Construction of a Predictive Model for Recurrence Risk of Childhood IgA Vasculitis and the Correlation Analysis between Chinese Medicine Syndrome Elements and Recurrence
ZHANG Boyan, et al
Dongzhimen Hospital, Beijing University of Traditional Chinese Medicine, Beijing 100700, China
Abstract：Objective: To explore the risk factors for recurrence of IgA vasculitis in children, establish a recurrence risk prediction line graph model, and investigate the distribution of Chinese medicine syndromes. Methods: A retrospective case-control study was conducted using 200 patients with initial diagnosis of IgAV who were treated at the outpatient and inpatient departments of Beijing Children's Hospital, Capital Medical University from 2018 to 2023. Patients were divided into a recurrence group of 100 cases and a non-recurrence group of 100 cases (no recurrence within one year after initial cure) based on whether they exhibited characteristic manifestations of IgA vasculitis again within one month after symptom resolution. The study used R software (version 4.4.2) to perform univariate and multivariate Logistic regression analysis to screen for independent characteristics associated with IgA vasculitis recurrence, constructing a line graph to predict the recurrence of children. If a variable was observed to have a statistically significant association with IgA vasculitis recurrence in univariate Logistic regression analysis, it would be included in multivariate Logistic regression analysis to further screen for independent characteristics associated with IgA vasculitis recurrence. The rms package was used to construct the line graph and visualize the data. Univariate and multivariate Logistic regression analyses were performed to compare the distribution of Chinese medicine syndromes between the two groups, identifying Chinese medicine syndromes associated with IgA vasculitis recurrence. P<0.05 was considered statistically significant. Results: The duration of the initial illness, the location of rash involvement, NLR, C-reactive protein, D-dimer, and IgE were independently associated with the occurrence of recurrence (P<0.05), based on the results of multi-factor Logistic regression analysis, a risk prediction model for recurrent IgA vasculitis in children was constructed, with an area under the ROC curve (AUC) of 0.868. The decision curve indicates that using a line plot to predict recurrence in children with IgA vasculitis offers high clinical net benefit. The calibration curve results showed that the line plot demonstrates good calibration across different prediction probability ranges, with minimal deviation between predicted values and actual recurrence rates. Single-factor Logistic regression analysis revealed statistically significant differences in four disease factors lung, large intestine, heat, and external wind in both the recurrent and non-recurrent groups (P<0.05). Incorporating these four disease factors into a multi-factor Logistic regression analysis, the large intestine in disease factors and heat in disease factors were independently associated with recurrence (P<0.05). Conclusion: The number of days since initial onset, affected areas of rash, NLR, C-reactive protein, D-dimer, and IgE are independent influencing factors for recurrent IgA vasculitis in children. The line plot model constructed based on these influencing factors has high clinical net benefit and is reliable and accurate in clinical application. "Large intestine" and "heat" in traditional Chinese medicine disease factors are independent risk factors for the recurrence of IgA vasculitis.
张博妍, 赵骞, 何松蔚. 儿童IgA血管炎复发风险预测模型构建及中医证素与复发的相关性分析[J]. 河北医学, 2025, 31(7): 1127-1133.
ZHANG Boyan, et al. Construction of a Predictive Model for Recurrence Risk of Childhood IgA Vasculitis and the Correlation Analysis between Chinese Medicine Syndrome Elements and Recurrence. HeBei Med, 2025, 31(7): 1127-1133.
2025, Vol. 31 
冀公网安备13080202000786号