A predictive nomogram for relapse risk in children with autoimmune encephalitis

Area of Science:

• Pediatric Neurology
• Immunology
• Clinical Prediction Modeling

Background:

• Autoimmune encephalitis (AE) is a significant cause of acute encephalopathy in children, with relapses leading to adverse neurodevelopmental outcomes.
• Predicting relapse risk in pediatric AE is crucial for timely intervention, but validated tools for the Chinese population are lacking.

Purpose of the Study:

• To develop and validate a nomogram for predicting the risk of relapse in children diagnosed with autoimmune encephalitis in China.

Main Methods:

• A retrospective cohort study of 127 children (aged 1-16 years) diagnosed with AE between 2017 and 2023.
• Logistic regression identified independent predictors of relapse.
• A nomogram was constructed and validated for discrimination, calibration, and stability.

Main Results:

• The overall relapse rate was 26.8%.
• Independent risk factors for relapse included older age (≥8 years), limb weakness, delayed immunotherapy, and positive N-methyl-D-aspartate receptor (NMDAR) antibodies.
• Rituximab and prolonged IVIG treatment (≥6 months) were protective factors. The nomogram demonstrated excellent predictive accuracy (C-statistic 0.936).

Conclusions:

• The developed nomogram is accurate and clinically useful for predicting individualized relapse risk in Chinese children with AE.
• This tool can aid clinicians in stratifying risk and tailoring management strategies for pediatric AE patients.