Establishment And Evaluation Of Prediction Models For The Analysis Of Vascular-related Risk Factors In Type 2 Diabetic Erectile Dysfunction

Chapter One:Introduction:Overview of related mechanisms of vascular endothelial dysfunction in diabetes induced erectile dysfunctionDiabetes induced erectile dysfunction is one of the common complications in patients with diabetes.The pathogenesis of diabetes induced erectile dysfunction involves pathological changes in multiple systems such as blood vessels,nerves,muscles,and endocrine systems.Vascular endothelial dysfunction is an important influencing factor.Factors such as high blood glucose levels,advanced glycation end products,and increased oxidative stress can cause changes in NO-dependent vasodilation,promoting endothelial cell damage and apoptosis,affecting endothelial progenitor cell mobilization and recruitment,causing testosterone levels to fall,leading to vascular endothelial dysfunction and eventually erectile dysfunction.Chapter Two:Establishment and evaluation of prediction model for the analysis of vascular damage related risk factors in type 2 diabetes patients who developed erectile dysfunctionObjective:Analyzing of risk factors related to vascular damage in patients with type 2 diabetes who developed ED,and providing some clues for the future exploration of establishing individualized diagnostic tools.Methods:A total of 181 patients with type 2 diabetes were included for sexual function assessment,and clinical data in the patient system were retrieved.After preprocessing,the data are described by the number of cases and percentages.Statistical analysis was performed by R software(Version 3.6.1;https://www.r-project.org/).The Lasso regression model was used to optimize feature selection;the EPV method was used to evaluate whether the sample size for Logistic regression meets the requirements;and multivariable logistic regression analysis was applied to build a Logistic model incorporating the feature selected in the Lasso regression model.Then,establishing the DIED nomogram of this risk model.Discrimination,calibration,and clinical usefulness of the predicting model were assessed using the C-index,calibration plot,and decision curve analysis.Internal validation was assessed using the bootstrapping validation.Results:Erectile Dysfunction(ED)was diagnosed in 90(49.7%)of 181 patients.Predictors contained in the prediction nomogram included duration of diabetes(OR=4.440,P<0.05;OR=7.667,P<0.05),Status of c-IMT(OR=3.767,P<0.05),DR(OR=5.382,P<0.05),DKD(OR=4.959,P<0.05),LDL-C(OR=8.210,P<0.05),RDW(OR=2.418,P<0.05),Fbg(OR=4.649,P<0.05).According to the results of internal validation,the C-index of nomograms is 0.911;the calibration Hosmer-Lemeshow test was P=0.517;DCA curve showed that the nomogram had good clinical use.Conclusion:The multivariable logistic regression model and the nomogram was constructed using the duration of diabetes,Status of c-IMT,DR,DKD,LDL-C,RDW,Fbg.The internal validation showed that the nomogram displayed good discrimination,good calibration and was clinically useful.It provides some clues for further external validation and optimization based on larger population and multi-center data in the future.Chapter Three:Establishment and evaluation of prediction model for the analysis of vascular damage related risk factors in type 2 diabetic erectile dysfunction patients who developed severe erectile dysfunctionObjective:The clinical data of patients with type 2 diabetes induced erectile dysfunction were analyzed to explore the vascular-related clinical risk factors of severe diabetic erectile dysfunction.Methods:The Lasso regression model was used to optimize feature selection.Using EPV method to calculate whether the sample size required for Logistic regression analysis meets the requirements.Multivariable logistic regression analysis was applied to figure out the regression coefficients and OR values of variables.The internal validation based on Bootstrap method was used to evaluate the discrimination,calibration,and clinical usefulness of the predicting model.Results:Severe erectile dysfunction was diagnosed in 31(34%)of 90 patients.Predictors contained in the nomogram included duration of diabetes,DR,DKD.Through internal validation,the nomogram had good discrimination,calibration and clinical use.Conclusion:The logistic model for the analysis of vascular damage related risk factors in type 2 diabetic erectile dysfunction patients who developed severe erectile dysfunction was established using duration of diabetes,DR,DKD.It provides a basis for further exploration of risk factors for severe diabetic erectile dysfunction in the future.