Effects of docosahexaenoic acid and eicosapentaoic acid supplementation on white matter integrity after repetitive sub-concussive head impacts during American football: Exploratory neuroimaging findings from a pilot RCT

CONTEXT: Repetitive sub-concussive head impacts (RSHIs) are common in American football and result in changes to the microstructural integrity of white matter. Both docosahexaenoic acid (DHA) and eicosapentaoic acid (EPA) supplementation exerted neuroprotective effects against RSHIs in animal models and in a prior study in football players supplemented with DHA alone. OBJECTIVE: Here, we present exploratory neuroimaging outcomes from a randomized controlled trial of DHA + EPA supplementation in American football players. We hypothesized that supplementation would result in less white matter integrity loss on diffusion weighted imaging over the season. DESIGN, SETTING, PARTICIPANTS: We conducted a double-blind placebo-controlled trial in 38 American football players between June 2019 and January 2020. INTERVENTION: Participants were randomized to the treatment (2.442 g/day DHA and 1.020 g/day EPA) or placebo group for five times-per-week supplementation for 7 months. Of these, 27 participants were included in the neuroimaging data analysis (n = 16 placebo; n = 11 DHA + EPA). EXPLORATORY OUTCOME MEASURES: Changes in white matter integrity were quantified using both voxelwise diffusion kurtosis scalars and deterministic tractography at baseline and end of season. Additional neuroimaging outcomes included changes in regional gray matter volume as well as intra-regional, edge-wise, and network level functional connectivity. Serum neurofilament light (NfL) provided a peripheral biomarker of axonal damage. RESULTS: No voxel-wise between-group differences were identified on diffusion tensor metrics. Deterministic tractography using quantitative anisotropy (QA) revealed increased structural connectivity in ascending corticostriatal fibers and decreased connectivity in long association and commissural fibers in the DHA+EPA group compared to the placebo group. Serum NfL increases were correlated with increased mean (ρ = 0.47), axial (ρ = 0.44), and radial (ρ = 0.51) diffusivity and decreased QA (ρ = −0.52) in the corpus callosum and bilateral corona radiata irrespective of treatment group. DHA + EPA supplementation did preserve default mode/frontoparietal control network connectivity (g = 0.96, p = 0.024). CONCLUSIONS: These exploratory findings did not provide strong evidence that DHA + EPA prevented or protected against axonal damage as quantified via neuroimaging. Neuroprotective effects on functional connectivity were observed despite white matter damage. Further studies with larger samples are needed to fully establish the relationship between omega-3 supplementation, RSHIs, and neuroimaging biomarkers. TRIAL REGISTRATION: ClinicalTrials.gov-NCT04796207