Estimating risk of venous-thromboembolic events in hospitalized medical patients: Comparison between 2008 and 2012 guidelines

Background: Prophylaxis for hospitalized venous-thromboembolic events (VTEs) is frequently underutilized, in part due to lack of a simple risk assessment model (RAM). Objectives: To compare patient selection and administration of VTE prophylaxis according to the American College of Chest Physicians (ACCP) 2008 guidelines versus the newer 2012 guidelines, and assess the feasibility of developing simpler local RAMs. Methods: We conducted a prospective assessment of VTE risk among 300 unselected consecutive patients admitted to a medical hospital ward, using the 2008 and 2012 ACCP guidelines. The frequency and relative weight of each risk factor in the 2012 ACCP guidelines were used to develop a local VTE RAM. Results: VTE prophylaxis was indicated by the 2008 and 2012 ACCP guidelines in 40% and 42% of the cohort respectively, and was administered in 28% and 26% of eligible patients, respectively. Contraindication to VTE prophylaxis was found in 29% of patients according to both guidelines. In comparison to the 2008 guidelines, sensitivity and specificity of the 2012 guidelines were 96% and 88%, respectively. A local RAM based on the following concise score, comprising age, malignancy and immobility, correctly identified 99% of at-risk patients based on the 2012 guidelines, with a sensitivity and specificity of 98% and 95%, respectively. Conclusions: Both guidelines performed to a similar degree and were poorly implemented in daily practice. A simplified RAM accurately identified the vast majority of these eligible patients. The development of local RAMs is feasible and may result in higher utilization rates.