Abstract Details

Title Comparative Performance of Glial and Neuronal Biomarkers in Predicting Outcomes Following Moderate to Severe Traumatic Brain Injury: A Network Meta-analysis

Topic Neuro Trauma and Critical Care

Presentation(s) S17 - Neurocritical Care (1:12 PM-1:24 PM)

Poster/Presentation
Number 002

Background TBI is considered one of the leading causes of morbidity and mortality worldwide. Biomarkers like S100B, Glial Fibrillary Acidic Protein (GFAP), Neuron-Specific Enolase (NSE), and Ubiquitin Carboxy-terminal Hydrolase L1 (UCH-L1) have shown promising prospects in outcome prediction, though their performance varies across studies.

Objective This network meta-analysis (NMA) evaluated the comparative performance of glial and neuronal biomarkers for mortality and unfavorable functional outcomes in adults with moderate to severe traumatic brain injury (TBI).

Design/Methods An NMA following the PRISMA guidelines was conducted on studies assessing multiple biomarkers to predict unfavorable outcomes following TBI, specifically mortality and poor functional outcomes (Glasgow Outcome Scale [GOS], GOS-Extended [GOSE]). A Bayesian framework with three Markov Chain Monte Carlo (MCMC) chains, a random-effects model, and Log Odds Ratios (LOR) transformed into Odds Ratios (OR) was used for comparisons, with GFAP as the reference. Heterogeneity was assessed via the I² statistic. Rank probabilities determined each biomarker's relative performance.

Results Seven studies comprising 522 patients were included. For mortality, GFAP vs. NSE (OR 1.003, 95% CI: 0.53-1.84), GFAP vs. S100B (OR 0.88, 95% CI: 0.50-1.55), and GFAP vs. UCH-L1 (OR 0.80, 95% CI: 0.45-1.46) showed minimal heterogeneity (I²=4%). NSE had the highest probability of being the most effective biomarker (39.96%), followed by GFAP (35.9%). For unfavorable functional outcomes, GFAP vs. NSE (OR 0.82, 95% CI: 0.28-2.14), GFAP vs. S100B (OR 1.1, 95% CI: 0.42-2.7), and GFAP vs. UCH-L1 (OR 1.1, 95% CI: 0.23-4.6) had low heterogeneity (I²=14%). UCH-L1 had the highest probability of being the most effective (40.2%), followed by S100B (36.2%).

Conclusions NSE was most effective for predicting mortality, while UCH-L1 ranked highest for predicting unfavorable functional outcomes. All biomarkers showed consistent findings with low heterogeneity across studies. Future research should focus on the clinical application of these biomarkers in decision-making settings.

Authors/Disclosures
Lilian Maria Godeiro Coelho, MD PRESENTER	Dr. Godeiro Coelho has nothing to disclose.
Fernanda Jacinto Pereira Teixeira, MD	Dr. Jacinto Pereira Teixeira has nothing to disclose.
Ayham M. Alkhachroum, MD (Columbia University Medical Center)	The institution of Dr. Alkhachroum has received research support from Miami CTSI.

��ɫ��

��ɫ��