Abstract
Background In patients with nonvalvular atrial fibrillation (AF), apixaban is given in doses of 5 or 2.5 mg twice daily, according to clinical characteristics. The usual on-treatment range of apixaban drug levels, as determined by apixaban-calibrated anti-factor Xa (anti-Xa) activity, has previously been measured in small cohorts; however, the association between anti-Xa activity and clinical outcomes and the predictors of variability in anti-Xa activity have not been well studied in the AF population. Methods and Results Anti-Xa activity was measured before taking the morning dose, 3 months after enrollment in the AVERROES study using a calibrated anti-Xa assay (Rotachrom). Patients with two of the following criteria-age >80; weight <60 kg; or creatinine >133 μg/L-received 2.5 mg twice daily ( n = 145), while all others received 5 mg twice daily ( n = 2,247). A total of 2,392 patients were included, with median follow-up of 1.1 years. Median apixaban anti-Xa activity was 122 ng/mL (interquartile range [IQR]: 63-198 ng/mL) for the entire group; 99 ng/mL (IQR: 60-146 ng/mL) for the 2.5-mg group; and 125 ng/mL (IQR: 64-202 ng/mL) for the 5-mg group ( p = 0.003). A relationship was evident between bleeding and anti-Xa activity ( p = 0.01), which was driven by minor bleeding. No relationship was evident between major bleeding or stroke/systemic embolism and anti-Xa activity. In those receiving the 5-mg dose, estimated glomerular filtration rate, sex, and age had the strongest association with anti-Xa activity. Conclusion There is considerable variability in anti-Xa activity among AF patients receiving apixaban. Rates of major bleeding and stroke/systemic embolism were low irrespective of anti-Xa activity. Clinical Trial Registration ClinicalTrials.gov NCT00496769; https://clinicaltrials.gov/ct2/show/NCT00496769 .