Prediction Of Risk Factors And Model Construction For Progression Of TCM Treatment Cohort For Diabetic Kidney Disease

ObjectiveThe aim of this study was to retrospectively analyze the data of patients with diabetic kidney disease(DKD)in our hospital from 2011 to 2016.To analyze the risk factors that affect the renal function progression and prognosis for patients with DKD under the treatment of traditional Chinese medicine(TCM),and respectively construct the prediction equation of renal function progression and prognostic nomogram for patients with DKD.Methods①Inpatients diagnosed with DKD at Guangdong Provincial Hospital of Traditional Chinese Medicine from 2011,to 2016 were enrolled in this study.The inclusion criteria were as follows:be treated with TCM;DKD as the primary diagnosis;no dialysis or kidney transplantation performed for DKD.Patients who had undergone renal replacement therapy less than 3 months after the initial diagnosis,those for whom the timing of the initial diagnosis of DKD was unclear,those with incomplete clinical data and time interval between serum creatinine data collection less than 1 month were excluded.The included patients were randomly divided into training cohort and validation cohort according to a ratio of 7:3.According to the median decline rate of renal function,the included patients were divided into renal function progressive group and non-progressive group.The logistic regression analysis was used to analyse risk factors associated withrenal function progression of DKD.And the prediction equation for renal function progression of DKD was constructed based on the risk factors above.ROC curve and Homser-Lemeshow test were used to test the prediction efficiency of the prediction equation.The external validation was verified by the validation cohort.②Patients enrolled in the study were followed up until December 31,2016.The endpoint was determined as renal replacement therapy or all-cause death.Cox regression analysis was used to analyse risk factors significantly related to the prognosis of DKD in the training cohort.The prognostic nomogram was designed based on the prognostic factors identified by Cox proportional hazards regression.The accuracy of the nomogram was tested by C-index and the consistency were tested by calibration curve.Internal validation was verified through Bootstrap method,and external validation was verified by the validation cohort.Results①340 patients with diabetic kidney disease were enrolled in the study and they were divided into training cohort(n=238)and validation cohort(n=102)according to a ratio of 7:3.According to the renal function decline rate of 0.64 ml·min-1·1.73m-2·year-1,there are 124 patients in the renal function progressive group and 114 patients in the non-progressive group in the training cohort.In the validation cohort,there were 46 patients in the renal function progressive group and 56 patients in the non-progressive group.Multivariate logistic regression analysis showed that,factors included syndrome of damp(OR:3.73,95%CI:1.30-10.72;P=0·014),BMI(OR:0.52,95%CI:0.29-0.92;P=0.026),24-hour urinary protein quantification[500~3500mg/24h(OR:3.51,95%CI:1.25-9.83;P=0.017),over 3500mg/24h(OR:6.74,95%CI:2.33-19.45;P<0.001)],low density lipoprotein(OR:2.19,95%CI:1.22-3.92:P=0.008),ACEI/ARB(OR:2.16,95%CI:1.21-3.85,P=0.009)were significantly associated with the renal function progression of DKD.However,no association was found between other TCM syndromes or the use of proprietary Chinese medicines and renal function progression of DKD.Using five risk factors above,the prediction equation of renal function progression of DKD was developed,and the area under ROC curve was 0.715(95%CI:0.651-0.779).The area under ROC curve of validation cohort is 0.723(95%CI:0.622-0.824).Hosmer-Lemeshow test showed good calibration in training cohort and validation cohort(P≥0.05).②Among 340 patients with DKD in survival analysis,136 were enrolled in renal replacement therapy,27 died,and 49 were lost to follow-up.The median renal survival duration was 25.63(95%CI:20.23-33.80)and 26.37(95%CI:18.93-38.50)months in the training cohort and validation cohort,respectively.Cox regression analyses demonstrated that the syndrome of damp-heat(HR:1.65,95%CI:1.07-2.56;P=0.024)、the baseline of eGFR(HR:0.97,95%CI:0.96-0.98;P<0.001),24-hour urine protein quantification(500~3500 mg/24 h(HR:4.85,95%CI:1.69-13.99;P=0.003),more than 3500 mg/24h(HR:5.61,95%CI:1.94-16.18,P=0.001),total cholesterol(HR:1.55,95%CI:1.01-2.37);P=0.046),the level of hemoglobin(HR:2.34,95%CI:1.54-3.56;P<0.001),diuretics(HR:1.57,95%CI:1.03-2.38;P=0.034)were prognostic factors for DKD.Although univariate Cox regression analysis found the use of proprietary Chinese medicines for alleviating renal failure was associated with the prognosis of DKD,but no significant association was found after multivariate Cox regression.The six risk factors were used to develop a prognostic nomogram for the prognosis risk of DKD.The Oindex for the training cohort and validation cohorts was 0.78(95%CI:0.74-0.82),and 0.79(95%CI:0.73-0.85),respectively.The calibration curves for the probability of renal survival at 1,2,3 years indicated good consistency between the prediction and the actual observation.ConelusionBased on the treatment of TCM,there are various risk factors that influence the progression and prognosis of DKD.The TCM syndrome was also significantly related to the progression of DKD,in which the single syndrome type syndrome of damp influenced the renal function progress of DKD,while the compound syndrome of damp-heat syndrome influenced the prognosis of DKD.In addition,urine protein quantification runs through the whole process of DKD and is significantly related to renal function progression and prognosis of DKD.The prediction equation for predicting renal function progression of DKD and the prognostic nomogram for predicting the prognosis of DKD we developed have important clinical value for early prediction of progression of DKD and early identification of high-risk patients.