Right drug, wrong dosage: insights from the PAVE-AF antithrombotic study in older patients with atrial fibrillation

Optimal antithrombotic treatment of older patients is usually impeded by several prevailing misconceptions. The aim of our study was to assess the type, dosage and predictors of antithrombotic therapy in older patients with non-valvular atrial fibrillation (NVAF). PAVE-AF was a prospective, cross-sectional study, including NVAF patients ≥ 80 years from 30 participating centers. Demographic data, comorbidities and treatment patterns were documented in a single visit. Patients treated with non-vitamin K oral anticoagulants (NOACs) were further classified into three dosing categories (recommended, underdosing and overdosing). Among 1018 patients (85.4±4.0 years), 88.4% received anticoagulants (AC), 8% antiplatelets (AP) and 3.6% no treatment. The primary reason for AP administration was physician concern of bleeding followed by patient denial. Patients ≥90 years had two times greater probability to receive AP therapy compared to patients < 90 years. Among patients treated with AC, one third received vitamin K antagonists, while two thirds received NOACs [34.6% apixaban, 9.5% dabigatran and 22.6% rivaroxaban]. Independent predictors of AC prescription over AP or no treatment were low HAS-BLED score, hypertension, labile INR, permanent AF, absence of uncontrolled hypertension, prior stroke/systemic embolism, age and male gender. In total, 37% of NOAC recipients received inappropriate dosage, while the number of patients receiving recommended dosing differed significantly among NOAC subgroups (p < 0.001). In our study, a minority of older NVAF patients received AP or no therapy for stroke prevention. Among patients treated with anticoagulants, two thirds were on NOAC treatment, though with a considerable proportion of inappropriate dosing. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1007/s11239-020-02167-8) contains supplementary material, which is available to authorized users.