Acute interleukin-10 predicts persistent post-concussion symptoms in children with mild traumatic brain injury:...

Children are a vulnerable population for mild traumatic brain injury (mTBI), accounting for a high burden of emergency department (ED) visits. A significant and unpredictable subset will develop persistent post-concussion symptoms (P-PCS), with consequences for daily quality of life. At ED discharge, no reliable tool exists to identify children requiring neuropsychological follow-up. Acute blood biomarkers represent a promising approach for early risk stratification.

In this prospective multicenter cohort study, children with mTBI had a blood sample collected within 24 h of injury for biomarker analysis (S100b, NfL, NTproBNP, GFAP, HFABP, IL6, IL10). The parent version of the Post Concussion Symptom Inventory (PCSI-P) was sent at 14 days and 3 months post-injury. P-PCS were defined as the presence of three or more symptoms worsening by at least one point above pre-injury baseline at both timepoints. This analysis included only children identified to be symptomatic at 14 days. Within this group, biomarker levels were compared between those with P-PCS at 3 months and those who had recovered.

Of 203 eligible children, 120 (60%) completed the 14-day questionnaire and 84 (41%) completed the 3-month questionnaire. Among the 38 children who remained symptomatic at 14 days, 16 (42%) met criteria for P-PCS at 3 months, 9 (24%) recovered, and 13 (34%) were lost to follow-up. IL10 levels were significantly elevated in children with P-PCS (p = 0.014), with an AUC of 79.9% (95% CI: 59.5-100%). At the optimal Youden index threshold, IL10 achieved 100% sensitivity and 55.6% specificity for ruling-out P-PCS. At 100% specificity, IL10 achieved a sensitivity of 31% for ruling-in P-PCS.

This exploratory study suggests that early IL10 measured at ED presentation may support two complementary risk-stratification strategies: a rule-out approach (100% negative predictive value [NPV]) to safely exclude children unlikely to develop P-PCS, and a rule-in approach (100% positive predictive value [PPV]) to identify those at high risk of requiring neuropsychological follow-up. Consistent with adult TBI literature, these findings highlight the acute neuroinflammatory response in pediatric recovery. Replication in larger cohorts is needed before clinical application.