Construction and efficiency analysis of prediction model for venous thromboembolism risk in the elderly after hip fracture.

OBJECTIVES:To screen the risk factors for predicting venous thromboembolism (VTE) risk after hip fracture in the elderly, to establish a prediction model based on these factors, and to analyze its prediction efficacy. METHODS:A total of 52 hip fracture patients over 60 years old with VTE admitted to the Department of Orthopaedic Trauma, Xiangya Hospital, Central South University from March 2017 to April 2019 were selected as a thrombus group, and another 52 hip fracture patients over 60 years old without VTE were selected as a control group. The differences of hospitalization data and examination results between the 2 groups were compared. Logistic regression model was used to explore the influence of risk factors on VTE risk after hip fracture in the elderly and construct the prediction model based on these factors. The receiver operating characteristic curve was used to analyze the predictive effectiveness of model, Hosmer-lemeshow goodness of fit test was used to evaluate the fitting degree of prediction model. RESULTS:Univariate analysis showed that injury-admission interval, Caprini score, WBC count, platelet count, neutrophil count, neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio, monocyte-to-lymphocyte ratio, systemic immune-inflammatory index (SII), and fibrinogen in the thrombus group were higher than those in the control group (all P<0.05). Logistic regression analysis showed that injury-admission interval, Caprini score, and SII were independent predictors of VTE risk after hip fracture in the elderly. The AUC was 0.949 (95% CI 0.901 to 0.996) when the sensitivity and specificity were 82.70% and 96.20%, respectively, which were significantly higher than each single index, and the prediction model had perfect fitting degree (Hosmer-lemeshow χ2=14.078, P>0.05). CONCLUSIONS:SII, Caprini score, and injury-admission interval are independent predictors of VTE after hip fracture in the elderly. The prediction model based on these 3 factors has a good efficacy on the prediction of VTE risk, and could provide important reference for the prevention, management, and treatment of VTE after hip fracture in the elderly.