Value of Padua Risk Assessment Model in Evaluating Venous Thromboembolism of Hospitalized Patients in the Department of Internal Medicine

  Objective   The aim of this study was to investigate the status of venous thromboembolism(VTE)in patients in the department of internal medicine and to evaluate whether Padua risk assessment modelcan be applied to this patient population.

  Methods   Baseline information, risk factors, prevention, and the incidence of VTE in in-patients of the department of internal medicine were collected and analyzed at Peking Union Medical College Hospital from May 17, 2016, to September 9, 2016. Patients with or without VTE were compared during hospitalization and within 3 months after discharge. Correlations between risk factors and VTE events were calculated and risk stratification was calculated by Padua risk assessment model. The predictive value was assessed by Logistic regression analysis.

  Results   A total of 3115 patients were enrolled in this study. The incidence of VTE was 2.5%(78/3115). The sensitivity and specificity of Padua risk assessment model were 83.3% and 62.3%, respectively, and 38.8% of patients had a high risk of VTE based on this model. The four risk factors embodied in the model including elderly age(≥ 70 years), acute myocardial infarction or ischemic stroke, obesity(body mass index ≥ 30 kg/m²), and recent(≤ 1 month)trauma and/or surgery, showed no statistical difference between patients with and without VTE(P>0.05). However, other risk factors that are not included in Padua risk assessment model, i.e. use of estrogen or progesterone, blood transfusion, and mechanical ventilation showed statistically different between VTE and non-VTE patients(P < 0.01). The ratio of receiving preventive treatment with anticoagulant drugs or machines was significantly higher in high-risk patients than those in low-risk ones(P < 0.01); 85.3% of high-risk patients did not receive anticoagulant or mechanical prevention and did not have VTE either.

  Conclusions   Padua risk assessment model shows low specificity. Several risk factors and their weight distribution in the model are not suitable, which leads to the limitation of this model in VTE risk assessment for in-patients of the department of internal medicine.