Real-world variability in dabigatran levels in patients with atrial fibrillation

BackgroundIn clinical practice, physicians are given the choice of selecting one of two dabigatran doses based on patient characteristics, with the lower dose typically used in patients at a higher risk of bleeding. ObjectivesThe objectives of the study were to (i) estimate the inter- and intra-patient variability in dabigatran levels with 110mg (DE110) and 150mg (DE150) doses, (ii) examine the effect of physicians' dose selection on levels in DE110 and DE150 subgroups, and (iii) explore whether a single trough measurement identifies patients with extreme levels on subsequent visits. MethodsIn this prospective observational study of 100 patients with atrial fibrillation (AF), peak and trough levels of dabigatran were measured with the Hemoclot((R)) assay at baseline and every 2months thereafter (maximum four visits). ResultsInter-patient variability in dabigatran levels (geometric coefficient of variation [gCV], 51-64%) was greater than intra-patient variability (gCV, 32-40%). Similar medians and distributions of levels were observed in DE110 and DE150 subgroups. Patients receiving DE110 were older, had lower renal function and weighed less than those receiving DE150. Up to 40% of patients whose trough levels were in the upper extremes, and up to 80% of patients whose trough levels were in the lower extremes at baseline, showed subsequent levels that fell in the middle quartiles. ConclusionsOur data support the practice of selecting the dabigatran dose based upon clinical characteristics because it results in similar levels of drug exposure in patients given DE110 or DE150. They do not support the concept that a single Hemoclot((R)) measurement reliably identifies patients with consistently high or low values.