Mortality benefit in the COMPASS trial: is it related to superior statistical power or better efficacy and safety?

Until recently, attempts to improve the benefit of aspirin by adding another antithrombotic agent have not resulted in a mortality reduction in patients with chronic symptomatic atherosclerosis. In this population, COMPASS is the only one among six trials to show a significant mortality reduction, thereby providing evidence of a clear net clinical benefit with the combination of low-dose rivaroxaban plus aspirin. In this systematic review, we sought to determine whether the mortality benefit of the combination arm in COMPASS is best explained by greater statistical power or by a more favorable efficacy-safety profile than the other regimens evaluated in patients with chronic symptomatic atherosclerosis.

Plasma Rivaroxaban Level to Identify Patients at Risk of Drug Overexposure: Is a Single Measurement of Drug Level Reliable?

Introduction Dose adjustment based on laboratory monitoring is not routinely recommended for patients treated with rivaroxaban but because an association has been reported between high drug level and bleeding, it would be of interest to know if measuring drug level once could identify patients at risk of bleeding who might benefit from a dose reduction. Objective This study was aimed to investigate the reliability of a single measurement of rivaroxaban level to identify clinic patients with persistently high levels, defined as levels that remained in the upper quintile of drug-level distribution. Methods In this prospective cohort study of 100 patients with atrial fibrillation or venous thromboembolism, peak and trough rivaroxaban levels were measured using the STA-Liquid Anti-Xa assay at baseline and after 2 months. Values of 395.8 and 60.2 ng/mL corresponded to the 80th percentile for peak and trough levels, respectively, and levels above these cut-offs were categorized as high for our analyses. Results Among patients with a peak or trough level in the upper quintile at baseline, only 26.7% (95% confidence interval [CI]: 10.9-52.0%), and 13.3% (95% CI: 2.4-37.9%), respectively, remained above these thresholds. Conclusion Our findings do not support the use of a single rivaroxaban level measurement to identify patients who would benefit from a dose reduction because such an approach is unable to reliably identify patients with high levels.