External validation of the simplified Geneva risk assessment model for hospital-associated venous thromboembolism in the Padua cohort

Background: The simplified Geneva risk assessment model (RAM) predicts the risk of hospitalization-related venous thromboembolism (VTE) in medical inpatients in its developmental cohort but has not been validated. Objectives: To externally validate the simplified Geneva RAM. Patients/Methods: For this secondary analysis of a prospective cohort set in Padua, we calculated the simplified Geneva RAM for all participants. They were followed up for 90 days for the occurrence of adjudicated VTE. Thirty- and 90-day risks of VTE were estimated by the Kaplan-Meier method, and categories of risks compared with a Cox regression model adjusted for the use of thromboprophylaxis. Results: Among 1180 medical inpatients, the 90-day risk of symptomatic VTE was 3.1%. The simplified Geneva RAM classified 56.9% as high risk (≥3 points; 90-day risk of VTE of 5.2%) and 43.1% as low risk (<3 points; 90-day risk of VTE of 0.4%). Compared with low-risk participants, high-risk participants had an 18-fold greater risk of VTE than low-risk participants (hazard ratio [HR] 17.9, 95% confidence interval [CI] 4.3-74.7). A very high-risk category (≥7 points) identified 5.3% of participants with a 9.5% probability of VTE at 90 days. Conclusions: In this external validation study, we confirm the excellent discrimination and clinically adequate calibration of the simplified Geneva RAM as a stratification tool to guide the use of thromboprophylaxis.