Search criteria

Data collection

To identify relevant studies, the following international databases were searched: PubMed, Embase, Web of Science, Scopus, and Cochrane Library. The search encompassed scientific literature on antithrombotic therapy following TMVR from January 2015 to January 2025. The search strategy involved the use of Medical Subject Headings terms and relevant keywords, including: ‘transcatheter mitral valve repair’, ‘TMVR’, ‘MitraClip’, ‘edge-to-edge repair’, ‘anticoagulation’, ‘antiplatelet’, ‘direct oral anticoagulants’, ‘vitamin K antagonists’, ‘stroke’, ‘thromboembolism’, ‘bleeding complications’, ‘hemorrhage’, and ‘mortality’.

Two independent reviewers independently screened the titles and abstracts of all the identified studies. Studies that did not meet the inclusion criteria were excluded. The full texts of potentially relevant studies were retrieved and assessed according to the inclusion and exclusion criteria. Any disagreements between reviewers were resolved through discussion.

Data extraction

A standardized data extraction form was developed, and the following information was extracted from each study. The study’s characteristics included authors, year of publication, study design, sample size, and intervention summary. Information on the antithrombotic regimen was also gathered, including the type of therapy (such as SAPT, DAPT, or OAC), the specific agent used (e.g., aspirin, clopidogrel, warfarin, or DOACs), dosage, and duration. The outcomes were recorded, including thromboembolic events (such as stroke, transient ischemic attack, and systemic embolism), bleeding complications (both major and minor), mortality, cardiovascular events, and hospitalization for heart failure. Finally, the follow-up duration or length of time the patients were monitored after the TMVR procedure was noted. Two reviewers independently performed data extraction, and any discrepancies were resolved through discussion.

Quality assessment

The methodological quality of the included studies was assessed using appropriate tools based on study design. The Cochrane risk of bias tool was used to assess the risk of bias in RCTs. This tool evaluates bias across several domains: selection, performance, detection, attrition, and reporting biases. The Newcastle-Ottawa Scale (NOS) was used to assess the quality of observational studies. NOS evaluates studies based on selection, comparability, and outcome assessment.

Data synthesis

Owing to the anticipated heterogeneity of study designs, patient populations, TMVR procedures, and antithrombotic regimens, a quantitative synthesis (meta-analysis) may not be appropriate. Instead, narrative synthesis was performed to summarize and synthesize the findings of the included studies. This involved organizing the data into meaningful categories and identifying key themes and patterns.

Study selection and characteristics

A comprehensive literature search was conducted across five major electronic databases: PubMed/MEDLINE, Embase, Cochrane Library, Scopus, and Web of Science, and 1,677 records were identified. After removing duplicates, 1,252 records were screened. Reports assessed for eligibility (n = 26). Finally, 15 studies were included in the study. These include various study types, such as prospective cohort studies, retrospective studies, RCTs, and registry-based observational studies. The PRISMA flow diagram (Figure 1) illustrates the study selection process.

Display full size

Figure 1. PRISMA-2020 flow diagram. PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses. Adapted from [13]. © The Author(s) 2021. Licensed under a Creative Commons Attribution (CC BY 4.0).

The total number of patients across all studies was 20,956, with individual study sample sizes ranging from 11 to 11,267 participants in the most extensive study by Mentias et al. [14], which analyzed a Medicare database. The patient population primarily consisted of individuals with severe MR or specific valve conditions. The mean age of participants was generally in their 70s and 80s, with the sex distribution varying across studies. The included studies evaluated a range of transcatheter mitral interventions, including TEER devices (e.g., MitraClip) and transcatheter mitral valve replacement devices (e.g., Tendyne and Tiara systems). The follow-up period ranged from 30 days to 4 years, depending on the study. A detailed overview of the study’s characteristics is provided in Table 1.

Antithrombotic therapy comparison

Across the studies, different antithrombotic therapy regimens were used, including DOACs, VKAs, and DAPT, etc. Several studies compared these therapies regarding bleeding risk, stroke, and mortality, providing insightful findings.

