We read with interest the comments from Dr Stefanidis et al., which point out some discrepancies between the recommendations in the recent guidelines on the management of patients with valvular heart disease1,2 and in particular those on the management of patients with atrial fibrillation.3
Guidelines are supposed to represent the best of our current knowledge, thus, they can only be supported by the knowledge existing in the field today. Unfortunately, in valvular disease, the level of evidence is more limited than in other domains and recommendations usually rely on expert consensus.
These limitations apply to the use of antiplatelet drugs on top of anticoagulation in patients with valve prostheses. Available trials show that antiplatelet drugs can be beneficial in patients with vascular disease alone as well as in those with vascular disease and prosthetic valves. The decrease in thrombo-embolic risk observed with the addition of antiplatelet drugs seems to be more related to a decrease in complications of coronary artery disease (myocardial infarction, heart failure, and sudden death) rather than to all embolic events.4 But it has not been demonstrated that this benefit is also present in those without vascular disease. In addition, all trials show consistently that when added to anticoagulation these agents increase the risk of bleeding. Thus, the ESC Task Force on Valvular Heart Disease propose to tailor this combined therapy to specific indications and not to expand to all patients.
We recommend tailoring the degree of anticoagulation to the thrombo-embolic risk of the specific prosthesis and the patient. In a given patient with atrial fibrillation, the degree of anticoagulation is higher than those in sinus rhythm with no other risk factors.
Finally, there is no evidence to support the long-term use of antiplatelet agents in patients with bioprostheses who have no other indications.
However, it should be noted that fortunately the ACC/AHA and the ESC guidelines on valve disease are overall consistent in most domains and convey the same messages, the differences being in some topics where evidence is lacking.
Generally speaking, we agree that the large number of published guidelines, even more so if their recommendations are discordant, may well cause confusion for the reader and therefore limit their use in clinical practice, which is the final goal. The comments made by Dr Stefanidis should support the performance of further trials to provide better evidence and also that of consensus papers between the different Tasks Forces on a given field to underline the main messages and explain the potential discrepancies.