ObjectiveTo develop a nomogram for the use of predicting prognosis of patients with LN.To explore risk factors for TIL,So as to provide ascientific basis for assessment of LN progression and to provide decision-making for the prevention and treatment of LN.Methods(1)A historical cohort study was used in this study.All patients wh o had been diagnosed as LN by renal biopsy were from Guangdong Province T raditional Chinese Medical Hospital(January 1,2006 to December 31,2016).A total of 202 patients were enrolled in the study.The related baseline indicators such as demography,clinicopathology and TCM syndrome were coll ected,and the follow-up data were collected.(1)Doubled Serum creatinine,eGFR decreased more than 50%,ESRD,dialysis,renal transplantation were def ined as the composite endpoint.The Cox proportional risk regression model was used to screen the factors associated with the composite endpoint of LN.The Kaplan-meier survival curve was used to calculate the cumulative survival rate of LN patients,and the logarithmic rank test(Log-rank)wa s used to compare the groups.(2)Based on the variables related to prognos is of LN patients screened by multivariate Cox regression,predictive mod els with different variables were established.C-index,AIC and ROC curves were used to evaluate the accuracy of different models.The optimal predic tion model was selected to develop a nomogram for the use of predicting p rognosis in LN patients at a certain time.(3)Multivariate logistic regres sion was used to evaluate related risk factors affecting the exacerbation of TIL in LN.Results(1)A total of 202 patients were included in this study.The median age was 27(22-38)years at biopsy time.173 cases were female,accounting for85.6%.The median course of disease was 7(2.5-15)months at biopsy time.Ac cording to ISN/RPS 2003 classification,6 patients were classified as clas s Ⅱ(3.0%),58 cases as class Ⅲ(28.7%,including 21 cases of class Ⅲ+Ⅴ),117 cases as class Ⅳ(57.9%,72 cases of class Ⅳ-global(Ⅳ-G)group,14 cases of class Ⅳ-segmental(Ⅳ-S)group,31 cases of class Ⅳ+Ⅴ),and 21 cases as class Ⅴ(10.4%).There was no cases of class Ⅰ and Ⅴ in our st udy.The median follow-up time was 32(18-46)months.A total of 23 patient s(11.3%)had composite endpoint.①As compared with the non-end-point group,the end-point group have m ore severe baseline renal function(P<0.05),a higher rate of 24-hour uri nary protein excretion and triglyceride(P<0.05),a lower haemoglobin con centrations(P<0.05).(2)As compared with the non-end-point group,the end-point group have m ore severe baseline renal pathology including an increased percentage of type IV,spherical sclerosis,cellular crescent and fibrous crescent,more a dvanced TIL(P<0.05).Activity index(AI)and chronic ity index(CI)score were also higher(P<0.05).③As compared with the non-end-point group,the end-point group have a greater proportion of blood stasis and number of subordinate syndromes(P<0.05).④Cox regression univariate analysis showed that hypertension,CKD sta ge,serum creatinine,eGFR,blood urea nitrogen,hemoglobin,blood complement C3,24-hour urinary protein excretion,spherical sclerosis,crescent,interst itial inflammation,interstitial fibrosis and tubular atrophy,AI,CI,blood stasis,number of subordinate syndromes are closely associated with compos ite endpoint of LN.Multivariate Cox regression analysis showed that serum creatinine,CI,CKD stage,fibrous crescent,interstitial inflammation,inters titial fibrosis and tubular atrophy,blood stasis are independent risk fac tors of composite endpoint of LN.⑤Kaplan-Meier survival analysis showed that the more severe TIL or t he worse renal function,the higher risk of composite endpoint.LN with blo od stasis have higher risk of composite endpoint than that without blood stasis.⑥According to multivariate Cox regression,four nomograms including i ndex for CKD stage,chronicity index CI,fibrous crescent,interstitial infl ammation,interstitial fibrosis and tubular atrophy,blood stasis were desi gned.The best model,within which included CKD stage,chronicity index CI(HR=1.426,95%CI:1.010~2.015,P=0.034)and blood stasis(HR=4.602,95%CI:1.266~16.726,P=0.020)demonstrated good discrimination to predict 3-year,5-year and 8-year non-outcome survival.The nomogram based on above model al so performed good calibration.(2)A total of 161 patients(79.7%)have TIL,which were divided into no n-lesions group(41 cases,20.3%),mild lesions group(108 cases,53.4%),mo derate lesions group(31 cases,15.4%)and severe lesions group(22 cases,10.9%)based on NIH semi-quantitative score.①Relationship between TIL and the clinical findings:With increase of the severity of TIL,renal function became worse,24-hour urinary proteinexcretion increased,the proportion of hypertension and mean arterial pres sure increased gradually,haemoglobin concentrations decreased gradually(P<0.05).②Relationship between TIL and pathologic findings:With increase of the severity of TIL,The proportion of type Ⅳ in moderate and severe lesi ons group was higher than that in mild and non-lesion groups.The proporti on of moderate to severe lesions in type Ⅲ and Ⅳ was higher than that i n type Ⅱ and Ⅴ,the proportion of non-mild lesions was lower than that i n type Ⅱ and Ⅴ(P<0.05).chronicity index CI,spherical sclerosis,segm ental sclerosis and crescent in moderate to severe lesions group was high er than that in non-mild lesions group(P<0.05).With increase of the sev erity of TIL,the percentage of spherical sclerosis,segmental sclerosis,cr escent and chronicity index CI increased gradually.③Relationship between TIL and TCM syndromes:With increase of the sev erity of TIL,the percentage of blood stasis and numbers of subordinate sy ndromes increased gradually.④Univariate and Multivariate logistic regression analysis showed tha t serum creatinine(β:0.014,OR:1.014,95%CI:0.010~0.039,P=0.015),24-hour urinary protein excretion(β:0.573,OR:1.774,95%CI:0.308~0.838,P=0.000),interstitial inflammation and blood stasis(p:1.158,OR:3.148,95%CI:0.125~2.892,P=0.021)were independent risk factors for inter stitial fibrosis and tubular atrophy.24-hour urinary protein excretion(β:0.396,OR:1.486,95%CI:0.143~0.872,P=0.012),fire-toxicity syndr ome(β:0.927,OR:2.527,95%CI:0.136~3.768,P=0.042),blood stasis(β:0.795,OR:2.214,95%CI:0.084~2.672,P=0.036)were independent risk fact ors for interstitial inflammation.Conclusion(1)Serum creatinine,CI,CKD stage,fibrous crescent,interstitial inflamm ation,interstitial fibrosis and tubular atrophy,blood stasis were indepen dent risk factors of composite endpoint of LN.The nomogram based on CI,CK D stage,blood stasis can accurately predict 3-year,5-year and 8-year non-outcome survival and provide scientific decision-making for the preventio n and treatment of LN.(2)With increase of the severity of TIL,the condition of LN patients became worse.TIL determined the reanl outcome in patients with LN.Multivariate logistic regression model analysis showed that serum creatinine,24-hour urinary protein excretion,interstitial inflammation and blood stasis were independent risk factors for interstitial fibrosis and tubular atrophy,24-hour urinary protein excretion,fire-toxicity syndrome,blood stasis were independent risk factors for interstitial inflammation.Early identification of risk factors and active intervention of reversible factors arefavourable for delaying or preventingthe progression of LN. |