Mild traumatic brain injury (mTBI) is a recognized injury in full-contact sports. Clinical assessments of injury severity typically include a self-reported survey of the most common concussion symptoms. While convenient, functional impairment does not necessarily translate to physiological injury severity. Further, recent studies suggest full clinical recovery precedes physiological recovery meaning injury can persist even after symptoms have subsided.¹ This discrepancy between recoveries is hazardous for athletes because they are within the post-acute stage or window of cerebral vulnerability where a successive concussion will worsen the outcome.² Therefore, sensitive and specific biomarkers are needed which rely upon physiological rather than functional measures.

Despite extensive research being conducted in this field, there is no accepted physiologically-based diagnostic measure of the severity and extent of mTBI.³ Recently, some neuroimaging techniques have shown promise for objective biomarkers of mTBI. Among them, diffusion kurtosis imaging (DKI) could be well-suited for mTBI due to its sensitivity to tissue microstructure. Microstructural changes associated with mTBI, such as axonal swelling or beading⁴, are below the resolution of CT and high-resolution MRI. However, DKI can extract information from the white matter fibers reflective of these changes. Early studies applying DKI to mTBI have shown strong correlation between kurtosis metrics and recovery⁵. Still, the acute and post-acute stages of mTBI in sport-related concussion have not been investigated. These stages are critical in determining safe return-to-play.

Data were acquired from 26 high school and college football players who sustained concussions and 26 non-injured matched control athletes (Table 1). The study was approved by the IRB and written consents were obtained from participants. Subjects were administered the SCAT3 postconcussion symptom checklist⁶ preseason, within 24 hours of injury, and 8 days post-injury (Table 2). SCAT3 has a score range of 0-132; high scores indicate more severe symptom burden. Standardized Assessment of Concussion (SAC)⁷ and Balance Error Scoring System (BESS)⁸ were also administered for reference.

Single-shot SE-EPI sequence was used with 3mm-isotropic voxels, four b=0 (reference images) and 60 diffusion-weighted images with b=1000s/mm2 and 2000s/mm2 (30 diffusion directions for each shell). Images were processed through software developed in-house to estimate DKI tensors based on the algorithm by Jensen et al⁹ to calculate DTI metrics of mean, axial, and radial diffusivity (MD, Dax, Drad) and fractional anisotropy (FA) as well as DKI metrics of mean, axial, and radial kurtosis (MK, Kax, Krad). Via Tract-Based Spatial Statistics (TBSS)¹⁰, independent samples t-tests were performed to compare DKI metrics of concussed subjects and controls at 24 hours and 8 days post injury. Paired-sample t-tests investigated within-group changes between the time points for each of the metrics.

At baseline, there were no differences between the concussed group and control on the SCAT-3, SAC, or BESS. At 24 hours post-injury, SCAT-3 and SAC scores differed significantly between the groups. At 8 days post-injury, no significant differences existed (Table 2).

At 24 hours post-injury, the concussed group demonstrated decreased MD compared to controls in major white matter tracts (Figure 1). Dax and Drad were also decreased in the concussed group compared to controls. At Day 8, MD, Dax, and Drad showed more widespread effects. No differences existed for FA between the groups at either time point. Neither group demonstrated significant changes from 24 hours to 8 days for any of the DTI metrics.

For DKI metrics, the concussed group showed increased Kax compared to controls (Figure 1), with more widespread effects after 8 days. No group differences existed for MK or Krad. Within-group comparisons did not reveal any significant Kax changes for either group between time points.

Results revealed that DTI and DKI metrics are altered in concussed athletes in the acute post-injury stage. Not only did significant group differences exist, but effects were more widespread at 8 days post-injury indicating an evolving neuropathological process. Conversely, clinical symptoms approached baseline over the week. This supports the proposition that clinical recovery precedes physiological recovery.

Decreased diffusion in concussed athletes compared to controls is consistent with known pathophysiological processes in acute SRC. Decreased diffusion has been reported in the acute phase of SRC.^11,12 Increased Kax is consistent with the previously proposed mechanism of axonal beading which would increase the complexity of diffusion parallel to fibers.

By utilizing a well-controlled sample of contact sport athletes with acute SRC, this study may be useful in illuminating the immediate neurophysiological consequences of SRC. Future work will aim to assess the full trajectory of physiological recovery.