Analysis Of High-risk Factors For Purpura Nephritis In Children And The Development Of Its Prediction Model Nomogram

Research Background and Purpose : Henoch-Schonlein purpura(HSP)is a systemic syndrome in which the lesion is an inflammation of small blood vessels.It is common in children and adolescents,and mostly presents with thrombocytopenic purpura,abdominal pain,gastrointestinal bleeding,arthralgias,angioneurotic edema and renal damage,but also with testicular inflammation and damage to the central nervous system.The presence Henoch-schonlein purpura nephritis(HSPN)and its severity determine the prognosis of children with HSP.The gold standard for confirming the diagnosis of HSPN is kidney puncture biopsy.However,the technique of confirming diagnosis by puncture biopsy in pediatrics has problems such as low cooperation rate of children and low acceptance rate of parents.Current studies on risk factors for purpura nephritis are numerous,but exploring a model that can predict the risk of developing HSPN based on high risk factors for the occurrence of HSPN is less common.Exploring the high-risk factors of children with purpuric nephritis,and establishing the HSPN risk nomogram(Nomogram)based on the high-risk factors screened out,providing a more reliable reference basis for doctors to prevent and treat HSPN early.Research Objects and Methods: Children diagnosed with HSP attending the Affiliated Hospital of Yan’an University from January 1,2015 to December 31,2019 with2≤ age ≤ 14 years were retrospectively collected,and the data of 692 children were finally enrolled in this study according to the inclusion and exclusion criteria.The data were randomly divided into a modeling group(n=500)and a validation group(n=192)according to the ratio of 7:3.The modeling group was used to analyze the risk factors for the occurrence of HSPN and to establish a model.The validation group was used to test the reliability of the model.The test of the model was expressed in terms of discrimination and calibration.Result:According to the results of univariate analysis,children’s gender,age,season,history of upper respiratory tract infection,allergy history,abdominal symptoms,onset-tovisit time,rash duration,recurrence,CRE,Cys-c,β2-MG,ASO,PLT,APTT,D-dimer,and hs-CRP were correlated with the occurrence of HSPN(P < 0.1),The results of the single factor analysis were then incorporated into the multivariate logistic regression analysis,and the independent risk factors for the onset of HSPN were screened out: Cysc,history of allergy,APTT,abdominal symptoms,ASO,hs-CRP,age,recurrence,duration of rash and CRE.based on the multifactorial screening variables,a nomogram of the risk of HSPN development was plotted using R language software,and the analysis yielded a model discrimination AUC=0.828 and AUC=0.802 in the validation group,indicating that the established model has good discrimination,and plotting the calibration curve can be observed to analyze that the regression fitting curve of this model and the standard curve basically overlap,indicating that the model has good calibration and the predicted value of the model is close to the true value.The model has good accuracy and clinical usability.Conclusion:1.Cys-c,history of allergy,APTT,abdominal symptoms,ASO,hs-CRP,age,recurrence,duration of rash,and CRE are independent risk factors for the onset of HSPN.2.HSPN nomogram can predict the risk of renal damage in children with HSP,and the model has good accuracy and clinical utility.