Four studies [14–17] compared DOACs with VKAs and found that DOACs were associated with a significantly lower bleeding risk [hazard ratio (HR): 0.21, p = 0.02] than VKAs, with similar stroke and mortality outcomes across both groups. Specifically, Mentias et al. [14] demonstrated that DOACs led to reduced mortality (HR: 0.67) and lower major bleeding rates compared to VKAs. Similarly, Guerrero et al. [16] reported a higher bleeding risk with VKAs but no significant differences in stroke or mortality outcomes, confirming the safety benefits of DOACs.

A recent single-center observational study by Schipper et al. [27] found no significant difference between DOAC and VKA for the composite of ischemic and bleeding events. Notably, overall mortality was higher in the VKA group (HR 2.56, p = 0.002) [27], suggesting a possible advantage of DOACs. Schipper et al. [27] conclude that DOACs may be a safe alternative but emphasize that randomized trials are needed.

In a similar analysis of transcatheter valve patients, Alaour et al. [28] examined DOAC vs. VKA after transcatheter aortic valve replacement (TAVR) in a large registry (n = 2,908) and found no difference in net clinical benefit or bleeding at 1–5 years. Consistent with Schipper et al. [27], they observed higher long-term mortality with VKAs (HR = 1.25 at 5 years). These findings across transcatheter valve cohorts suggest DOACs are at least comparable to VKAs, although all data remain observational.

In addition, Paukovitsch et al. [19] and Guerrero et al. [16] explored the efficacy of triple therapy (OAC + DAPT) in patients post-TMVR. Although triple therapy resulted in a higher risk of bleeding (24%), it did not provide additional protection against thromboembolic events or reduce mortality. This suggests that triple therapy carries substantial bleeding risks without clear benefits compared to DOAC/DAPT alone.

Several studies [19–22] examined DAPT and found it associated with a higher bleeding rate than SAPT. However, DAPT did not offer any additional benefit for stroke prevention, highlighting the need for careful consideration of bleeding risks when prescribing DAPT. For a detailed comparison of antithrombotic therapies, see Table 2.

Bleeding complications

Four studies [15, 18–20] reported bleeding complications, revealing significant differences based on the anticoagulation regimen. Patients receiving triple therapy or DAPT had the highest rates of bleeding, whereas those treated with DOACs had significantly lower rates of major bleeding.

Two studies [14, 15] confirmed that DOACs were associated with a lower incidence of major bleeding events than VKAs without increasing the risk of stroke or mortality. Specifically, El Bèze et al. [15] reported that DOACs reduced bleeding risk by 79% (9% vs. 40% with VKAs), while Mentias et al. [14] found that DOACs were associated with a lower risk of major bleeding (HR: 0.79) compared to VKAs. Additionally, Paukovitsch et al. [19] found that combining DOACs with DAPT reduced bleeding complications without compromising thromboembolic protection, making it a safer alternative to triple therapy (Table 2).

Mortality and stroke outcomes

Four studies [14, 22, 23, 26] reported mortality outcomes, indicating that DOACs are associated with reduced mortality after TMVR. Mentias et al. [14] and Malaisrie et al. [23] found that patients treated with DOACs had significantly lower mortality rates than those treated with VKAs. Specifically, Mentias et al. [14] reported a HR of 0.67 for mortality with DOACs compared to VKAs, highlighting the superior survival benefits of DOACs in the post-TMVR population.

Interestingly, Tichelbäcker et al. [26] observed a 1.1% incidence of left ventricular thrombus formation even in patients treated with DOACs, underscoring the need for vigilant monitoring of thromboembolic events despite effective anticoagulation therapy. This finding suggests that while DOACs are effective in reducing bleeding and mortality, they may not eliminate the risk of thrombotic events in high-risk patients (Table 2).

Quality of life and functional status

Regarding quality of life (QoL), several studies reported significant improvements in patients’ New York Heart Association class and Kansas City Cardiomyopathy Questionnaire scores following TMVR. Two studies [22, 23] demonstrated improvements in functional capacity, particularly in heart failure symptoms, after undergoing TMVR. These findings suggest that TMVR improves mortality and complications and significantly enhances patients’ overall well-being.

