In some clinical situations, measurements of anticoagulant effect of apixaban may be needed. We investigated the inter- and intra-individual apixaban variability in patients with atrial fibrillation and correlated these results with clinical outcome. We included 62 patients receiving either 5 mg (A5, n = 32) or 2.5 mg (A2.5, n = 30) apixaban twice-daily. We collected three trough and three peak blood samples 6-8 weeks apart. Apixaban concentration was measured by liquid chromatography-tandem mass-spectrometry (LC-MS/MS) and by anti-Xa. Patients on A2.5 were older, had lower creatinine clearance, higher CHA2DS2VASc (4.7 ± 1.0 vs. 3.4 ± 1.7) and lower trough (85 ± 39 vs. 117 ± 53 ng/mL) and peak (170 ± 56 vs. 256 ± 91 ng/mL) apixaban concentrations than patients on A5 (all p < 0.01). In patients on A5, LC-MS/MS showed a significant difference between through levels and between peak levels (p < 0.01). During apixaban treatment, 21 patients suffered bleeding (2 major). There was no association between bleeding and apixaban concentrations or variability. Four patients who suffered thromboembolic event had lower peak apixaban concentrations than patients without it (159 ± 13 vs. 238 ± 88 ng/mL, p = 0.05). We concluded, that there was a significant intra- and inter-individual variability in apixaban trough and peak concentrations. Neither variability nor apixaban concentrations were associated with clinical outcomes.