Background The link between head-impact exposure in football and neurological sequelae remains controversial. Blood-based biomarkers can provide valuable information in traumatic brain injuries, reflecting e.g. axonal damage.

Objective To investigate if repetitive headers or accidental head impacts in football could cause structural damage to the brain, detected as an increase in serum concentrations of neurofilament light (NfL) or tau proteins.

Design Prospective cohort study.

Setting Elite division football.

Participants Male football players in the Norwegian premier league, including a total of 621 player seasons.

Independent Variables Short- and long-term exposures in football, with and without head impacts.

Main Outcome Measurements Baseline NfL and tau were measured in Norwegian professional football players in pre-season. Then, the effect of short-term exposures (independent variable) was assessed by measuring biomarker levels (dependent variable) after three conditions: (1) high-intensity exercise, (2) repetitive headers, and (3) head-impact incidents in a match. The effect of long-term head impact exposure was assessed by comparing two groups with relative differences in previous concussions and headers at baseline (low vs. high levels).

Results We analyzed 354 samples. Mean (±SD) NfL concentration was 6.8±2.6 pg/mL; mean tau concentration was 1.2±0.7 pg/mL. We observed no short-term effects on NfL after exposure from either of the three conditions. Tau levels rose in response to high-intensity exercise and repetitive headers, but not after accidental head-impact incidents; the highest absolute values were seen 1 h after high-intensity exercise. We did not detect any long-term effects on serum NfL or tau concentrations from previous concussions and headers.

Conclusions NfL and tau in serum were unaffected by head impacts in football, after both short-term and long-term exposure. Importantly, tau levels seem to rise in response to exercise, emphasizing the need for appropriate control groups in future studies. Our findings highlight important characteristics and limitations for using NfL and tau as biomarkers in sports.