Adequacy between prescriptions and CHADS2 score recommendations in geriatric patients

Atrial fibrillation (AF) is a major risk factor for stroke. Thromboprophylaxis with anticoagulant reduces the incidence of stroke and is warranted by the CHADS(2) recommendations when score >= 2. But such therapy remains underused particularly among elderly patients. The aim of our study was to evaluate the adequacy between prescriptions and CHADS(2) recommendations in geriatric hospitalised patients with AF Method: retrospective study set in the Geriatrics Department of the University hospital of Poitiers (France), of patients >75 y with AF, between July and December 2009. The description of the patients taken into the count: epidemiological data, functional daily activities (score GIR), cognitive assessment, antithrombotic treatment, and evaluation of the CHADS(2) and HEMORR(2)HAGES scores. Results: in this study161 hospitalisations were analysed, mean age of the patients was 874 +/- 5.4 years. Antithrombotic treatment was prescribed in 84% of cases. The overall conformity to CHADS(2) recommendations was 44%. Most of hospitalisations (88.9%) included patients with CHADS(2) score >= 2. Non-conformity rate was up to 60% in this group with 5 significant variables: MMSE score <26 (OR: 3.17 [95%IC: 1.23-8.17]), high risk of bleeding (2.88 [1.28-6.47]), dementia (2.46 [1.01-5.95]), functional impairment (GIR score <= 4) (2.43 [1.23-4.84]) and history of fall (2.42 [1.13-5.17]). Still, the variables explained only 19% of the non-conform prescriptions and no variable was significantly relevant on multivariate analysis. Discussion: our study show 56% of prescriptions that are non-conform to CHADS(2) recommendations, particularly in the elderly with a predictive embolic score >= 2. Antithrombotic treatment seems to be all the less conform as patients appear to be more vunerable. Due to the new score CHA(2)DS(2)-VASc, physicians really have to discuss the benefit-risk balance at an individual level.