Outcomes of long-term anticoagulation in frail elderly patients with atrial fibrillation

Objective: To assess the outcomes of long-term anticoagulation in elderly nursing home patients with atrial fibrillation (AF) and investigate the influence of age, gender and co-morbid conditions. Design: A retrospective chart review was performed in five randomly selected nursing facilities. Patients: Thirteen males and 74 females with nonrheumatic AF, and a mean age of 82.4 years (SD 7.7 years) were included; 74 patients (85%) were older than 75 years. 92% were Caucasian, 74% had a prior cerebral event [61 disabling strokes and three transient ischaemic attacks (TIA)]. All had one to five co-morbid conditions known to be stroke risk factors (mean 2.7, SD 0.9). Two and more risk factors were present in 91% of the patients, and three and more risk factors were present in 63% of the patients. Interventions: Treatment with warfarin was given for 12 to 72 months, (mean 25.8 months, SD 4.3 months). The mean dose of warfarin was 2.8mg (SD 1.1mg). Twelve monthly international normalised ratio (INR) values were extracted for each patient. Outcome Measures: A stroke/TIA or a major bleed was considered an adverse outcome. A multivariate logistic regression model was used to analyse the data. Results: Sixteen patients (18.3%) had adverse outcomes: four strokes, one TIA and 11 bleeds. Half of the events, three strokes and five bleeds, were fatal. Gastro-intestinal bleeds had a 56% mortality rate. Patients with and without adverse outcomes were similar in terms of age and intensity of anticoagulation (mean INR 2.1, SD 0.5 and 0.4). Adverse events were seen in 12% of females and in 54% of males. The influence of gender was independent of age, number of co-morbid conditions and intensity of anticoagulation entered in the multivariate model. Conclusions: The patients in this study had a rate of bleeds higher than in reported clinical trials. Their older age (85% older than 75 years) and greater co-morbidity (91% had more than two stroke risk factors) were the likely causes of this difference. The outcomes of anticoagulation were better in females than in males. Further studies are required on outcomes of anticoagulation in elderly patients with AF.