Multivariate Logistic Regression Analysis And Risk Prediction Of Chronic Venous Leg Ulcers

Objective:To investigate the risk factors associated with chronic venous leg ulcers,in order to improve the understanding of chronic venous leg ulcers and the level of prevention and management.Method:A retrospective study was conducted on the clinical diagnosis and treatment data of 102 patients with chronic venous leg ulcer collected in the inpatient department of our hospital from December 2018 to September 2020,and 154 patients with non-venous leg ulcer were selected in the same period.The factors were measured retrospectively included age,sex,weight,diabetes,smoking,activity restriction,hypertension,Lipid-lowering treatment,hypoproteinemia,renal insufficiency,uric acid,urinary protein,hyperlipidemia,hemoglobin,neutrophils count,lymphocyte count.The above factors(independent variables)were first screened by univariate analysis,and then the independent variables that were significant by univariate analysis were subjected to multivariat logistic regression analysis.After the prediction model is established and the feasibility is confirmed,read the best segmentation point in the ROC curve of all data,and then obtain the predicted probability value of each patient according to the prediction model.The probability value is greater than the best segmentation point for chronic venous leg ulcers of high risk.Result : the measurement data were statistically analyzed by T-tests or U-tests,the enumeration data were statistically analyzed by Chi-square test or Fisher’s exact test,showed that sex,diabetes,hyperlipidemia,neutrophils count,activity restriction and hypoproteinemia(p<0.1)were influencing factors of chronic venous calf ulcers.Multifactorial logistic regression analysis showed that diabetes(p<0.05)、activity restriction(p<0.05)、hyperlipidemia(p<0.05)and neutrophils count(p<0.05)were independent risk factors for chronic venous leg ulcers.The lower neutrophils count,the higher the possibility of ulcer.The patients whose predicted probability value is greater than the best segment point 0.398 in the prediction model are those at high risk of developing chronic venous leg ulcers.Conclusion:Factors such as diabetes,activity restriction,hyperlipidemia and neutrophils count are risk factors for the development of chronic venous leg ulcers.The lower neutrophils count,the higher the possibility of ulcer.Patients with a predicted probability value greater than 0.398 are at high risk of chronic venous leg ulcers.Finding risk factors associated with chronic venous leg ulcers is a guide to reduce the incidence of clinical venous leg ulcers.