Comparison of efficacy and safety between VKAs and DOACs in patients with atrial fibrillation after transcatheter aortic valve replacement: A systematic review and meta‐analysis

Abstract In the past decade, direct oral anticoagulants (DOACs) have proven to be the best option for patients with nonvalvular atrial fibrillation. Nevertheless, evidence for the use of DOACs for anticoagulation in valvular atrial fibrillation, particularly after aortic valve replacement, remains inadequate. Thus, we conducted a meta‐analysis to compare the efficacy and safety of vitamin K antagonists (VKAs) and DOACs in patients with atrial fibrillation after transcatheter aortic valve replacement (TAVR). We conducted a comprehensive search of online databases, and 11 studies were included in the final analysis. The primary endpoint was all‐cause mortality. Secondary endpoints included stroke and cardiovascular death. The safe endpoint is major and/or life‐threatening bleeding. Subgroup analysis was conducted according to the different follow‐up time of each study. Random‐effects models were used for all outcomes. Statistical heterogeneity was assessed using χ 2 tests and quantified using I 2 statistics. Patients in the DOACs group had a significantly lower risk of all‐cause mortality compared with patients in the VKAs group (relative risk [RR]: 1.20, 95% confidence interval [CI]: 1.01–1.43, p = .04). This benefit may be greater with longer follow‐up. In a subgroup analysis based on the length of follow‐up, a significantly lower risk of all‐cause mortality was found in the DOACs group in the subgroup with a follow‐up time of >12 months (RR: 1.50, 95% CI: 1.07–2.09, p = .001). There were no significant differences between the two groups in cardiovascular death, stroke, and major and/or life‐threatening bleeding. For patients with atrial fibrillation after TAVR, the use of DOACs may be superior to VKAs, and the benefit may be greater with longer follow‐up. The anticoagulant strategy for atrial fibrillation after TAVR is a valuable direction for future research.


| INTRODUCTION
As the population ages, the incidence, prevalence, and mortality of aortic valve disease, particularly calcific aortic valve disease, has risen substantially, contributing significantly to the disease burden among the elderly. 1,2 Patients with mild aortic stenosis (AS) may remain asymptomatic for many years. 3 Once severe aortic stenosis develops, the symptoms and condition deteriorate, and conservative medical treatment tends to have a poor prognosis, quality of life, and longterm survival unless surgery or intervention is performed. 4,5 Surgical aortic valve replacement (SAVR) is a traditional treatment for AS, but it is characterized by high surgical trauma and high surgical risk.
Structural cardiac interventions have advanced by leaps and bounds in recent years, with transcatheter aortic valve replacement (TAVR) emerging as an alternative treatment for patients with symptomatic AS, inoperable aortic valve stenosis, or high risk of SAVR. 6 With the gradual maturity of TAVR therapy, the therapeutic effect has been significantly improved and the indications are becoming more and more extensive. For symptomatic severe AS patients with low surgical risk, there was no significant difference in all-cause mortality, stroke or myocardial infarction, or prosthetic valve failure in patients after TAVR compared to those who received SAVR. 7 Prior and new atrial fibrillation (AF) is common in patients with severe AS receiving TAVR, and AF is associated with increased mortality and adverse ischemic and bleeding events. 8 Warfarin is the main anticoagulant therapy for stroke prevention in patients with AF, but it has disadvantages such as narrow treatment window, variable dose-response, interaction with drugs and food, and the need for international normalized ratio (INR) detection. 9,10 Randomized clinical trials have demonstrated that DOAC is as good or better than warfarin for antithrombotic therapy in patients with AF. [11][12][13][14] The guidelines recommend DOACs as the first choice for anticoagulant therapy in patients with DOACs indications. 15 DOACs are increasingly being used in place of warfarin, but the evidence for their effectiveness and safety in patients with valvular AF remains limited. 9

| Inclusion and exclusion criteria
Studies were included in our meta-analysis when the following criteria were met: (1) The study is limited to English and human subjects. (2) The study compared the efficacy or safety of VKAs and DOACs in patients with AF after TAVR. (3) At least one of these outcomes was reported in the study: all-cause mortality, death from cardiovascular causes, stroke, and major and/or life-threatening bleeding.

