Real-world persistence to direct oral anticoagulants in patients with atrial fibrillation: a systematic review and network meta-analysis

Abstract

Objective To conduct a systematic review and network meta-analysis of real-world evidence comparing adherence, persistence, cost, and utilization between oral anticoagulant (OAC) in non-valvular atrial fibrillation (NVAF) patients. Methods A systematic search of MEDLINE and Embase (inception-July 2019) was conducted for published observational cohort studies comparing outcomes between ≥2 OACs. A network meta-analysis was performed to estimate odds ratios for non-persistence using a random-effects model. Results There were 80 studies evaluating the outcomes of interest. However, due to a paucity in adherence studies and heterogeneity in adherence, cost, and utilization definitions, persistence was the focus of this network meta-analysis. There were 36 studies evaluating non-persistence in 395,593 participants, 24 of which used 3 gap definitions (30-, 60-, and 90-days); 18 unique studies evaluating non-persistence at 12 months were included in the network meta-analysis. Using 30- and 90-day gaps, all NOACs, when compared with VKAs, had lower odds of non-persistence (30-day OR (95%CI): apixaban: 0.63 (0.58, 0.69); rivaroxaban: 0.69 (0.62, 0.76); dabigatran: 0.89 (0.82, 0.97); 90-day OR (95%CI): apixaban: 0.33 (0.22, 0.47); rivaroxaban: 0.47 (0.36, 0.61); dabigatran 0.61 (0.44, 0.85)). When using a 60-day gap, dabigatran had higher odds of non-persistence vs VKAs (OR: 1.35; 95%CI: 1.12, 1.61), but there were no significant differences for apixaban and rivaroxaban. Apixaban had the lowest probability of non-persistence across the 3-gap definitions (95.7% with 30-day gap, 76.9% with 60-day gap, 98.4% with 90-day gap). Conclusions The current findings, despite multiple limitations, can raise awareness and understanding of real-world persistence associated with OAC therapy in NVAF patients.