Multi-modal assessment of outcomes in pediatric mild traumatic brain injury

Abstract

Objective: Mild traumatic brain injury is the most common type of head injury among children and adolescents. Symptoms are highly heterogeneous and are affected by several factors pre- and post-injury, making diagnosis and symptom management challenging. This study aims to look at using multi-modal analysis to determine relationships between symptoms and contribute to the understanding of how acute symptoms transition to a chronic pathology. Study design: Participants aged 8-18 years and with a medically diagnosed mTBI were recruited through the Alberta Children’s Hospital Brain Injury Clinic. Age- and sex-matched healthy controls were recruited through word of mouth, sibling matches, and the Healthy Infants and Children's Clinical Research Program (HICCUP). Participants were categorized as follows: 1) symptomatic mTBI (n=26), 2) repetitive mTBI (n=14), and controls (n=27). Study appointments occurred within 6-16 weeks from the initial injury where the following assessments were administered: symptom assessment, balance assessment, neuropsychological evaluation, ERP assessment, neuroimaging, and blood collection for cytokine analysis. Each assessment was analyzed individually, and then used to build a multi-modal elastic-net regression model to identify significant predictors among all of the outcomes. Results: mTBI participants had higher PCSI scores and lower PedsQL scores compared to controls, indicating higher symptom burden. The mTBI groups had more difficulties with school, as reflected by the BASC assessment. No differences were found between groups for the balance assessment or the ERP assessment. On the CNSVS test, female mTBI participants had lower Neurocognition Index scores and faster reaction times compared to female controls. Repetitive mTBI males had faster psychomotor speed compared to symptomatic mTBI males. CTACK levels were elevated in female mTBI groups, SCGF-B levels were lower in male mTBI groups, and MDC levels were lower in female mTBI groups when compared to controls. No differences in FA and MD scores were found between groups and sexes scores for the left and right CPC tracts. Multi-modal assessment revealed important model predictors from the neuropsychology and cytokine assay modalities. Conclusion: Multi-modal assessment is a necessary tool to understand the key factors involved in the secondary injury and symptoms that arise after a pediatric mTBI. It is more informative than looking at individual assessments and highlights the limitations of looking at specific biomarkers in isolation when studying a complex injury affecting multiple systems in the body.