Superficial-vein thrombosis can lead to deep-vein thrombosis and pulmonary embolism. Rivaroxaban, an oral factor Xa inhibitor, might simplify treatment compared with fondaparinux because it does not require daily subcutaneous injection and is cheaper. We compared efficacy outcomes in patients with superficial-vein thrombosis and additional risk factors given either rivaroxaban or fondaparinux to assess whether rivaroxaban is non-inferior to fondaparinux in the prevention of thromboembolic complications.

In this open-label, masked endpoint, randomised, non-inferiority phase 3b trial, we recruited patients aged 18 years or older with symptomatic superficial-vein thrombosis from 27 sites (academic, community hospitals, and specialist practices) in Germany. We randomly assigned patients (1:1) to receive 10 mg oral rivaroxaban or 2·5 mg subcutaneous fondaparinux once a day for 45 days. Patients were eligible if they had symptomatic thrombosis (at least 5 cm in a supragenual superficial-vein segment) and at least one additional risk factor (older than 65 years, male sex, previous venous thromboembolism, cancer, autoimmune disease, thrombosis of non-varicose veins). Main exclusion criteria were: symptoms for longer than 3 weeks, thrombus within 3 cm of the sapheno-femoral junction, indication for full-dose anticoagulation therapy, and substantial hepatic or renal impairment. Randomisation was done with a central block randomisation process. The primary efficacy outcome was a composite of symptomatic deep-vein thrombosis or pulmonary embolism, progression or recurrence of superficial vein-thrombosis, and all-cause mortality at 45 days in the per-protocol population (all randomly assigned patients without major protocol violations). We used a non-inferiority margin of 4·5% (absolute difference between rivaroxaban and fondaparinux). The main safety outcome was major bleeding. This study is registered with ClinicalTrials.gov, number NCT01499953.

Between April 25, 2012, and Feb 18, 2016, 485 patients were enrolled in the study and 472 were randomly assigned to the rivaroxaban group (n=236) or the fondaparinux group (n=236). In the 435 patients included in the per-protocol analysis set, the primary efficacy outcome occurred in seven (3%) of 211 patients (95% CI 1·6-6·7) in the rivaroxaban group and in four (2%) of 224 patients (0·7-4·5) in the fondaparinux group (hazard ratio [HR] 1·9, 95% CI 0·6-6·4; p=0·0025 for non-inferiority) at day 45. There were no major bleeds in either group. There was one death in the rivaroxaban group; this patient died from cardiogenic shock on day 50 after a type A aortic dissection, not related to treatment.

Rivaroxaban was non-inferior to fondaparinux for treatment of superficial-vein thrombosis in terms of symptomatic deep-vein thrombosis or pulmonary embolism, progression or recurrence of superficial vein-thrombosis, and all-cause mortality, and was not associated with more major bleeding. Therefore, rivaroxaban could offer patients with symptomatic superficial-vein thrombosis a less burdensome and less expensive oral treatment option instead of a more expensive subcutaneous injection.

GWT-TUD and Bayer Vital.