Furthermore, Geis et al. [22] reported a reduction in ventricular arrhythmias and a decreased need for implantable cardioverter defibrillator therapies following the procedure, highlighting the potential of TMVR to improve cardiac rhythm control and reduce the arrhythmic burden. These improvements in functional status and QoL are critical for patients with severe MR, as they often experience significant limitations in daily activities due to heart failure symptoms. For more details, see Table 3.

Procedural success and device-specific outcomes

Two studies [17, 25] found that MitraClip implantation resulted in high success rates of MR reduction, with procedural success rates exceeding 95% in most cases. However, Tendyne and Tiara devices showed more challenges in procedural outcomes, including device migration and valve leakage, suggesting that some devices may carry a higher risk of complications, potentially affecting long-term results.

Moreover, Nathan et al. [20] investigated anticoagulation strategies after TMVR and found that VKAs and DOACs were associated with a reduced risk of thromboembolic events. However, the optimal anticoagulation strategy after TMVR remains a topic of ongoing debate. These findings highlight the importance of tailoring anticoagulation therapy to individual patient needs, particularly in high-risk populations (Table 3).

Quality assessment

The risk of bias across the 15 studies was assessed using two different tools: the Cochrane risk of bias tool for RCTs and the NOS for other non-randomized studies. This RCT demonstrated a low risk of bias in all key domains, including random sequence generation, allocation concealment, and blinding, making it a high-quality study. The remaining 15 non-randomized studies were assessed using the NOS, where most of the studies scored six stars, indicating moderate quality. These studies provided valuable data on treatment outcomes, complications, and effectiveness but were limited by their observational nature and the absence of control groups, which introduced potential biases due to confounding factors. Overall, an RCT had the lowest risk of bias, while the non-randomized studies, despite their limitations, provided important evidence on the effectiveness and safety of various interventions (Table 4).

Heterogeneity of TMVR procedures and patients

The challenges of antithrombotic management following TMVR are compounded by the heterogeneous nature of the patient population and the multitude of TMVR procedures and devices. TMVR is performed in patients with various clinical profiles, including degenerative mitral valve disease, functional MR due to heart failure, and ≤ 50% of surgical candidates [4]. These patients are often elderly and have several comorbidities [6, 9], including AF, chronic kidney disease, and a history of bleeding, making the decision-making process even more challenging. The increasing complexities and comorbidities highlight the need for personalized risk assessments and antithrombotic strategies. Therefore, a comprehensive review of the existing data is required to inform clinical decision-making and optimize outcomes for this high-risk group.

Heterogeneity of procedures and implications for therapy

A key finding of this review is the critical distinction between TEER and replacement. TEER with the MitraClip involves native valve tissue with a lower intrinsic thrombotic risk; thus, antithrombotic therapy is often dictated by concomitant indications like AF [14, 18]. In contrast, TMV replacement implants a prosthetic valve, carrying a higher risk of valve thrombosis analogous to surgical bioprostheses or TAVR, typically mandating at least temporary OAC [15, 28]. Similarly, valve-in-valve or valve-in-ring procedures have unique risk profiles influenced by the underlying substrate [16]. Pooling these populations is problematic, and our stratified synthesis highlights the need for procedure-specific guidelines.

Comparison of DOACs and VKAs

Evidence suggests that DOACs also demonstrate a superior safety profile regarding bleeding prevention compared to VKAs. Nonetheless, considering its efficacy against thromboembolic events and safety profile, the best antithrombotic regimen in TMVR patients is still an essential choice.

