Management of atrial fibrillation with oral anticoagulants and predictors in clinical practice in Italy

Objective: To determine the prevalence of atrial fibrillation (AF) and its management with oral anticoagulants (OACs), and to investigate clinical variables associated with their use. Methods: Primary care data were obtained from Italian general practitioners (GPs). This study included data from patients with AF on 31 December 2000 or 31 December 2004 with a ≥1 year continuous history with the same GP, and any prescription for OAC or antiplatelet drug (AP) in the calendar year immediately subsequent to these dates. Clinical variables were also recorded as potential correlates of OAC use, and assessed using a logistic regression analysis. Results: The observed prevalence of AF increased from 0.9% in 2000 to 1.7% in 2004, and 2.7% in 2000 and 6.4% in 2004 for patients aged >75 years. In 2000, prevalence may have been under-estimated if many patients had short histories with their GPs. The 2004 data are consistent with a US study (1.0% in the general population and 9.0% for patients aged >80 years [Go et al, 2001]) and a prior Italian study (1.8% in those aged >40 years [Fillippi et al, 2000]). In 2000, 28.8% of patients received OACs, 33.5% received APs, and 41.2% received no antithrombotic agent; in 2004, the use of antithrombotics increased: 32.9% received OACs, 39.6% received APs, and 32.1% received neither. Dichotomous risk classifications according to the CHADS score (≤1 vs ≥2) slightly affected OAC prescribing (Odds Ratio [OR] 1.28 [95% CI: 1.03–1.58]); however, an increasing CHADS2 risk score did not correlate with increased use of either an OAC or an AP. Stronger correlations were observed with concurrent valve disease (OR 1.78 [1.26–2.52]) and the NICE stroke risk classification (moderate risk OR 2.04 [1.57–2.66]; high risk OR 2.42 [1.63–3.58]). By 2004, OAC use was highest in those with concomitant valve disease (49.6%), heart failure (44.0%), and previous stroke or transient ischaemic attack (42.1%). Conclusion: Guidelines recommend the use of OAC for stroke prevention in AF. Our study shows that prescription of OACs correlates with variables that may influence GP prescribing behaviours. However, approximately 70% of eligible patients who might benefit from OAC therapy do not receive it: ∼40% received an AP and ∼32% received no antithrombotic therapy. While current improvements in management strategies may result in more patients receiving OAC, our findings highlight the significant gap between guideline recommendations and routine practice, and the need for improved strategies to extend appropriate OAC therapy to eligible patients.