No benefit of thromboprophylaxis in medically ill patients MARINER trial

Rivaroxaban for thromboprophylaxis after hospitalization for medical illness

Spyropoulos AC et al.

The results of the multicenter randomized controlled MARINER trial (671 participating centers) were discussed at the ESC congress 2018 and published in the New England Journal of Medicine on 20^th September 2018. The MARINER trial assessed the clinical benefit of a prolonged prophylactic anticoagulation with rivaroxaban versus placebo with respect to the prophylaxis of venous thromboembolism (VTE) in acutely medical ill patients.

The background of this study refers to the controversy on a prolonged thromboprophylaxis due to an ongoing risk of venous thromboembolism after hospital discharge. The inclusion criteria of MARINER were a hospitalization for at least 3 days (no longer than 10 days) for heart failure, respiratory insufficiency, stroke, an infectious or an inflammatory disease and an additional risk factor for VTE (defined as total modified International Medical Prevention Registry on Venous Thromboembolism risk score of ≥ 4, or a risk score of 2 or 3 plus a D-dimer > twice the upper limit). Included patients were randomized to receive either once-daily 10 mg rivaroxaban (or 7.5 mg in renal insufficiency) or placebo for 45 days, starting on the day of discharge (or the next day).

As main result, there was no benefit of prolonged prophylaxis with once-daily 10 mg rivaroxaban regarding a reduction of the primary efficacy endpoint, which was

the composite of symptomatic VTE or VTE-related death. Among secondary endpoints, the risk of the occurrence of symptomatic VTE (analyzed separately) and the composite of symptomatic VTE and death from any cause were lowered by the prophylactic dose of rivaroxaban. However, regarding safety, nonmajor clinically relevant bleedings, other bleedings as well as transfusions of ≥ 2 units of packed red blood cells occurred more frequent in the rivaroxaban group. Referring to these observations, the number needed to treat to prevent one symptomatic VTE was 430 and the number needed to treat to cause one bleeding was 856.

In conclusion, the MARINER trial showed no benefit of a prolonged rivaroxaban administration in medically ill patients as VTE prophylaxis. Despite this negative trial result, MARINER provides clearer insights in the discussion on prolonged VTE prophylaxis with non-vitamin K antagonist oral anticoagulants (NOAC) in medically ill patients. This is an important contribution, since previous trials, such as MAGELLAN and ADOPT, primarily compared prophylactic NOAC regimens with shorter durations of prophylactic low-molecular weight heparins instead of comparing with placebo.