DOACs have replaced VKAs. They are preferentially chosen because of their ease of use, fewer dietary restrictions, and lower risk of drug interactions, thus facilitating management and improving patient compliance [15, 16]. Using a recent analysis of the National Inpatient Sample national dataset between 2016 and 2020, malignancy did not increase mortality, length of stay, or procedural costs for both TMVR and TEER [29]. However, the number of bleeding events was lower in the DOAC arm, suggesting a possible survival benefit with newer anticoagulants. This is especially important in the older population, who often undergo TMVR procedures, where bleeding complications result in worse outcomes and increased mortality. In addition, the study by Mentias et al. [14] confirmed these findings, demonstrating lower mortality rates in DOACs versus VKAs patients, potentially due to lower bleeding risk. Schipper et al. [27] reported no difference in stroke or bleeding between DOAC and VKA after mitral valve-TEER, but higher mortality with VKA. Likewise, Alaour et al. [28] found equivalent composite outcomes after TAVR under DOAC vs. VKA (with higher long-term mortality on VKAs).

However, a recent meta-analysis by Zorman et al. [29] has indicated that VKAs may be more effective in preventing bioprosthetic mitral valve thrombosis than their safety. VKAs, with their more exhaustive coverage of anticoagulation, are likely to yield better protection against valve thrombosis than DOACs, which inhibit only particular clotting factors. Given these contradictory observations, a more sophisticated view must consider the prospect of both subclinical and clinical thromboses.

Antiplatelet therapy: DAPT vs. SAPT

The role of antiplatelet therapy, particularly DAPT, in addition to anticoagulation after TMVR, remains a matter of debate. Although DAPT is frequently used to reduce device-related thrombosis risk compared with SAPT, several studies have demonstrated that DAPT is associated with an increased risk of bleeding complications without a significant protective effect against stroke [30, 31].

DAPT was implemented to prevent device-related thrombosis but was consistently associated with increased bleeding. Data from studies such as POPular transcatheter aortic valve implantation (TAVI) [32] and those by Mentias et al. [14] demonstrated that although DAPT has the potential to reduce thrombosis in the device profile, its risk of bleeding outweighs any theoretical benefit that may be observed compared with single-agent therapy (acetylsalicylic acid alone). The POPular TAVI trial [33] demonstrated that DAPT following TAVR did not reduce stroke outcomes but was associated with a higher rate of bleeding than SAPT. Based on these results, the current recommendation for the management of post-TMVR patients leans towards SAPT as the standard regimen, and only patients with additional cardiovascular risk factors (e.g., recent myocardial infarction, stent implantation) should receive DAPT.

Triple therapy: risks and benefits

In post-TMVR patients with high thromboembolic event risk, such as those with AF, triple therapy (OAC plus DAPT) is occasionally employed. However, multiple studies, such as those conducted by Paukovitsch et al. [19] and Zorman et al. [29], have cast some doubt on the high bleeding risk associated with triple therapy with no concomitant reduction in thromboembolic events. Given the relative lack of benefits and increased bleeding risk, this approach should be limited to the highest-risk patients and for the minimum time possible [30].

Subclinical thrombosis: an emerging concern and the role of imaging in diagnosis

In the post-TMVR care landscape, subclinical valve thrombosis, presenting as leaflet thickening and reduced leaflet motion on imaging, has surfaced as an important consideration. Even if asymptomatic, it may cause device dysfunction and systemic embolization, affecting long-term outcomes [16, 29].

Diagnosing and treating subclinical valve thrombosis are challenging; otherwise, they can lead to clinical events. Recent studies have shown that this augments transvalvular gradients and accelerates valve degeneration, requiring repeated interventions [34]. The early detection and treatment of subclinical thrombosis may enhance long-term valve durability and reduce adverse events [35].

Subclinical thrombosis detection often relies on the use of advanced imaging modalities. Transthoracic echocardiography (TTE) is the standard initial imaging modality; however, it has limited sensitivity for detecting subtle leaflet abnormalities. High-resolution transesophageal echocardiography (TEE) combined with four-dimensional computed tomography (4D-CT) has also been integrated into pathways for early leaflet thickening and decreased leaflet mobility [36]. For example, four-dimensional CT provides enhanced spatial resolution and the capability to quantify leaflet motion to enhance the detection of subclinical thrombosis. Specifically, four-dimensional CT allows evaluating various structural features to improve device assessment.

