Appropriateness of antithrombotics in geriatric inpatients with atrial fibrillation: a retrospective, cross-sectional study

Background Atrial fibrillation occurs in nearly half of geriatric inpatients and is a major cause of morbidity and mortality. Suboptimal anticoagulation use is an important concern in this population. This study aimed to evaluate the appropriateness of antithrombotic therapies in this patient cohort. Methods A retrospective analysis was conducted on the geriatric wards of a teaching hospital in Belgium, on a background of clinical pharmacy services. The first 90 atrial fibrillation patients from 2020 to 2022 were included if they received an oral anticoagulant. We assessed utilisation and appropriateness of antithrombotics at discharge, examined reasons for guideline deviations, and explored factors associated with underdosing. Temporal associations for appropriateness and type of anticoagulant (vitamin K antagonist (VKA) vs direct oral anticoagulant (DOAC)) were assessed. Results The mean age of patients was 86.5 (±5.3) years and the median CHA2DS2-VASc score was 5 (interquartile range (IQR) 4–6). At discharge, 256 (94.8%) patients used a DOAC; nine (3.3%) used a VKA; one (0.4%) a DOAC-antiplatelet combination, and in four patients (1.5%) all antithrombotics were discontinued. The majority (64.4%) of patients received reduced DOAC doses with apixaban prescribed in 40.7%. In 39 (14.4%) patients, antithrombotic use was considered inappropriate, mostly without a rationale (23/39). Year 2022 (odds ratio (OR) 0.104; 95% confidence interval (CI), 0.012–0.878) was the sole determinant for underdosing. No significant differences were found with respect to appropriateness (p=0.533) or anticoagulant class (p=0.479) over time. Conclusion Most geriatric inpatients received a justified reduced DOAC dose. A significant proportion was managed inappropriately with underdosing (= unjustified reduced dose) being most common. Frequently no rationale was provided for deviating from trial-tested doses. The datasets generated and analysed during the study are available from the corresponding author on reasonable request.