Impact of anticoagulation in patients with dementia and atrial fibrillation. Results of the CardioCHUVI-FA registry

Abstract Introduction and objectives Population aging is associated with an increased prevalence of atrial fibrillation (AF) and dementia. This study aimed to analyze the impact of oral anticoagulation in elderly patients with AF and moderate-severe dementia. Methods We conducted a single-center retrospective study analyzing patients aged ≥ 85 years with a diagnosis of AF between 2013 and 2018. The impact of anticoagulation on mortality, embolisms, and bleeding events was assessed by multivariate Cox analysis. In patients with dementia, this analysis was complemented by propensity score matching, depending on whether the patients were prescribed anticoagulant treatment or not. Results Of the 3549 patients aged ≥ 85 years with AF, 221 had moderate-severe dementia (6.1%), of whom 88 (60.2%) were anticoagulated. During a follow-up of 2.8 ± 1.7 years, anticoagulation was associated with lower embolic risk and higher bleeding risk both in patients with dementia (hazard ratio [HR]embolisms, 0.36; 95%CI, 0.15-0.84; HRbleeding, 2.44; 95%CI, 1.04-5.71) and in those without dementia (HRembolisms, 0.58; 95%CI, 0.45-0.74; HRbleeding, 1.55, 95%CI, 1.21-1.98). However, anticoagulation was associated with lower mortality only in patients without dementia (HR, 0.63; 95%CI, 0.53-0.75) and not in those with dementia (adjusted HR, 1.04; 95%CI, 0.63-1.72; P = .541; HR after propensity score matching 0.91, 95%CI, 0.45-1.83; P = .785). Conclusions In patients aged ≥ 85 years with moderate-severe dementia and AF, oral anticoagulation was significantly associated with a lower embolic risk and a higher bleeding risk, with no differences in total mortality.