Early Predictors of Long-Term Outcomes in Pediatric "Mild" Traumatic Brain Injury: A Machine Learning Approach

Area of Science:

• Neuroscience
• Pediatrics
• Machine Learning

Background:

• Pediatric mild traumatic brain injury (pmTBI) affects many children, with up to one-third experiencing persistent symptoms after concussion (PSaC).
• Accurate prognosis of PSaC is challenging due to low incidence rates and nonspecific symptoms, even in uninjured children.
• Identifying reliable indicators for poor recovery in pediatric TBI is crucial for timely intervention.

Purpose of the Study:

• To utilize machine learning to identify predictors of poor recovery in a large prospective cohort of children with pmTBI.
• To assess the prognostic value of various assessments, including demographics, injury factors, symptom ratings, cognitive tasks, and neurosensory performance.
• To determine the most robust indicators of persistent symptoms after concussion at 4 months and 1-year postinjury.

Main Methods:

• A prospective cohort of 321 children with pmTBI underwent comprehensive assessments within 11 days of injury.
• Machine learning models were employed, using variable importance scores from 150 bootstraps to identify key prognostic indicators.
• Assessments included demographics, injury details, child/parent symptom reports, cognitive tests, objective performance measures, and symptom provocation on neurosensory tasks.

Main Results:

• Retrospective self-report of symptom burden and vulnerability to symptom provocation were the strongest predictors of PSaC at both 4 months and 1 year.
• These factors outperformed established risk scores in predicting persistent symptoms.
• Other significant predictors included near point convergence, long-term memory, household size, and parental symptom burden reports.

Conclusions:

• Retrospective symptom burden and acute symptom provocation are promising indicators for early risk stratification in pediatric TBI.
• These measures, combined with existing risk scores, can potentially guide individualized care for children with pmTBI.
• Further research is needed to integrate these findings into clinical practice and validate their effectiveness across different settings.