Is it the time for direct oral anticoagulants in bioprosthetic heart valves? Free

     The quality of life is more important than life itself

                 Alexis Carrel

The issue of bioprosthetic valve thrombosis and thromboembolism is becoming progressively relevant in the current cardiovascular practice. Of course, prophylaxis of valve-related thrombosis and subsequent thromboembolic events after biologic valve replacement is recommended [1–3], especially shortly after the intervention. Indeed, thromboembolic complications occur with higher incidence in the first 3 months after biologic valve replacement due to lack of or incomplete endothelialization [4]. This risk decreases significantly already 90 days after surgery [1], but in many patients with biological prostheses and increased risk of thromboembolic complications [e.g. atrial fibrillation (AF)], oral anticoagulation therapy is often continued after this landmark.

Currently, vitamin K antagonists (VKAs) are the only oral anticoagulants (OACs) with an established effectiveness profile and thus indication for prosthesis-related thromboembolism prophylaxis [2, 3]. However, they have many limitations, including challenging titration. Accordingly, the interest for direct oral anticoagulants (DOACs) in this surgical setting has increased over the past years, given also their ever expanding role in other thrombotic conditions, such as AF [5, 6]. Notrably, DOACs are also used in patients with AF and valvular heart disease, with the exception of significant mitral stenosis and mechanical heart valves [2, 5, 6].

The double-edged sword of OAC is clearly pivoting on the competing (but occasionally synergistic) risks of bleeding and thromboembolism. Indeed, thrombosis has recently become a challenging issue in prosthetic heart valves [7], as clinically significant valve thrombosis, despite being rare, can lead to major adverse events, ranging from heart failure and systemic thromboembolism to death. Furthermore, it can adversely affect valve durability [8]. According to the recent scholarly literature, capitalizing on refined non-invasive imaging techniques, it is apparent that asymptomatic or pauci-symptomatic prosthetic valve thrombosis is not at all rare. Hypo-attenuated leaflet thickening (HALT) can be identified by evidence of increased thickness of the bioprosthetic leaflet on cardiac computed tomography, and the restriction caused by HALT can be described as reduced leaflet motion (RLM). Even HALT and RLM could be the result of phenomena such as thrombosis, endocarditis, leaflet deterioration, and, according to the recent scientific literature, they have been used mostly as a synonym of subclinical leaflet thrombosis. The prevalence of HALT and RLM has been described in 5–10% (but up to 40%) of patients undergoing computed tomography assessment post-aortic valve replacement [9, 10].

This issue of the European Journal of Cardio-Thoracic Surgery features a landmark systematic review and meta-analysis by Eikelboom et al. [11], including 4 randomized controlled trials comparing DOACs versus VKAs in patients with recent bioprosthetic valve replacement and clinical indication for OAC. In this study-level pooled analysis including 2284 patients, the authors have focused on thrombotic events (cerebral, systemic or valve-related), major bleeding, death, aortic valve reintervention and subclinical valve thrombosis. Their results showed no difference in outcomes, among the 2 populations examined, regarding thrombosis, bleeding or death in the first 90 days after bioprosthetic valve implantation. Indeed, this work has several strengths, including the comprehensive search strategy and detailed selection criteria, as well as the rigorous statistical analysis and the reasonably large sample size (2284 patients), and thus can be relied upon to validly inform decision-making.

Some key drawbacks should be taken into account as well, though. First, as highlighted by the authors, the prevalence of transcatheter aortic valve replacement patients (83%) versus surgical valve replacement (SAVR) of the mitral valve (8%) and SAVR of aortic or mitral valves (10%) and, second, the relatively short follow-up (<18 months). Albeit with the objective of comparing the different outcomes the first 90 days, a long-term follow-up would of a paramount importance in order to define the clinical impact of 2 primary outcomes: subclinical thrombosis and valve durability.

Indeed, the authors did not aim to disentangle the conundrum of the limits and possible serious adverse effects that anticoagulant therapy involves as well as the resulting, severe limitations of the lifestyle of patients with heart valve prostheses. Nevertheless, they provided further evidence that DOACs do not adversely impact on the outcomes in patients with heart valve prosthesis in terms of thrombosis, bleeding or death compared to VKAs, at least in the first 3 months after valve replacement procedure. Therefore, such patients could benefit from this remarkably user-friendly treatment.

In conclusion, early to early-to-mid follow-up results support the role of DOACs as an alternative to VKAs in patients undergoing bioprosthetic valve implantation. Yet, long-term follow-up of available trials and future randomized studies are crucial to precisely gauge the comparative risk–benefit profile of DOACs in this challenging clinical setting, especially in case of SAVR.

Conflict of interest: Giuseppe Biondi-Zoccai has consulted for Amarin, Balmed, Cardionovum, Crannmedical, Endocore Lab, Eukon, Innovheart, Guidotti, Meditrial, Microport, Opsens Medical, Terumo and Translumina. All other authors report no conflict of interest.

REFERENCES