis Be Used for a Tricuspid Valve Replacement? A Meta-Analysis. J Card Surg 2016;31:294-302.
6. Cheng Z, Fang T, Wang D, et al. Tricuspid Valve Replacement: Mechanical or Biological Prostheses? A Systematic Review and Meta-Analysis. Heart Surg Forum 2021;24:E209-14.
7. Abdul Qadeer M, Abdullah A, Noorani A, et al. Tricuspid valve replacement with mechanical versus biological prostheses: a systematic review and meta-analysis. J Cardiothorac Surg 2024;19:636. [Crossref] [PubMed]
8. Huang HL, Xie XJ, Lu C, et al. Long-term outcomes of mechanical and bioprosthetic valve replacement in the tricuspid position. Chinese Journal of Clinical Thoracic and Cardiovascular Surgery 2015;22:418-22.
9. Cho WC, Park CB, Kim JB, et al. Mechanical valve replacement versus bioprosthetic valve replacement in the tricuspid valve position. J Card Surg 2013;28:212-7. [Crossref] [PubMed]
10. Ke Y, Hua L, Pang S, et al. Long-term prognosis and risk factors in tricuspid valve replacement surgery: a single-center study. Front Surg 2025;12:1532945. [Crossref] [PubMed]
11. Li W, Fang WT, Kong Y, et al. Perioperative and mid-term clinical outcomes of tricuspid valve replacement. Shanghai Medical Journal 2017;40:759-63.
12. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Med 2009;6:e1000100. [Crossref] [PubMed]
13. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372: [Crossref] [PubMed]
14. Tierney JF, Stewart LA, Ghersi D, et al. Practical methods for incorporating summary time-to-event data into meta-analysis. Trials 2007;8:16. [Crossref] [PubMed]
15. Tierney JF, Burdett S, Fisher DJ. Practical methods for incorporating summary time-to-event data into meta-analysis: updated guidance. Syst Rev 2025;14:84. [Crossref] [PubMed]
16. Fang L, Li W, Zhang W, et al. Mid-term results and risks of isolated tricuspid valve reoperation following left-sided valve surgery. Eur J Cardiothorac Surg 2018;53:1034-9. [Crossref] [PubMed]
17. Liang W, Yue H, Li T, et al. The better substitute for tricuspid valve replacement in patients with severe isolated tricuspid regurgitation. Anatol J Cardiol 2019;22:172-6. [Crossref] [PubMed]
18. Hua K, Yang X, Zhou M, et al. Comparison of short-term and long-term survival rates in patients undergoing tricuspid valve replacement with mechanical versus bioprosthetic valves. Chinese Journal of Thoracic and Cardiovascular Surgery 2020;36:263-7.
19. Yang L, Yang YC, Yan Q, et al. Propensity score matching analysis of long-term follow-up results in 608 patients undergoing tricuspid valve replacement. Chinese Journal of Clinical Thoracic and Cardiovascular Surgery 2021;28:777-87.
20. Liu P, Xia DS, Qiao WH, et al. Which is the best prosthesis in an isolated or combined tricuspid valve replacement? Eur J Cardiothorac Surg 2021;59:170-9. [Crossref] [PubMed]
21. Yuan Y, Sun Y, Li N, et al. Long-Term Outcomes of Bioprosthetic and Mechanical Tricuspid Valve Replacement after Left-Sided Valves Surgery. Ann Thorac Cardiovasc Surg 2023;29:307-14. [Crossref] [PubMed]
22. Liew A, Wijesuriya N, Howell S, et al. Cardiac implantable electronic devices in the setting of tricuspid valve intervention: Risks, options, extraction and future perspectives: A state-of-the-art review. Heart Rhythm O2 2025;6:880-92. [Crossref] [PubMed]
23. Taramasso M, Gavazzoni M, Pozzoli A, et al. Tricuspid Regurgitation: Predicting the Need for Intervention, Procedural Success, and Recurrence of Disease. JACC Cardiovasc Imaging 2019;12:605-21. [Crossref] [PubMed]