| Data extraction and endpoints
All data were independently extracted from text, tables, and graphs by two authors (J. Y. and M. L.). Disagreements among reviewers were resolved through discussion to reach a consensus. The primary endpoint of this meta-analysis was all-cause mortality (death from all causes). Secondary endpoints were stroke (ischemic stroke) and cardiovascular death (death from cardiovascular causes). The safety endpoint is major and/or life-threatening bleeding. Scale (NOS). 21 The comprehensive effect used relative risk (RR), with a confidence interval (CI) of 95%. A two-sided p value was used, and p < .05 was considered significant. Random-effects models were used for all outcomes. Statistical heterogeneity was assessed using the chisquared test (p < .10 was considered statistically significant for heterogeneity) and was quantified using the I 2 statistic. Subgroup analysis by follow-up time was performed to further analyze the statistical results and explore possible sources of heterogeneity. The publication bias was tested by Egger regression to test the asymmetry of the funnel chart. We conducted sensitivity analysis by eliminating each included study one by one, looking for potential sources of heterogeneity. The above data were analyzed using Review Manager  Table 1. The RCTs were evaluated as high quality (Supporting Information: Figure 1), the CCT had a global ideal score being 21 (>16) (Supporting Information: Table 1), and all cohort studies were considered of high quality because of the scores ranging from 7 to 9, with an average of 7.875 (Supporting Information: Table 2). We

| Safety endpoint
There was no significant difference in major and/or life-threatening between the VKAs and DOACs groups (RR: 1.03, 95% CI: 0.84-1.25, p = .79) ( Figure 3A), with mild statistical heterogeneity (I 2 = 24.11%, p = .469). After excluding each study, the results remained stable.  Heart Disease, which recommends that patients without oral anticoagulant indications after TAVR should be given lifelong single-drug antiplatelet therapy, and patients with oral anticoagulant indications should be given lifelong oral anticoagulant therapy. 16 TAVR is mainly applied to elderly high-risk patients, who are not only at high risk of thrombosis but also at high risk of bleeding. The incidence of thrombosis and bleeding events is closely related to the prognosis, and postoperative ischemic and hemorrhagic complications are very common. 31 Many patients have AF before TAVR, which may be related to the cardiovascular pathophysiological conditions of elderly patients, such as atrial fibrosis and left atrial diameter enlargement. 32 Meanwhile, new-onset AF after TAVR is also common, which may be related to the operation itself. AF increases the risk of postoperative cardiogenic embolic events and increases the incidence of cardiovascular adverse events, cerebrovascular events, and mortality. 33 In addition, bleeding events are also important events that should be widely concerned, which increases the difficulty of clinical antithrombotic therapy. Some risk factors, such as old age, frailty, falls, kidney disease, liver disease, malignant tumors, coagulation disorders, and antithrombotic therapy, may increase the risk of bleeding, which affects the patient's prognosis. 34 Our study found that DOACs reduced the risk of all-cause death in patients with AF after TAVR. In addition, subgroup analysis based on different follow-up times showed that patients with longer followup times benefited more significantly. The biggest difference between DOACs and VKAs is that DOACs can only inhibit one step in the coagulation process, while VKAs can prohibit multiple steps.

| DISCUSSION
Vitamin K is a cofactor of activation of coagulation factors ⅱ, ⅶ, ⅸ, ⅹ, and VKAs can reduce the synthesis of vitamin K-dependent coagulation factors. 35  investigated the feasibility of DOACs as an antithrombotic after TAVR compared to standard regimens. 17 The ATLANTIS study showed that for all patients undergoing TAVR with or without an indication for oral anticoagulation, the efficacy of apixaban after TAVR was not superior to the current standard antithrombotic regimen in terms of net clinical benefit. The ENVISAGE-TAVI AF study was designed to investigate the efficacy of edoxaban in patients with recurrent or new-onset AF after TAVR. 18 The primary endpoint event rate was not inferior to the VKAs group in the edoxaban group, but the incidence of major bleeding was higher than in the VKAs group, mainly due to more gastrointestinal bleeding in the edoxaban group. Compared with the previous meta-analysis, we concluded that anticoagulation with DOACs in patients with TAVR combined with AF may be superior to VKAs, but we need more evidence to prove our point in the future, and the quest for the best antithrombotic treatment option after TAVR will continue.
Our meta-analysis has several limitations. First, our study included two RCT, one CCT, and eight cohort studies. Although we adopted some quality evaluation methods, bias will inevitably occur, and more RCTs will be needed in the future to verify our conclusions.
Second, some studies used antiplatelet drugs combined with anticoagulant therapy, while others did not. Due to limited data, we did not conduct a separate subgroup analysis. Also, we did not conduct a subgroup analysis of the patient population, so the heterogeneity of the patient population should be considered when interpreting the study results. Finally, due to the lack of data, we did not conduct a subgroup analysis of the types of DOACs and could not evaluate the individual category effect of each DOAC.

| CONCLUSION
For patients with AF after TAVR, the use of DOACs may be superior to VKAs, and the benefit may be greater with longer follow-up. The anticoagulant strategy for AF after TAVR is a valuable direction for future research.