Kuperman et al. found that patients with anemia had a higher risk of major bleeding (RR 2.84; 95% CI 2.52-3.39) in RIETE database. Anemia appeared to be an independent predictive factor for major bleeding [hazard ratio (HR) 1.95; 95% CI 1.72-2.20] in this registry. Unfortunately, selection bias due to enrolled patients does not allowed us to use these major results in ambulatory care. The aim of SCORE study was to refine bleeding risk estimation in French vitamin K antagonist (VKA) treated patients and to identifying one or several parameters of prognostic significance. We conducted a prospective, multi-center cohort study of 962 consecutive outpatients from private angiologic offices, clinics and hospitals enrolled in grenoble angiologic network for thromboembolic diseases between May 2009 and December 2010, followed during 1 year by their general practitioner. Main outcome was the occurrence of major bleeding or clinically non major relevant bleeding (CNMRB). Incidence rates major bleeding and CNMRB were 2.86 (95% CI 1.95-4.2) events per 100 patient-years and 12% (95% CI 9.89-14.11) respectively. Cox multivariate analyses showed that only anemia was strongly associated with a risk of major bleeding (HR 6.1; 95% CI 2.7-13.8; p = 0.001). Logistic regression analyses performed in CNMRB showed that anemia, prior gastro-intestinal bleeding and antiplatelet drug use were strongly associated with a risk of CNMRB at 1 year, respectively OR 2.53, 95% CI (1.4-4.56); p = 0.002, OR 3.32, 95% CI (1.51-7.31); p = 0.003 and OR 1.77, 95% CI (1.1-2.83); p = 0.017. These new data were consistent between major and CRNM bleeding in VKA treated patients. The key role of anemia should be confirmed in other prospective cohort studies, with different anticoagulants use such as direct oral anticoagulant in ambulatory care settings.
Keywords: Bleeding; Primary care; Prognosis; Risk assessment; Vitamin K antagonist.