| Objective:1.The clinical dataand literatures of children with Henoch-Schonlein purpura(HSP)were analyzed,to screen out the risk factors for children with Henoch-Schonlein purpura nephritis(HSPN),and to establish a risk prediction nomogram model for HSPN.2.To explore the mechanism of Xiaofeng Powder in the treatmet of HSP through network pharmacology,bioinformatics and in vivo experiments.Methods:1.Clinical data of children with HSP admitted to the affiliated hospital of Shandong university of TCM on January 2016 to October 2019 were studied.Observation tables were made and clinical data were recorded.The cases were divided into two groups according to the urine test results(HSPN group and NHSPN group).The univariate and multivariate analysis were used to determine the risk factors.2.Relevant studies on risk factors associated HSPN were searched through CNKI,CBM,VIP,Wang Fang database,Pub Med,Embase,and Web of Science from January 2000 to January 2020.The quality of all eligible studieswas assessed using the Newcastle-Ottawa scale(NOS)criteria and the data were extracted.Fixed or random effect models were used to analyse the risk factors after heterogeneity analysis.Publication bias were evaluated by funnel plot and Egger’s test.3.The results of clinical studies and literature research were compared and analyzed,and the prediction model of HSPN was established according to the selected risk factors.The area under the ROC curve(AUC),sensitivity and specificity were used to assess the clinical efficacy.4.The treatment of HSP was analyzed,and the law of clinical medication was summarized.The ingredients of Xiaofeng Powder and HSP related genes were searched by TTD、 Drug Bank、Pharm GKB and TCMSP databases.The mechanism of Xiaofeng Powder in the treatment of HSP was explored from the perspective of ingredients – gene-disease network.The animal experiments were designed to observe the effects of Xiaofeng Powder on HSP rats and to verify the key targets based on the results of network pharmacology analysis.Results:1.In this study,227 cases of HSP were studied.Clinical characteristics: HSP sends in cold season,with more boys than girls,mainly in school-age children,and the main inducement were infection.The incidence of skin purpura,abdominal pain,joint symptoms and renal invovement were 84.14%,49.34%,40.53% and 29.52% respectively.The mainly TCM syndrome types were wind-heat syndrome(29.52%)and damp-heat syndrome(49.78%),which often accompanied with blood stasis syndrome(48.02%).The univariate and multivariate analysis showed that,age ≥ 8 years old(OR:2.396,95%CI:1.217～4.718),persistent purpura ≥4 weeks(OR:3.387,95%CI:1.688～6.796),ALB(OR:0.876,95%CI:0.800～0.959),Cys-C level(OR:76.450,95%CI:3.987～1466.091)and damp-heat stasis syndrome(OR:2.309,95%CI:1.186～4.495)were the independent risk factors for HSPN(P<0.05).2.A total of 39 studies involving 6892 HSP cases and 2543 HSPN childrenwere included.They were all case-control studies,with definite diagnostic criteria and follow-up time and NOS score were 6～8.Elder age(OR:0.466,95%CI:0.329～0.602),persistent purpura(OR:3.860,95%CI:2.713 ～ 5.493),relapse(OR:3.104,95%CI:1.648 ～ 5.847),abdominal pain(OR:1.493,95%CI:1.167 ～ 1.911),severe abdominal pain(OR:2.036,95%CI:1.103～3.760),gastrointestinal bleeding(OR:2.025,95%CI:1.736～2.363),CNS involvement(OR:4.694,95%CI:1.492～14.766),obesity(OR:3.436,95%CI: 2.404～4.910),decreased complement 3(OR:2.251,95%CI:1.569 ～ 3.227),ALB(OR:-0.270,95%CI:-0.406 ～-0.135),TC(OR:0.597,95%CI:0.200 ～ 0.995)and ASO positive(OR:2.651,95%CI:1.101～6.384)were risk factors for HSPN(P<0.05).3.4 prediction models were established with age ≥ 8 years,abdominal pain,gastrointestinal bleeding,persistent purpura ≥ 4 weeks,relapse,ALB,Cys-C and dampness-heat stasis syndrome as parameters according to clinical and literature research.We found that,with the increase of parameters in the model,the AUC gradually increases by comparing the AUC.Finally,the most accurate prediction model included six factors:age,persistent purpura ≥ 4 weeks,relapse,ALB,Cys-C,and dampness-heat stasis syndrome.The AUC for the prediction model(Model 4)was 0.862(95%CI: 0.809～0.916),the sensitivity was 70.15% and specificity was 89.38%.4.Clinical analysis shows that Xiaofeng Powder was one of the main prescriptions for the treatment of HSP.Network pharmacology analysis shows that 64 active ingredients were screened from Xiaofeng Powder,which are acted on 25 related targets of HSP.Interleukin-6,Tumor necrosis factor,vascular endothelial growth factor,Prostaglandin G/H synthase 2 and Interleukin-1 beta might be important targets of Xiaofeng Powder in treatment of HSP.There were 193 GO items(P<0.05)in functional enrichment analysis of GO,11 signal pathways(P < 0.05,FDR < 0.05)in the KEGG pathway enrichment screening,involving Tumor signaling pathway,NOD-like receptor signaling pathway,VEGF signaling pathway,PI3K-Akt signaling pathway,NF-k B signaling pathway.5.Animal experiments showed that Xiaofeng Powder could reduce the proteinuria of HSP rats,lessen the pathological changes of kidney,and decrease the expression of serum IL-6 and IL-1B.Conclusion:1.Age≥8 years old,persistent purpura ≥4 weeks,ALB,Cys-C level and damp-heat stasis syndrome were independent risk factors for HSPN according to clinical analysis.2.Attention should be paid to the occurrence of kidney damage for children with older age,persistent purpura,relapse,abdominal pain,severe abdominal pain,gastrointestinal bleeding,CNS involvement,obesity,decreased C3,ALB,TC and ASO positive according to literature research.3.A model to predict the risk probability for children with renal involvement of HSP according to clinical and literature research.The model included age,persistent purpura ≥4 weeks,relapse,ALB,Cys-C,and dampness-heat stasis syndrome in the form of a nomogram.4.Xiaofeng Powder is effective in treating HSP in multi-component,multi-target and multi-link way.The therapeutic mechanism might be to regulate the expression of IL-6,TNF,IL-1B,CCX2 and other inflammatory targets by regulating NOD2-mediated MAPK and/or NF-k B signaling pathways. |
PDF Full Text Request