Research On The Application Of Venous Thromboembolism Risk Assessment Tools In Hospitalized Patients

Venous thromboembolism has brought about a threat to the lives of patients, and caused a huge burden to the patients’ life and medical resources. Therefore, venous thromboembolism has always been a focus in the medical field. How to effectively identify the risk factors of venous thromboembolism, timely and accurately assess the risk of venous thromboembolism, is highly concerned by many medical professors. Study on venous thromboembolism risk assessment abroad has become an important research direction. At present, our country has no risk assessment tools in hospitalized patients with venous thromboembolism. It is necessary to introduce the venous thromboembolism risk assessment tool which has been widely validated abroad. Venous thromboembolism risk assessment tool can provide evaluation criteria and intervention measures, reduce the incidence of venous thromboembolism in hospitalized patients.Objectives:(1) To investigate the risk factors associated with VTE in hospitalized patients and the difference of risk factor distribution between medical and surgical department, providing evidence for appropriate precautions in clinical nursing practice.(2) To translate the Caprini risk assessment model into Chinese, and evaluate the validity of Caprini risk assessment model in identifying venous thromboembolism risk patients among hospitalized patients. To compare the validity of Caprini risk assessment model in identifying venous thromboembolism risk patients between hospitalized medical and surgical patients.(3) To translate the Padua risk assessment model into Chinese, and compare the validity between Caprini risk assessment model and Padua risk assessment model in identifying venous thromboembolism risk patients among hospitalized medical patients.Methods:There are two parts in this study. Case-control study is adopted in every part, which is a kind of observational study.PartⅠ:To translate the Caprini risk assessment model into Chinese, use this tool to assess 1078 hospitalized patients, and evaluate the sensitivity, specificity and ROC curve. To compare the validity of Caprini risk assessment model in identifying venous thromboembolism risk patients between hospitalized medical and surgical patients. Collecting patients diagnosed with VTE according to their medical records, totally 539, in a certain top three hospital as VTE group form September 2011 to September 2014. Using mechanical sampling method to choose another 539 patients without VTE as the control group, who were treated in the same spot in the period. Recording Baseline information, risk factors related to VTE and laboratory indicators of patients as primary data. T test was used for measurement data, Chi-square test or Fisher’s test for categorical data and Logistic regression analysis was used for multivariate analysis. Using Caprini risk assessment model to assess 1078 hospitalized patients, and evaluate the sensitivity, specificity and ROC curve of this tool. And to compare the validity of Caprini risk assessment model in identifying venous thromboembolism risk patients between hospitalized medical and surgical patients.PartⅡ:To translate the Padua risk assessment model into Chinese, use this tool to assess 640 hospitalized medical patients, and compare the the sensitivity, specificity, positive predict value, negative predict value and ROC curve between Caprini risk assessment model and Padua risk assessment model in identifying venous thromboembolism risk patients among hospitalized medical patients.Results:PartⅠ:The result of Logistic regression analysis showed that there are significant difference in age, BMI>25, surgery, history of surgery, malignant tumor, central venous catheter, stroke, fracture of lower limb or pelvis, knee arthroplasty between groups(P<0.05). In terms of medical patients, we found significant difference in age, BMI>25, malignant tumor, lying in bed(>72h), central venous catheter, stroke (P<0.05). For surgical patients significant difference was observed in BMI>25, surgery, history of surgery, fracture of lower limb or pelvis, knee arthroplasty between groups(P<0.05).Compared with a low risk, those with a moderate risk were associated with 2.13 fold increased risk of VTE, a high risk was associated with 5.67 fold increased risk of VTE, a highest risk was associated with 18.70 fold increased risk of VTE for Caprini risk score in hospitalized patients. The receiver operating characteristics(ROC) yielded an area under the curve(AUC) of 0.774[95%CI:0.746-0.802] for hospitalized patients. At the cut-off point of 2, its sensitivity/specificity was 75.14%/65.86%, and the corresponding positive/negative predictive value was 68.76%/72.60%. For hospitalized medical patients, the receiver operating characteristics(ROC) yielded an area under the curve(AUC) of 0.770[95%CI:0.733-0.806]. For hospitalized surgical patients, the receiver operating characteristics(ROC) yielded an area under the curve(AUC) of 0.793[95%CI:0.751-0.835]. The sensitivity of hospitalized surgical patients is higher than hospitalized medical patients, was 81.27%/70.94%.Part Ⅱ:For medical patients, the scores of Cprini risk assessment model were inconsistent with the scores of Padua risk assessment model. For hospitalized medical patients, Padua risk assessment model can identify patients with VTE 32.19% for high-risk patients, Caprini risk assessment model can identify patients with VTE 70.94% for high-risk patients; for Caprini risk assessment model, the receiver operating characteristics (ROC) yielded an area under the curve (AUC) of 0.770[95%CI:0.733-0.806].At the cut-off point of 2, its sensitivity/specificity was 70.94%/73.44%. But for Padua risk assessment model, the receiver operating characteristics (ROC) yielded an area under the curve (AUC) of 0.609[95%CI:0.556-0.653]. At the cut-off point of 3, its sensitivity/specificity was 32.19%/85.63%. For Caprini risk assessment model, when the cut-off point was 2, youden’s index got the maximum value of 0.443, but for Padua risk assessment model, youden’s index could get the maximum value, when the cut-off point was 2 instead of 3.Conclusion:Caprini risk assessment model can effectively and quantitatively assess the risk of VTE among hospitalized patients based on their individual VTE risk factor. Compare with Padua risk assessment model, Caprini risk assessment model can be more effective for assessing hospitalized medical patients.The distribution of risk factors the between medical and surgical department is different, providing evidence for appropriate precautions in clinical nursing practice.