Predictors for identifying autoimmune encephalitis in pediatric patients

Area of Science:

• Neurology
• Pediatrics
• Immunology

Background:

• Autoimmune encephalitis (AE) is a serious neurological condition in children.
• Accurate and early diagnosis of pediatric AE is crucial for effective treatment.
• Predictive models can aid in identifying children at risk for AE.

Purpose of the Study:

• To identify independent predictors of autoimmune encephalitis (AE) in pediatric patients.
• To develop and validate a predictive model for diagnosing pediatric AE.
• To improve the diagnostic accuracy of AE in children.

Main Methods:

• Retrospective study of 88 pediatric patients (37 AE, 51 non-AE).
• Lasso regression, univariate, and multivariate logistic analyses were used to identify risk factors.
• A predictive model was developed and validated using ROC curves, calibration plots, and decision curve analysis.

Main Results:

• Four independent predictors for pediatric AE were identified: age, cerebrospinal fluid protein, cerebrospinal fluid chloride, and spontaneous remission.
• The predictive model demonstrated excellent diagnostic performance with an AUC of 0.976.
• The model showed strong discrimination and calibration, indicating high reliability.

Conclusions:

• A high-performance predictive model for pediatric autoimmune encephalitis was successfully established and validated.
• The model incorporates four clinically accessible parameters, facilitating practical application in diagnosis.
• This tool can significantly aid clinicians in the early and accurate diagnosis of AE in children.