Evaluation of GFAP, UCH-L1 and S100B Blood Biomarkers for the Management of Sports-Related Concussion in Professional Rugby Players: A Prospective Observational Study
Sports medicine - open · 2026
This blood panel is great at confirming concussion but misses about half of true cases, so it can't replace or override the HIA yet.
The paper
Prospective observational cohort, n=68 professional rugby players (23 sports-related concussion episodes), single French elite club, 2022-2024.
What they found
At 36h post-match, a rise in S100B of at least 15% from baseline gave 100% specificity but only 39% sensitivity for SRC; GFAP at least 30 ng/L gave 100% specificity but 17% sensitivity; combining both kept 100% specificity while raising sensitivity to 52%. UCH-L1 stayed below the quantifiable range at 36h post-SRC.
The appraisal
This is a well-conceived real-world cohort with serial sampling timed to the actual HIA protocol, but the numbers that matter are the sensitivities, not the specificities. High specificity is unsurprising here because baseline GFAP and UCH-L1 sat below assay detection limits in almost everyone, so any signal above the floor will look specific by construction. The real problem is sensitivity: even combined, the panel misses roughly half of confirmed concussions, which is not good enough to act on alone. Only 23 SRC episodes in one club is a small, likely underpowered sample for deriving cut-offs that will generalise, and the 15% and 30 ng/L thresholds appear to have been chosen and tested in the same dataset, so the 100% specificity is probably optimistic until it is checked in an independent cohort with a clearly reported control denominator.
The gap
Sensitivity of 52% even with a two-marker panel means this cannot be used to rule out concussion, so the paper doesn't answer the question clinicians actually need answered: can a normal blood result safely let a player return sooner or override an equivocal HIA?
Landmark context
This builds on the general mTBI literature establishing GFAP and UCH-L1 as an FDA-cleared panel for ruling out intracranial injury after head trauma (the Banyan/ALERT-TBI work), and on decades of S100B use in Scandinavian mild-TBI CT-triage guidelines; this study is novel mainly in showing how exercise and contact load (not just head injury) drive S100B and UCH-L1 up within hours in contact-sport athletes, a confounder that non-sport mTBI biomarker studies don't have to contend with.
What to do Monday
No change to current HIA practice yet. The useful, actionable finding is that 36h post-match is a 'clean' sampling window free of exercise-induced biomarker elevation, which is worth knowing if a team is already collecting research bloods, but the low sensitivity means a normal biomarker result cannot be used to clear a player or override clinical judgement.
In practice
This doesn't change what happens pitch-side on Monday: HIA1 removal-from-play decisions still rest on the standard clinical battery, and a normal S100B/GFAP at 36h must never be used to fast-track a return or override a positive HIA, sensitivity is too low for that. Where it's genuinely useful is for club medical staff already running research bloods, draw at 36h rather than 2h post-match, since 2h samples pick up exercise-induced S100B and UCH-L1 rises from heavy contact work that have nothing to do with concussion. For S&C, the finding that contact frequency correlates with S100B/UCH-L1 rise at 2h is the interesting bit, it confirms that heavy-contact training blocks produce a measurable physiological hit even with no diagnosed concussion, worth factoring into recovery scheduling around contact-heavy weeks rather than treating it as a diagnostic tool.
The five papers that matter — every week, free.
We read the whole sports-medicine, rehab & performance literature and appraise the papers worth your time.
Get the free weekly issueRelated appraisals
Read in your language: English · Français · Deutsch · Español · Italiano · Nederlands · Dansk · Svenska