The VICTORIA Study (Vascular CalcIfiCation and sTiffness induced by ORal antIcoAgulation) is a comparative, parallel, prospective, controlled and randomized study of the structural and functional impact of rivaroxaban versus anti-vitamin K drugs on the arterial vasculature.

Long term oral anticoagulant treatment (> 12 month) is mainly indicated for atrial fibrillation, prosthetic valves and conditions with high risk for recurrent or deep venous thrombosis. For more than 60 years, vitamin K antagonists have been the only oral anticoagulant drugs available to prevent thrombus formation. The use of vitamin K antagonists is associated to the major constraint of a well-adjusted anticoagulation leading to minor/major risk of life threatening bleeds. They also exhibit other rare side-effects including skin eruption and necrosis, hepatic disorders, alopecia. A less known side effect is an increase in soft tissue calcification, including the cardiac valves and the peripheral arterial system. This side effect is explained by the inhibitory effect of vitamin K antagonists on the central (liver) and peripheral (e.g. vascular) carboxylation cycle synthesis of several vitamin K-dependant calcification inhibiting factors, such as the matrix gamma-carboxyglutamate protein, osteocalcin or Gas6 (1). The active form of gamma-carboxyglutamate protein is now identified as a potent local tissue inhibitor of vascular calcification. The calcifying effect of a decrease in gamma-carboxyglutamate protein or the ratio of carboxylated (i.e. active) /uncarboxylated (i.e. inactive) forms of gamma-carboxyglutamate protein have been reported in various acquired metabolic diseases such as chronic renal insufficiency, aging and of genetic origin (e.g. Cutis Laxa, Keutel syndrome,…) (2, 3) as well as in mouse gamma-carboxyglutamate protein -/- models (4). Furthermore, administration of warfarin in rats is a well-known pharmacological model to induce a vascular calcification within 2-4 weeks with an increase in systolic and pulse arterial blood pressure (5).

Vascular calcification is an independent risk factor for cardiovascular morbi-mortality and it is well-demonstrated that an increase in coronary calcium, as measured by the scan Agatston score, is independently linked to a higher risk for events (6, 7). The lower limb mediacalcosis (i.e. Monckeberg disease) is also a risk factor for limb amputation and calcification (8) of the atheromatous plaque represents a risk factor for plaque instability and rupture (9). The pathophysiological mechanisms linking the dystrophic calcification process to morbi-mortality are still unclear. Calcium deposit within the arterial intimal layer is generally associated to atherosclerosis with an increased risk for plaque rupture whereas deposit of calcium within the medial layer of the peripheral arteries (i.e. mediacalcosis) is rather responsible for an increased arterial stiffness and the development of arterial hypertension (10). Recent data from the investigators laboratory have showed site heterogeneity assessed by scan scoring in the calcifying process in the general population and also in a genetically-determined calcifying disease (i.e the pseudoxantoma elasticum).

Two recently published studies have pointed out a link between the use of vitamin K antagonists and an enhanced coronary (11) and extra-coronary (6) calcifications. Although the conclusions of these studies remains limited by a cross-sectional and retrospective design, a small number of patients and a large range of exposure to vitamin K antagonists (from 6 to 143 months - mean 46) they questioned a potential deleterious effect on the peripheral vasculature mainly for the long term use of non-vitamin K antagonists anticoagulants. One prospective controlled study in post-menopaused woman has demonstrated a long-term beneficial effect only of a supplement containing vitamins K1 and D on the elastic properties of the carotid artery (12). Therefore, in the present study, the investigators propose to determine the structural (i.e. calcification) and functional (i.e. stiffness) impact of the anti-Xa inhibitor rivaroxaban compared to vitamin K antagonists on the arterial structure in a longitudinal, prospective comparative study.

Inclusion Criteria:

• Male or female patient aged > 18 years
• Female patient capable of bearing children with highly effective methods of birth control
• Creatinine clearance > 30 ml/min
• Normal hepatic function based on hepatic enzymes
• Treated for atrial fibrillation according a score superior at 1
• Treatment duration 12 months according to the actual recommendations
• Treated by vitamine K antagonist less than 2 months before entering the study
• Patient willing to participate with a signed informed consent
• Patient covered by a healthcare insurance

Exclusion Criteria:

• Patient has any clinical condition which does not allow initiation of long-term including all contraindications such as hypersensitivity to active ingredient or other excipients, clinically relevant acute bleedings and all other risk circumstances according to Summary of Medicinal Product in which all warnings and preventive measures and precautions are described and have to be kept.
• Patients had a previous coronary stent implantation
• Creatinine clairance <30 mL)
• Liver disease with coagulopathy or other bleeding disorders including cirrhotic patients with Child Pugh
• Hyperthryroidism
• Hypercalcemia
• Hyperphosphatemia
• Acute gastrointestinal diseases
• Mental condition rendering the patient unable to understand the nature, scope and possible consequences of the study
• Patient is unwilling or unable to give informed consent
• Patient is unlikely to comply with protocol, e.g. uncooperative attitude, inability to return for follow-up visits, and unlikelihood of completing the study
• Participation in a parallel interventional clinical trial
• Patient has been committed to an institution by legal or regulatory order
• Pregnant or lactating women
• Female patient capable of bearing children without highly effective methods of birth control
• Patient with history of myocardial infarction and/or coronary disease.

Data and biological samples could be shared upon acceptance between both organizations