Influencing Factors And The Development And Validation Of A Predictive Model Of The Progression Of Diabetic Kidney Disease

The First Part-Analysis of risk factors for renal complications inpatients with type 2 diabetes mellitusDiabetic kidney disease(DKD)refers to chronic kidney disease(CKD)caused by diabetes.It is a clinical diagnosis based on an increase in urinary albumin and/or an decrease in estimated glomerular filtration rate while excluding other CKD.DKD is a common microvascular complication of diabetes.With the increase in the incidence of diabetes,DKD has gradually become an important cause of end-stage renal disease(ESRD).In developed countries such as the United States and Canada,DKD is the leading cause of end-stage renal disease,while the main cause of ESRD in China is still primary glomerulonephritis.However,with the increase in the incidence of diabetes,ESRD caused by DKD has been significantly increasing in recent years.DKD has become the leading cause of ESRD in areas with higher economic levels such as Beijing and Guangzhou.With economic development and improvement of living standards,DKD may become the leading cause of ESRD in China in the future.Therefore,finding risk factors for the progression of DKD is particularly important for the prevention of DKD.Based on the clinical characteristics and disease stage of DKD,this study is divided into the following two sections to explore the risk factors for the development of type 2 diabetes to DKD and the risk factors for the development of DKD patients in renal replacement therapy and establish the clinical prediction model.ObjectiveThe purpose of this study is to identify risk factors for renal complications in type 2 diabetes mellitus to guide the prevention of DKD.MethodsThe data of patients with type 2 diabetes mellitus who were hospitalized and followed up with detailed clinical data from October 2013 to January 2017 in the First Affiliated Hospital of Zhengzhou University were retrospectively collected.About 18 months of follow-up,the patients were divided into DM group and DKD group according to whether they have progressed to diabetic kidney disease.COX regression analysis was used to analyze the risk factors of progression to diabetic kidney disease in diabetic patients.The predictive value of risk factors in the risk of progression to diabetic kidney disease in diabetic patients was analyzed by plotting receiver operating characteristic curve.Results(1)There were 215 cases,102 cases in the DM group and 113 cases in the DKD group.Comparison of plateletocrit and β2 microglobulin between two groups was statistically significant(P<0.05).(2)Multivariate COX regression results showed that plateletocrit was an independent risk factor for the progression to diabetic kidney disease in diabetic patients(HR=1.588,95CI%1.071～2.355).(3)The area under the receiver operating characteristic curve for predicting the progression to diabetic kidney disease in diabetic patients of plateletocrit was 0.602(95%CI 0.527～0.678,p<0.05).ConclusionPlateletocrit is an independent risk factor for the progression of diabetes to diabetic kidney disease.With each 0.1%increase in plateletocrit,the risk of progression to diabetic kidney disease increases by 0.588 times.The Second Part-Development And Validation of A Predictive Model for the Progression of Diabetic Kidney Disease to Kidney FailureObjectiveWe aimed to use variables routinely measured in patients with diabetic kidney disease to create a model to predict progression to kidney failure.MethodsWe retrospectively assessed 641 patients with type 2 diabetic kidney disease.We used a combination of clinical guidance and univariate logistic regression with a backward stepwise method to select the relevant variables.The area under the receiver operating characteristic curve,net reclassification improvement and integrated discrimination improvement were used to assess the discriminatory ability of the models.The calibration was assessed by the Hosmer-Lemeshow(H-L)test.The goodness of fit of the models was evaluated by the Akaike information criterion(AIC).The best model was selected according to the optimal combination of discrimination and calibration.The we constructed a nomogram of the optimal model.ResultsWe found four convincing prognostic factors in the final nomogram:serum cystatin C levels,24-h urine protein levels and the neutrophil:lymphocyte ratio and the eGFR.The discrimination and calibration of the model comprised by these four factors performed better than any other models.ConclusionWe constructed a nomogram to predict the risk of patients with diabetic kidney disease initiating renal replacement in 3 years.Higher total points in the nomogram indicates a worse prognosis for the patient.