Direct Oral Anticoagulants for Stroke Prevention in Patients with Atrial Fibrillation and Bioprosthetic Heart Valves

Abstract

There are limited data to guide the most effective anticoagulation strategy in patients with a bioprosthetic heart valve and concomitant atrial fibrillation (AF). The use of vitamin-K antagonists (VKA) has been supported by limited data from randomized trials.1Guimarães HP Lopes RD De Barros EGM et al.Rivaroxaban in patients with atrial fibrillation and a bioprosthetic mitral valve.N Engl J Med. 2020; 383 (doi:10.1056/NEJMoa2029603.): 2117-2126Google Scholar Direct oral anticoagulants (DOACs) have emerged as an effective and safe alternative to VKA and are now considered the preferred agents for stroke prevention in most patients with non-valvular AF. However, the use of DOACs for the prevention of stroke or systemic embolism in patients with bioprosthetic valve and AF remains uncertain. As a result, many physicians continue to prescribe VKA in this setting. Thus, we conducted a meta-analysis leveraging clinical data from prior and recently completed randomized controlled trials (RCTs) to evaluate the net clinical benefit (NCB) of DOACS as compared with VKA for prevention of stroke or systemic embolism in patients with AF and bioprosthetic heart valve. We queried PubMed and Embase from their inception through February 10, 2021 to identify randomized trials that reported on the comparison of DOACs versus VKA in patients with bioprosthetic valves and AF. Data were extracted regarding stroke or systemic embolism events, bleeding events, and mortality. The 2 groups (DOACs and VKA) were compared for the following outcomes: major or life-threatening bleeding, stroke or systemic embolism, and all-cause mortality at the longest available follow-up. We used Mantel-Haenszel method with DerSimonian-Laird estimator for tau2 for random effects model and Q-profile method for confidence interval of tau2 and tau to calculate risk ratio (RR) with 95% confidence interval (CI). Heterogeneity was assessed using Higgins I2 statistics. NCB for DOACs was defined as stroke or systemic embolism events prevented by DOACs minus major or life-threatening bleeding caused by warfarin multiplied by a weighting factor (Net Clinical Benefit DOACs = (Stroke rate VKA – Stroke rate DOACs) – Weight factor (Major or life-threatening bleeding DOACs – Major or life-threatening bleeding VKA). This approach has previously been validated.2Singer DE Chang Y Fang MC et al.The net clinical benefit of warfarin anticoagulation in atrial fibrillation.Ann Intern Med. 2009; 151 (doi:10.7326/0003-4819-151-5-200909010-00003.): 297-305Google Scholar The weighting factor reflects the relative impact, with regard to death and disability, of major or life-threatening bleeding compared with stroke. In addition to a weighting factor of 1.5, sensitivity analyses by using weighting factors 1.0 and 2 were also performed. Four RCTs1Guimarães HP Lopes RD De Barros EGM et al.Rivaroxaban in patients with atrial fibrillation and a bioprosthetic mitral valve.N Engl J Med. 2020; 383 (doi:10.1056/NEJMoa2029603.): 2117-2126Google Scholar,3Carnicelli Anthony P Raffaele DC Halperin Jonathan L et al.Edoxaban for the prevention of thromboembolism in patients with atrial fibrillation and bioprosthetic valves.Circulation. 2017; 135 (doi:10.1161/CIRCULATIONAHA.116.026714.): 1273-1275Google Scholar, 4Durães AR de Souza Roriz P de Almeida Nunes B et al.Dabigatran versus warfarin after bioprosthesis valve replacement for the management of atrial fibrillation postoperatively: DAWA pilot study.Drugs R D. 2016; 16 (doi:10.1007/s40268-016-0124-1.): 149-154Google Scholar, 5Guimarães PO Pokorney SD Lopes RD et al.Efficacy and safety of apixaban vs warfarin in patients with atrial fibrillation and prior bioprosthetic valve replacement or valve repair: insights from the ARISTOTLE trial.Clin Cardiol. 2019; 42 (doi:10.1002/clc.23178.): 568-571Google Scholar including 1,381 patients (723 in DOAC group and 656 in VKA group) were included in the final analysis. The mean follow-up duration was 18 months (range 3 to 34 months). DOACs were associated with a lower risk of major bleeding (RR: 0.53; CI: 0.30–0.93; p < 0.05; I2 = 0%) compared with VKA (Figure 1a). There were no statistically significant differences in the risks of stroke or systemic embolism (RR: 0.48; CI: 0.22–1.01; p > 0.05; I2 = 12%) or all-cause mortality (RR: 0.95; CI: 0.57–1.60; p > 0.05; I2 = 0%) with DOACs relative to VKA (Figure 1b-c). With a weight factor of 1.5, pooled effect of RCTs showed a NCB of DOACs over VKA of 3.5% (95% CI: −6.08% to 13.05%). The NCB observed in individual trials were as follows: ENGAGE AF-TIMI 48: 18.7% (95% CI: 2.25% to 26.47%), ARISTOTLE: 1.4% (95% CI: −12.14% to 9.42%), and RIVER: 3.7% (95% CI: 1.42% to 6.03%). The patterns of results (pooled NCB) were preserved using a weight factor of 1 [NCB: 2.9% (95% CI:-6.66% to 12.46%)] and 2 [NCB: 4.07% (95% CI: −5.49% to 13.63%)]. In this study-level meta-analysis of data from four RCTs, including nearly 1,400 patients, we found that DOACs were non-inferior to VKA in preventing stroke or systemic embolism and carried a net clinical benefit that was driven by a lower risk of major bleeding in patients with bioprosthetic mitral or aortic valves with concomitant AF. Despite its proven efficacy, VKA use is limited by its narrow therapeutic index, multiple food and drug interactions and need for frequent monitoring. DOACs offer a better pharmacological profile and are non-inferior or even superior to VKA in terms of preventing strokes in patients with AF. Furthermore, most DOACs have proven to have a lower risk of major bleeding than VKA. In the RE-ALIGN trial, dabigatran as compared with VKA resulted in an increased risk of bleeding and thromboembolic events in patients with mechanical prosthetic valves. This study raised the concern that perhaps DOACs should not be used in either forms of prosthetic valves (mechanical or bioprosthetic). The superiority of VKA over dabigatran in this trial was attributed to VKA’s ability to inhibit activation of both tissue factor and contact pathway induced coagulation in addition to factor X and thrombin. Both tissue factor and contact pathway play a significant role in mechanical prosthesis induced thrombosis. In contrast to VKA, dabigatran only inhibits thrombin. Our analysis has certain limitations: the participants in the included studies had both aortic and mitral bioprostheses, which carry different risk of thromboembolism, target INR levels for anticoagulation, and consequently different bleeding predispositions. However, the present study sheds new light on a question often faced by clinicians. It supports a risk assessment strategy that incorporates both risk for thromboembolism and the risk for major bleeding provides a practical basis for the decision on antithrombotic therapy in patients with atrial fibrillation and bioprosthetic valves. These data are intended to be a springboard for future studies and for physician-patient conversations pertaining to anticoagulation post-procedurally in a shared decision-making process.