# A diagnostic prediction model was established based on the clinical characteristics of multicenter children with Kawasaki disease in Xinjiang

## TL;DR

The study developed a model to predict treatment response in children with Kawasaki disease using blood lymphocyte subsets in Xinjiang.

## Contribution

A novel diagnostic prediction model using lymphocyte subsets for IVIG treatment sensitivity in Kawasaki disease is proposed.

## Key findings

- CD4+ cell count is an independent risk factor for IVIG non-responsiveness in Kawasaki disease.
- Lymphocyte subsets like CD8+ and CD+ show high sensitivity and specificity for distinguishing Kawasaki disease from febrile illnesses.

## Abstract

Kawasaki disease (KD) is a serious pediatric systemic vasculitis that may cause cardiovascular complications, including coronary artery lesions (CAL). This study aimed to evaluate the predictive value of peripheral blood lymphocyte subsets for intravenous immunoglobulin (IVIG) treatment sensitivity in children with KD in Xinjiang.

This multicenter retrospective study collected clinical and laboratory data from 142 children with KD and 120 controls with infectious fever across several hospitals in Xinjiang from June 2022 to December 2024. Peripheral blood lymphocyte subsets and immunoglobulins were analyzed. Multivariate logistic regression was used to identify risk factors for IVIG non-responsiveness, and ROC curve analyses were performed to evaluate diagnostic efficacy.

Significant differences in lymphocyte subsets were observed between KD and control groups, including CD3+, CD4+, CD8+, CD19+, and the CD4+/CD8+ ratio. Multivariate logistic regression identified CD4+ cell count as an independent risk factor for IVIG non-responsiveness. ROC analyses suggested that lymphocyte subsets and immunoglobulins (including CD3+, CD+, CD16+CD56+, IgA, and IgM) have diagnostic potential, with CD8+ and CD+ showing high sensitivity and specificity.

Peripheral blood lymphocyte subsets, particularly CD4+, may serve as useful biomarkers for predicting IVIG treatment response and distinguishing KD from other febrile illnesses. Further studies with larger sample sizes are warranted to refine predictive models and improve KD management.