(P182) DIRECT ORAL ANTICOAGULANTS VERSUS VITAMIN K ANTAGONISTS IN THE FIRST 3 MONTHS AFTER BIOPROSTHETIC VALVE REPLACEMENT: A SYSTEMATIC REVIEW AND META-ANALYSIS

Background: Guidelines recommend vitamin K antagonists (VKAs) in patients who require anticoagulation in the first three months after bioprosthetic valve replacement. Limited observational data suggest no increase in bleeding with use of DOACs early after surgery. The efficacy and safety of DOACs compared with VKAs early after bioprosthetic valve replacement, when thrombotic and bleeding risk are highest, are uncertain.

METHODS AND RESULTS: We systematically searched EMBASE, MEDLINE and CENTRAL until November 16, 2021 for randomized trials comparing DOACs to VKAs in the first three months after bioprosthetic valve replacement. After duplicate title, abstract and full text screening, data abstraction, and risk of bias evaluation, we pooled data using the Mantel-Haenzel method and random effects modelling. We assessed the certainty of evidence using Grading of Recommendations, Assessments, Development and Evaluation (GRADE). We examined thrombotic events (a composite of stroke, systemic embolism, and valve thrombosis), major bleeding, death, and net clinical adverse events (a composite of thrombotic events and major bleeding). We preplanned a sensitivity analysis of studies that initiated treatment prior to hospital discharge. We included 5 trials with 2311 patients and a median follow-up of 12 months (range three to 18 months; Table 1). Two studies included patients with transcatheter valve replacement and three included patients with surgical bioprosthetic valve replacement. Three studies initiated therapy prior to hospital discharge. Risk of bias was low or presented some concerns in all studies. There were fewer thrombotic events and more bleeding events with DOACs compared to VKAs but these results were not statistically significant (Table 1). There was no difference in risk of deaths between groups. Net adverse clinical events were similar between groups. Sensitivity analyses were consistent with overall results; risk of major bleeding was higher with DOACs than VKAs in patients who initiated treatment prior to hospital discharge but the estimate is imprecise.

Conclusion: Moderate certainty evidence suggests that use of DOACs in the first 3 months after bioprosthetic valve replacement may be associated with higher risk of bleeding when initiated prior to hospital discharge. Assessment of net adverse clinical events reveals no significant difference: decreased numbers of thrombotic events with DOACs are counterbalanced by increased numbers of bleeding events. Future studies should focus on bleeding associated with initiation of DOAC therapy prior to hospital discharge. Conclusions about net adverse clinical events may require consideration of additional patient-important factors such as cost, ease of administration, and subclinical valve thrombosis.