Development and validation of a nomogram to predict recurrence in children with Henoch-Schönlein purpura.

Background: To screen risk factors for the recurrence in children with Henoch-Schönlein Purpura (HSP) and to develop and validate a nomogram for recurrence in children with HSP.

Methods: During September 2019 and September 2021, 212 children with HSP were selected in this study. The children were divided into two sets in a proportion of 7:3 using R language, with the first group as the training sets and the second as the internal validation sets. The related variables were analyzed by univariate and multivariate logistic regression analyses, and a nomogram for predicting the recurrence in HSP children was established. The nomogram was evaluated by ROC curve, calibration curve and decision curve, and 1000 times bootstrap resampling method was used to verify the model internally.

Results: Univariate and multivariate regression analyses identified respiratory infection, without preventive medication and diet restriction, age, allergen positive and abnormal urine routine as risk factors for the recurrence in children with HSP. Those risk factors were used to construct a predictive nomogram. The calibration curves revealed excellent accuracy of the predictive nomogram model, internally and externally.

Conclusions: We constructed and validated a clinical nomogram to predict the recurrence in children with HSP. We confirmed that respiratory tract infection, without preventive medication and diet restriction, age, allergen positive and abnormal urine routine were independent recurrence risk factors. This nomogram had a good performance in clinical decision-making.