The impact of a direct oral anticoagulant in cardiovascular disease

Abstract

Novel oral anticoagulants (NOACs) have uncomplicated dosing with no need for INR monitoring and fewer possible drug and food interactions compared to warfarin, which may result in improved adherence to treatment and may offset the high retail cost of NOACs. Rivaroxaban 10mg is the only NOAC available on the Maltese National Health Service (NHS) formulary and is not available for cardiology patients. The aim was to compare the NOAC rivaroxaban to warfarin with respect to incidence and severity of bleeding, treatment adherence, drug-drug interactions (DDIs), alcohol consmption and costs. Following ethics approval, 100 patients (50 warfarin, 50 rivaroxaban) were recruited by convenience sampling from Mater Dei Hospital (MDH) outpatients and from community pharmacies. Bleeding complications were classified according to the Bleeding Academic Research Consortium (BARC) criteria. Medication adherence was assessed using a validated adherence questionnaire. Time in therapeutic range (TTR) for warfarin was calculated using the Rosendaal Linear Interpolation Method. Micromedex® and Medscape® drug interaction checker tools were used to identify potential DDIs. Cost analysis incorporated drug, INR monitoring and physician costs. For the total population of 100 patients, mean age was 65 ±12.91 (range 27-85) years, 53 were female and 47 were male, the main indications for anticoagulation were atrial fibrillation (59 patients) followed by deep vein thrombosis (30 patients) and mean duration of anticoagulant therapy was 10 ±5.97 (range 1-33) months. Twenty-four patients reported BARC Type 1 bleeding (18 warfarin and 6 rivaroxaban) and 10 patients reported Type 2 bleeding (6 warfarin and 4 rivaroxaban) (p<0.001). Mean adherence score was 41 ±3.92 (range 30-45) for warfarin and 44 ±1.41 (range 39-45) out of 45 for rivaroxaban (p<0.001). A total of 768 INR tests were processed in a six-month period, with a mean of 2.56 ±1.58 (range 1-7) INR tests per patient per month. Thirty-seven percent of these INR tests were not in therapeutic range. A total of 91 (mean 1.8 ±1.03, range 0-4/patient) and 19 (mean 0.4 ±0.52, range 0-2/patient) potential DDIs were identified in patients on warfarin and rivaroxaban respectively (p<0.001). The mean retail cost/patient/month was €4.20 ±1.63 (range €1.82-€8.46) for patients on warfarin and €97.50 ±0.00 (range €97.50-€97.50) for patients on rivaroxaban Patients on warfarin 1) had an increased incidence of Type 1 and Type 2 bleeding, 2) were less adherent to treatment, 3) had a lower TTR and 4) had a higher risk of potential DDIs. These factors should be incorporated into a pharmacoeconomic model when justifying the inclusion of rivaroxaban (NOACs) in the NHS formulary for cardiology patients.