Rivaroxaban is an oral direct factor Xa inhibitor that has been effective in prevention of venous thromboembolism in patients undergoing elective orthopaedic surgery. However, its use after acute coronary syndromes has not been investigated. In this setting, we assessed the safety and efficacy of rivaroxaban and aimed to select the most favourable dose and dosing regimen. In this double-blind, dose-escalation, phase II study, undertaken at 297 sites in 27 countries, 3491 patients stabilised after an acute coronary syndrome were stratified on the basis of investigator decision to use aspirin only (stratum 1, n=761) or aspirin plus a thienopyridine (stratum 2, n=2730). Participants were randomised within each strata and dose tier with a block randomisation method at 1:1:1 to receive either placebo or rivaroxaban (at doses 5-20 mg) given once daily or the same total daily dose given twice daily. The primary safety endpoint was clinically significant bleeding (TIMI major, TIMI minor, or requiring medical attention); the primary efficacy endpoint was death, myocardial infarction, stroke, or severe recurrent ischaemia requiring revascularisation during 6 months. Safety analyses included all participants who received at least one dose of study drug; efficacy analyses were by intention to treat. This study is registered with ClinicalTrials.gov, number NCT00402597. Three patients in stratum 1 and 26 in stratum 2 never received the study drug. The risk of clinically significant bleeding with rivaroxaban versus placebo increased in a dose-dependent manner (hazard ratios [HRs] 2.21 [95% CI 1.25-3.91] for 5 mg, 3.35 [2.31-4.87] for 10 mg, 3.60 [2.32-5.58] for 15 mg, and 5.06 [3.45-7.42] for 20 mg doses; p<0.0001). Rates of the primary efficacy endpoint were 5.6% (126/2331) for rivaroxaban versus 7.0% (79/1160) for placebo (HR 0.79 [0.60-1.05], p=0.10). Rivaroxaban reduced the main secondary efficacy endpoint of death, myocardial infarction, or stroke compared with placebo (87/2331 [3.9%] vs 62/1160 [5.5%]; HR 0.69, [95% CI 0.50-0.96], p=0.0270). The most common adverse event in both groups was chest pain (248/2309 [10.7%] vs 118/1153 [10.2%]). The use of an oral factor Xa inhibitor in patients stabilised after an acute coronary syndrome increases bleeding in a dose-dependent manner and might reduce major ischaemic outcomes. On the basis of these observations, a phase III study of low-dose rivaroxaban as adjunctive therapy in these patients is underway. Johnson & Johnson Pharmaceutical Research & Development and Bayer Healthcare AG.