Adverse events in clinically complex elderly patients with atrial fibrillation according to oral anticoagulation status

Background Few data are available about the impact of oral anticoagulants (OAC) in patients with Atrial Fibrillation (AF) and clinical complexity (CC). Methods We conducted a retrospective study utilising data from the TriNetX network. Based on ICD-10-CM codes entered between 2020 and 2022, AF patients aged ≥75 years on long-term OAC with CC were categorised into two groups based on OAC use in the year before entering the study (maintained vs discontinued). CC was defined as BMI ≤23 kg/m2, and/or history of bleeding, and/or chronic kidney disease. The primary outcomes were the one-year risk of all-cause death, major cardiovascular events (MACE), and major bleeding. Cox regression analyses were used to calculate hazard ratios (HRs) and 95% CIs before and after 1:1 propensity score matching (PSM). Findings We identified 6554 AF CC patients who discontinued OAC (mean age 81.5 ± 6.0 years, 46.7% females) and 23,212 AF patients with CC who maintained OAC (81.3 ± 6.0 years, 49.4% females). Before PSM, AF CC patients who discontinued OAC had a higher prevalence of intracranial, gastrointestinal haemorrhages, and antiplatelet use, with no significant differences after PSM. OAC discontinuation was associated with a higher risk of all-cause death (HR 1.22, 95% CI 1.11–1.35) and MACE (HR 1.38, 95% CI 1.25–1.53). The one-year risk of major bleeding was similar in those who discontinued or maintained OAC (HR 1.05, 95% CI 0.94–1.18), although it was significantly higher during the early follow-up (HR 1.51, 95% CI 1.24–1.83). The risk of primary outcomes decreased over time, with the risk of bleeding becoming not significant. Interpretation AF CC patients who discontinued OAC have a high risk of adverse events. New antithrombotic and integrated care approaches to reduce thrombotic risk without increasing bleeding risk are needed in these patients.