predicted the LAAT resolution rate for VKA and DOAC (mean age of participants: β = −0.042, *p* = 0.343; male ratio: β = 0.248, *p* = 0.860).

**Figure 2.** (**A**) The forest plot of random-effect meta-analysis of the incidence of left arterial appendage thrombus formation under the use of vitamin K antagonist (VKA) and direct oral anticoagulants (DOACs). (**B**) The funnel plot of the studies included in the meta-analysis of the incidence of left arterial appendage thrombus formation under the use of vitamin K antagonist (VKA) and direct oral anticoagulants (DOACs).

**Figure 3.** (**A**) The forest plot of random-effect meta-analysis of the likelihood of resolution of left arterial appendage thrombus under the use of VKA and DOACs. (**B**) The funnel plot of the studies included in the meta-analysis of the likelihood of resolution of left arterial appendage thrombus under the use of VKA and DOACs.

#### **5. Discussion**

Various guidelines recommend DOAC as a preferable anticoagulant option to VKA for stroke prevention in AF. However, evidence on optimal anticoagulant selection in patients with LAAT is lacking. Although DOACs are safer than VKAs, their efficacy remains unverified. After analyzing pooled data from studies conducted in the past decade, we discovered that DOACs showed superiority over VKAs in increasing the likelihood of LAAT resolution and reducing the chances of LAAT development in high-risk patients. By monitoring patients taking anticoagulants, the incidence of thrombus formation in patients taking DOACs was lower than in those taking VKAs. In addition, patients with LAAT showed a higher thrombus resolution rate with DOACs use in comparison to VKAs. The superiority of the safety of DOAC compared with VKA is well established; these findings provide evidence for developing future treatment recommendations.

AF is associated with a high incidence of LAAT [45]. Previous reports have demonstrated that the CHADS2 score is an independent predictor of LAAT, and the prevalence of LAAT increases with the CHADS2 score [46,47]. However, this parameter does not have a significant skew; therefore, its exclusion from this meta-analysis did not affect the risk of LAAT and resolution obtained from each included study.

Clinically, when the efficacy of anticoagulants in LAAT resolution is comparable, safety concerns, including the risk of bleeding and drug–drug interactions, become the priority of medicine choice. DOACs have not yet been approved for patients with mechanical mitral valves, thrombus in locations other than the LAA, and antiphospholipid syndrome. However, a growing body of evidence has demonstrated that DOACs lead to fewer bleeding complications than VKAs. Therefore, DOACs are more likely to be selected for the prevention of thromboembolism events in people with LAAT.

Many large-scale clinical phase III trials have demonstrated that the efficacy of DOACs in preventing stroke is superior to VKAs and that DOACs have lower rates of bleeding. However, conclusive data on the recommended type and duration of anticoagulant use in LAAT is limited. Two recent meta-analyses have demonstrated that DOACs are as efficacious as and safer than VKAs in the treatment of LAAT in patients with nonvalvular AF. In addition, two ongoing prospective randomized trial registry studies are seeking to compare DOAC and VKA in patients with LAAT. One of these is a randomized control trial in China (NCT03792152), in which the effectiveness of rivaroxaban and VKA are compared. The other randomized control trial (RE-LATED AF (NCT02256683)) is a comparison of dabigatran and VKA in patients with nonvalvular AF of LAAT.

The strength of this study is its analyses of both the risk of LAAT development in high-risk patients and the likelihood of LAAT resolution with the use of VKAs and DOACs. The development of LAAT is an indicator of subsequent systemic thromboembolism events, which require anticoagulation therapy. However, the presence of LAAT requires emergent anticoagulation therapy until the thrombus resolves. The comparable efficacy of DOAC and VKA indicates an opportunity to provide patients with safer prescriptions. This study has several limitations. First, no subgroup analyses were performed for the different classes of DOAC in the 18 studies due to the small number of studies for each DOAC. Further studies are warranted to investigate the effectiveness of different DOACs in patients with different comorbidities. Second, several studies reported adverse effects, bleeding risk, and thromboembolism events, which prevented these events from being included in the meta-analysis. Third, the relative weight of the included studies varied in analyses, which is why the random-effects model was employed. Fourth, although the anti-IIa/Xa activity effectiveness and possible drug–drug interactions are crucial in the issues investigated in the current study, it is not feasible to conduct relevant analyses considering the very limited information provided in the included studies. Another related concern is the ratio/period of the effective therapeutic range, such as the international normalized ratio (INR) values in patients taking VKA. The distinctive ways of reporting INRs in the included studies also hinder further analyses from examining the underlying origins of heterogeneity.

#### **6. Conclusions**

This meta-analysis of observation data revealed significant differences in LAAT development in high-risk patients and the likelihood of LAAT resolution in patients treated with DOAC or VKA. With respect to safety profiles, DOACs are preferable to VKAs in patients with LAAT and without absolute DOAC contraindication. As the role of DOACs expands, further studies should be conducted to provide clinicians with a practical reference for optimization of the selection of appropriate DOACs and the duration of treatment.

**Supplementary Materials:** The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/jcdd9100351/s1, Table S1: Search history and results.

**Author Contributions:** Conceptualization, Y.-Y.C., S.T. and L.C.; methodology, S.T., C.-C.Y. and L.C.; software, C.-C.Y.; validation, C.-C.Y. and L.C.; formal analysis, S.T., C.-T.H., C.-C.Y. and L.C.; investigation, S.T. and C.-T.H.; resources, S.T. and C.-T.H.; data curation, Y.-Y.C., S.T. and C.-C.Y.; writing—original draft preparation, Y.-Y.C. and S.T.; writing—review and editing, C.-T.H., C.-C.Y. and L.C.; visualization, S.T. and C.-C.Y.; supervision, C.-T.H.; project administration, S.T., C.-T.H. and L.C. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Ethical review and approval were waived for this study due to the nature of this meta-analytic study. All the included studies have been approved by their IRBs.

**Informed Consent Statement:** Patient consent was waived due to the nature of this meta-analytic study. No identifiable personal information was obtained in the included studies.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding authors.

**Conflicts of Interest:** The authors report no conflict of interest.
