Factors affecting adherence to guidelines for antithrombotic therapy in elderly patients with atrial fibrillation admitted to internal medicine wards

Current guidelines for ischemic stroke prevention in atrial fibrillation or flutter (AFF) recommend Vitamin K antagonists (VKAs) for patients at high-intermediate risk and aspirin for those at intermediate-low risk. The cost-effectiveness of these treatments was demonstrated also in elderly patients. However, there are several reports that emphasize the underuse of pharmacological prophylaxis of cardio-embolism in patients with AFF in different health care settings. AIMS: To evaluate the adherence to current guidelines on cardio-embolic prophylaxis in elderly (> 65 years old) patients admitted with an established diagnosis of AFF to the Italian internal medicine wards participating in REPOSI registry, a project on polypathologies/polytherapies stemming from the collaboration between the Italian Society of Internal Medicine and the Mario Negri Institute of Pharmacological Research; to investigate whether or not hospitalization had an impact on guidelines adherence; to test the role of possible modifiers of VKAs prescription. METHODS: We retrospectively analyzed registry data collected from January to December 2008 and assessed the prevalence of patients with AFF at admission and the prevalence of risk factors for cardio-embolism. After stratifying the patients according to their CHADS(2) score the percentage of appropriateness of antithrombotic therapy prescription was evaluated both at admission and at discharge. Univariable and multivariable logistic regression models were employed to verify whether or not socio-demographic (age >80years, living alone) and clinical features (previous or recent bleeding, cranio-facial trauma, cancer, dementia) modified the frequency and modalities of antithrombotic drugs prescription at admission and discharge. RESULTS: Among the 1332 REPOSI patients, 247 were admitted with AFF. At admission, CHADS(2) score was ≥ 2 in 68.4% of patients, at discharge in 75.9%. Among patients with AFF 26.5% at admission and 32.8% at discharge were not on any antithrombotic therapy, and 43.7% at admission and 40.9% at discharge were not taking an appropriate therapy according to the CHADS(2) score. The higher the level of cardio-embolic risk the higher was the percentage of antiplatelet- but not of VKAs-treated patients. At admission or at discharge, both at univariable and at multivariable logistic regression, only an age >80 years and a diagnosis of cancer, previous or active, had a statistically significant negative effect on VKAs prescription. Moreover, only a positive history of bleeding events (past or present) was independently associated to no VKA prescription at discharge in patients who were on VKA therapy at admission. If heparin was considered as an appropriate therapy for patients with indication for VKAs, the percentage of patients admitted or discharged on appropriate therapy became respectively 43.7% and 53.4%. CONCLUSION: Among elderly patients admitted with a diagnosis of AFF to internal medicine wards, an appropriate antithrombotic prophylaxis was taken by less than 50%, with an underuse of VKAs prescription independently of the level of cardio-embolic risk. Hospitalization did not improve the adherence to guidelines