Objective To construct a Nomogram model for predicting purpuric nephritis in children based on the age, neutrophil/lymphocyte ratio (NLR) and mean platelet volume (MPV).
Methods The clinical data of 332 children with HSP were retrospectively collected, and the children were randomly divided into the modeling group and verification group using the R language sample function at a ratio of 7∶3. According to the diagnostic criteria of HSPN, the two groups were divided into the HSP group and HSPN group. The modeling group was used to analyze the risk factors of children's HSPN, and the variables selected by R language software were used to establish a nomogram prediction model. The receiver operating characteristic (ROC) curves of the modeling group and validation group were plotted by MedCalc software to obtain the AUC of two groups. The calibration curve and decision curve were drawn to evaluate the efficiency of the prediction model.
Results The results of multivariate logistic regression analysis showed that the age, NLR and MPV were the independent risk factors of HSPN (OR = 1.204, 1.716, 1.426, P ＜ 0.05 to P ＜ 0.01). The C index of the modeling group and verification group were 0.84 and 0.86, respectively. The area under ROC curve (AUC) of the modeling group and verification group were 0.844 (95%CI: 0.791 ~ 0.888) and 0.860 (95%CI: 0.776 ~ 0.921), respectively. The results of calibration curve showed that there was a good agreement between the predicted probability and actual probability, and the results of decision curve analysis showed that the nomogram had a high value in clinical application.
Conclusions The age, NLR and MPV are associated with the renal damage in HSP, and risk factors of renal damage. The Nomogram has high predictive accuracy, and can more reliably predict the risk of kidney damage in children with anaphylactoid purpura, which has good clinical application value.
DownLoad: