NEUROLOGY | NOVEMBER 07, 2018

Asken BM, Bauer RM, DeKosky ST, Svingos AM, Hromas G, Boone JK, DuBose DN, Hayes RL and Clugston JR.

Neurology. 2018 Nov 7. pii: 10.1212/WNL.0000000000006617

DOI: 10.1212/WNL.0000000000006617

ABSTRACT

Objective To evaluate changes in serum biomarker concentrations (β-amyloid peptide 42 [Aβ42], total tau, ubiquitin carboxy-terminal hydrolyzing enzyme L1, S100 calcium binding protein B [S100B], glial fibrillary acidic protein [GFAP], microtubule associated protein 2 [MAP2], and 2′,3′-cyclic-nucleotide 3′-phosphodiesterase [CNPase]) after sport-related concussion (SRC) in a sample of collegiate athletes. Associations with clinical outcomes were also investigated.

Methods Participants in this case-control study included 36 athletes (50% male, 61% white, aged 19.7 ± 1.0 years) with SRC. Twenty-nine also had baseline blood drawn, allowing for within-patient analyses of concentration changes. Between-group analyses incorporated 86 demographically matched controls (51% male, 63% white, aged 19.6 ± 1.1 years). Biomarker sensitivity/specificity for SRC vs controls and relative to standardized normative cutoffs was evaluated (receiver operating characteristics). We also analyzed associations between post-SRC clinical outcomes and both biomarker change from baseline and post-SRC concentrations.

Results The majority of blood samples had concentrations of GFAP, MAP2, and CNPase below limits of quantification. Within-patient analyses indicated elevated S100B after SRC (p = 0.003, 67% of patients elevated), especially for blood samples collected <4 hours post-SRC (88% of patients). Significant between-group differences were limited to blood draws <4 hours post-SRC: Aβ42 (81% of SRC > control median, area under the curve [AUC] = 0.75 [95% confidence interval 0.59–0.91]), total tau (75% SRC > control, AUC = 0.74 [0.56–0.79]), and S100B (88% SRC > control; AUC [specific to white race] = 0.82 [0.72–0.93]). Using standardized normative cutoffs (z > 1.0), specificity ranged from 79.1% to 89.3% while sensitivity was <70%. Biomarkers were not associated with clinical outcomes.

Conclusion For SRC, diagnostic accuracy of serum biomarkers appears best if blood is collected within a few hours. Accurate blood marker identification of SRC appears somewhat dependent on the “healthy” comparison. Additional research must evaluate whether physiologic changes in the absence of clinical changes, or vice versa, are relevant for concurrent or future neurologic health.

Classification of evidence This study provides Class III evidence that certain serum biomarkers are elevated from baseline and higher than demographically matched controls after sport-related concussion.