Value of Padua risk assessment model for evaluating venous thromboembolism of stroke patients

Abstract

Objective To explore the value of Padua risk assessment modelfor evaluating venous thromboembolism (VTE) of stroke patients. Methods VTE screening was performed in 169 patients with stroke who were hospitalized in Beijing Shijitan Hospital between January 2018 and June 2019. Fifty⁃six patients were identified in VTE group and 113 patients in non⁃VTE group. The difference of risk factors distribution in Padua risk assessment model was compared between 2 groups. Univariate and multivariate Logistic regression were used to analyze the risk factors for VTE in stroke patients. The risk stratification of VTE was calculated by Padua risk assessment model and different scores were used as the boundary value of risk stratification to fit the receiver operating characteristic curve (ROC), which evaluated the validity of Padua risk assessment model in predicting VTE risk patients. Results The proportions of atrial fibrillation (P=0.024), elderly age (P=0.000), heart and/or respiratory failure (P=0.000), acute infection and/or rheumatologic disorder (P=0.000), active cancer (P=0.016), previous VTE (P=0.007), reduced mobility(P=0.009)in VTE group were significantly higherthan thosein non⁃VTE group. Univariate and multivariate Logistics regression analysis were used to evaluate the risk factors of VTE. In VET group, atrial fibrillation (OR= 3.203, 95%CI: 1.172-8.751; P= 0.023), elderly age (OR= 3.469, 95%CI: 1.603-7.508; P=0.002), heart and/or respiratory failure (OR=4.017, 95%CI: 1.315-12.274; P=0.015), acute infection and/or rheumatologic disorder (OR=3.472, 95%CI: 1.457-8.271; P=0.005), previous VTE (OR=5.884, 95%CI: 1.068-32.408; P=0.042) were significantly associated with VTE. For Padua risk assessment model,the ROC yielded an area underthe curve (AUC) of 0.762 and the standard error ofthe area is 0.040 (95%CI: 0.689-0.854, P=0.000). At the cutoff point of 4, Youden index got the value of 0.404 and its sensitivity and specificity was 58.93% and 85.83% and corresponding positive and negative predictive value was 53.34% and 83.08%. Padua risk assessment model can effectively assess the risk of VTE at the cutoff point 4. Conclusions Elderly age, previous VTE and critical and severe stroke are VTE high⁃risk factors. Padua risk assessment model can effectively assess the risk of VTE among patients afterstroke.

DOI：10.3969/j.issn.1672⁃6731.2020.08.013