| Objectives:To explore the risk factors of renal insufficiency in patients with idiopathic membranous nephropathy(IMN),and establish a predictive model in order to provide reference for early treatment.Methods:A retrospective analysis was made on the patients who were pathologically diagnosed as membranous nephropathy by renal biopsy in the Department of Nephrology,the First Affiliated Hospital of Jilin University from May 2017 to October 2020.According to estimated glomerular filtration rate(e GFR),patients were divided in two groups:normal e GFR group(e GFR≥90m L/min/1.73m~2)and decreased e GFR group(e GFR<90m L/min/1.73m~2).R4.0.2 software was used for statistical analysis.The independent risk factors of renal insufficiency in patients with IMN were screened and predicted,based on which to establish a predictive model,and comprehensively evaluated the model.Clinical data included gender,age,height,weight,body mass index(BMI),history of hypertension and diabetes.Laboratory indexes included pathological stage,aspartate aminotransferase(AST),alanine aminotransferase(ALT),total protein(TP),albumin(ALB),blood urea nitrogen(BUN),serum creatinine(SCr),retinol binding protein(RBP),cystatin C(CysC),uric acid(UA),total cholesterol(TCHO),triglycerides(TG),high-density lipoprotein cholesterol(HDL-C),low-density lipoprotein cholesterol(LDL-C),white blood cell count(WBC),hemoglobin(HGB),platelets(PLT),thyroid-stimulating hormone(TSH),free triiodothyronine(FT3),free thyroxine(FT4),immunoglobulin G(Ig G),immunoglobulin A(Ig A),immunoglobulin M(Ig M),complement 3(C3),urinary red blood cells(URBC),24-hour urine total protein(UTP),anti-phospholipase A2receptor(PLA2R)antibody.Results:1.Among the 594 patients with IMN,394 were males and 200 were females,the male-female ratio was 1.97:1.The age range was from 18 to 75 years old,including152 elderly patients(age≥60 years old),277 middle-aged patients(45 years old≤age<60 years old),and 165 young patients(18 years old≤age<45 years old).The mean height was 1.69±0.64m,the mean weight was 71.93±12.67kg,and the mean BMI was25.85±3.93kg/m~2.There were 270 patients with a history of hypertension,accounting for 45.45%,and a total of 56 patients with a history of diabetes,accounting for 9.43%.There were 444 patients in the normal e GFR group and 150 patients in the decreased e GFR group.Decreased e GFR group had higher incidence of diabetes and BMI,and there was statistically different between the two groups(P<0.05),Compared with the normal e GFR group,the proportion of males in the decreased e GFR group was higher,the proportion of elderly patients was larger,body weight and the incidence of hypertension were higher.The difference between the two groups was statistically significant(P<0.01).There was no difference in height between two groups(P>0.05).2.FT4 level of the decreased e GFR group was lower and TCHO level of the decreased e GFR group was higher than that of the normal e GFR group,which were statistically different(P<0.05).TP,ALT,ALB,FT3,Ig G levels of the decreased e GFR group were lower,and blood RBP,HGB,BUN,SCr,CysC,TG,UTP,PLA2R antibody levels of the decreased e GFR group were higher than that of the normal e GFR group,and the difference was statistically significant(P<0.01).There was no difference between the two groups in pathological stage,UA,AST,HDL-C,LDL-C,WBC,PLT,TSH,Ig A,Ig M,C3,URBC levels(P>0.05).3.Univariate logistic regression analysis showed that compared with the normal e GFR group,the decreased e GFR group had lower ALT,FT4,Ig G levels and higher incidence of diabetes and TG levels,which were statistically different(P<0.05).The incidence of hypertension,weight,BUN,RBP,CysC,UTP,PLA2R antibody levels in the decreased e GFR group were higher than those in the normal e GFR group,and the levels of TP,ALB,HGB,and FT3 level of the decreased e GFR group were lower than those in the normal e GFR group,and the difference was statistically significant(P<0.01).There was no statistical difference in height,BMI,AST,UA,TCHO,HDL-C,LDL-C,WBC,PLT,TSH,Ig A,Ig M,C3,URBC between the two groups(P>0.05).In the results of univariate logistic regression analysis,a total of 16indicators with statistical differences and significant statistical differences between the two groups were selected for multivariate logistic regression analysis.The results showed that the decreased e GFR group had higher incidence of hypertension and UTP levels compared with the normal e GFR group,which were statistically different(P<0.05).The levels of BUN and CysC in the decreased e GFR group were higher than those in the normal e GFR group,and the FT3 level of the decreased e GFR group was lower than those in the normal e GFR group and the difference was statistically significant(P<0.01).With hypertension,blood urea nitrogen,cystatin C,24-hour urine total protein increased by 1 unit,free T3 decreased by 1 unit,the risk of renal insufficiency increased by 1.695,1.266,26.275,1.067,and 0.573 times respectively.For diabetes history,weight,ALT,TP,ALB,RBP,TG,HGB,FT4,Ig G,PLA2R antibody,there was no difference between the two groups(P>0.05).4.Establishment of the model:based on the five indicators with statistical differences and significant statistical differences selected by multivariate logistic regression analysis,a clinical prediction model for predicting the risk of renal insufficiency in IMN patients was constructed.This model suggested that IMN patients with a previous history of hypertension,high levels of BUN,CysC,UTP,and low level of FT3 were at greater risk of renal insufficiency.In addition,RBP was added as a supplementary model according to references and clinical experience.5.Evaluation of the model:1)net reclassification index(NRI)and integrated discrimination improvement(IDI)were used to compare the original model and the supplementary model.The P values of NRI,NRI+,and NRI-were 0.809,0.526,0.518(P>0.05),respectively,and the P value of IDI was 0.105(P>0.05),which indicated that there was no significant difference between the two models,so the original model was finally selected.2)The AUC value of the ROC curve of the prediction model was 0.845(95%CI:0.810～0.880),and the specificity,sensitivity,and cutoff value were 0.736,0.800,and 0.388,respectively.3)Hosmer-Lemeshow test:the chi-square value was 8.784,the degree of freedom was 8,and the P value was 0.361(P>0.05).4)The calibration curve showed that the model was well calibrated,and the decision curve analysis(DCA)showed that IMN patients might get desirable clinical benefits from this model.Finally,a nomogram was used to display the prediction model.Conclusions:1.A history of hypertension,high levels of BUN,blood CysC,UTP and low level of FT3 are independent risk factors of renal insufficiency in patients with IMN.2.The blood RBP level also has a certain predictive effect on the renal insufficiency of IMN patients,but its predictive power has no obvious advantage over other screened predictors.3.The establishment of prediction model based on multivariate logistic regression analysis can help predict the risk of renal insufficiency in IMN patients,so as to more actively intervene in risk factors. |
PDF Full Text Request