Stroke and mortality with antithrombotic strategies

The antithrombotic regimen chosen significantly influenced both stroke and mortality outcomes in patients with TMVR. Stroke and mortality rates differ among anticoagulation strategies, and this quality of evidence underlines the necessity of refining treatment goals. Mentias et al. [14]. demonstrated that patients on DOACs experienced significantly lower mortality rates than their VKA counterparts, suggesting a potential survival advantage with newer anticoagulants. This can be ascribed to the lower risk of bleeding associated with DOACs, as bleeding events have been shown to predict all-cause mortality in this cohort strongly.

However, DOACs may increase asymptomatic thromboembolism, which could offset some of their advantages in stroke prevention. Tomšič et al. [37] noted the importance of weighing the thromboembolic prevention benefits of DOACs against the subclinical thrombosis risk. Patient selection and monitoring are critical for optimizing the outcomes, highlighting the inherent limitations of this study.

Approach to antithrombotic therapy

The complex balance of thrombotic and bleeding risks in patients undergoing TMVR makes it almost certain that a one-size-fits-all approach to antithrombotic therapy will not be optimal. Clinicians should adopt a tailored approach that addresses patient-specific traits, procedural considerations, and device-specific properties.

Careful pretreatment risk stratification and regimen tailoring to the individual patient profile are crucial for the successful management of antithrombotic therapy. This involves assessing patient-specific risk factors, such as age, comorbidities (e.g., AF, chronic kidney disease, and previous bleeding), frailty, and concomitant medications. Additionally, procedural considerations should be considered, such as the type of TMVR procedure performed (e.g., valve-in-valve, valve-in-ring, and valve-in-native annulus), access site, and procedural success [38–40].

Equally important are device-specific characteristics, including valve design, material, and implantation techniques. These factors influence the risk of thromboembolic events and bleeding; therefore, the choice of antithrombotic therapy must be personalized based on a comprehensive assessment of these variables.

This approach includes risk assessment tools, such as the HAS-BLED score to assess bleeding risk and the CHA2DS2-VASc score to assess stroke risk, which can assist in rational decision-making [41, 42]. These tools can help assess each patient’s risk and guide the most appropriate antithrombotic strategy choice. Nonetheless, it is important to note that these scores have not yet been validated in the TMVR population. They should be used with clinical judgment and considering individual patient-specific variables.

Current guidelines & future direction

Despite the growing body of literature, there are multiple unanswered questions regarding antithrombotic therapy following TMVR. The rarity of substantial RCTs comparing various antithrombotic strategies has made it difficult for many clinical providers to base their decisions on high-level evidence. Additionally, current guidelines are often not tailored to patients undergoing TMVR, resulting in diverse practices and possibly inadequate outcomes [7].

Limitations

This review has several limitations. First, the studies included are highly heterogeneous, encompassing both transcatheter repair and replacement procedures with fundamentally different thrombotic risks. Although we have narratively synthesized findings by procedure type where possible, the inability to perform a quantitative meta-analysis due to this heterogeneity limits the strength of pooled conclusions. Second, all but one study was observational, susceptible to unmeasured confounding and selection bias. The reported HRs are derived from individual studies, not from a pooled meta-analysis. Third, definitions for bleeding and thromboembolic outcomes varied (e.g., BARC, Mitral Valve Academic Research Consortium).

Conclusion

The ideal antithrombotic regimen after TMVR remains an active area of investigation. Although DOACs have a better safety profile with a lower risk of bleeding and greater convenience for the patient, VKAs remain the drug of choice for thromboembolic prevention in high-risk populations. The roles of DAPT and triple therapy remain debatable, with data indicating that these regimens should be reserved for carefully selected patients. Subclinical valve thrombosis, the most recent complication of TMVR, requires attention and imaging after device implantation. This highlights the principle of a personalized approach to antithrombotic management that focuses on the particularities of individual patient characteristics, procedural factors, and device properties. Filling these gaps will allow clinicians to make informed decisions and improve the safety and effectiveness of TMVR. It can also be useful for highlighting the emerging biomarker image modalities, such as transesophageal echocardiography for subclinical thrombosis.