RESUMO
BACKGROUND: Direct oral anticoagulants (DOACs) have a favourable risk-benefit profile compared to vitamin K-antagonists (VKAs) in atrial fibrillation (AF). Dosing is based on age, weight and renal function, without need of routine monitoring. METHODS AND RESULTS: In two prospective, multicentre AF cohorts (Swiss-AF, BEAT-AF) patients were stratified as receiving VKAs or adequately-, under- or overdosed DOACs, according to label. Primary outcome was a composite of major adverse clinical events (MACE), defined as cardiovascular death, myocardial infarction (MI), ischaemic stroke and systemic embolism. Secondary outcomes included major bleeding. Adjustment for confounding was performed. Median follow-up was 4 years. Of 3236 patients, 1875 (58%) were on VKAs and 1361 (42%) were on DOACs, of which 1137 (83%) were adequately-, 134 (10%) over- and 90 (7%) under-dosed. Compared to adequately dosed individuals, overdosed patients were more likely to be older and female. Underdosing correlated with concomitant aspirin therapy and coronary artery disease. Both groups had higher CHA2DS2-VASc scores. Patients on overdosed DOACs had higher incidence of MACE (HR 1.75; CI 1.10-2.79; adjusted-HR: 1.22) and major bleeding (HR 1.99; CI 1.14-3.48; adjusted-HR: 1.51). Underdosing was not associated with a higher incidence of MACE (HR 0.94; CI 0.46-1.92; adjusted-HR 0.61) or major bleeding (HR 1.07; CI 0.46-2.46; adjusted-HR 0.82). After adjustment, all CIs crossed 1.0. CONCLUSION: Inappropriate DOAC-dosing was more prevalent in multimorbid patients, but did not correlate with higher risks of adverse events after adjusting for confounders. DOAC prescription should follow label.