Alterations in Regional Distribution of Cerebral Amyloid Plaques following Head Trauma in Former College Athletes: Relationship to Long-Term Symptoms

Introduction: We previously described regionally specific changes in brain perfusion patterns in professional football players and their associations with neuropsychologic parameters, and others have reported involvement of cerebral tau protein in such players, while relationship of sports injuries to patterns of cerebral amyloid deposition remain poorly defined. In this study, we employed PET to examine the regional distribution of amyloid in the brains of former college athletes with histories of head trauma, and their relationships to symptomatology.

Methods: Fourteen former college athletes (mean age 29 years) who had each suffered at least one previous concussion and completed 16-20 years education, were administered computerized neurocognitive testing and a structured questionnaire of present symptoms. [F-18]Florbetaben PET images were acquired dynamically, with data binned in 20 time frames of 1 min each (early phase), followed by 4 frames of 5 min each (late phase), with image frames summed within each phase after linearly coregistering to subjects’ mean images, then elastically transformed to template space of a florbetaben scan used in standardized volume of interest quantification, and intensity normalized to mean voxel value within each phase. For each image set, statistical parametric mapping (spm) methods were used to assess significance of differences between the more symptomatic and less symptomatic subjects on a voxel-by-voxel basis by paired t-test analyses, to identify clusters of greatest peak significance and size. Significance of demographic group comparisons were tested by two-tailed t-test (parametric) and chi-square (non-parametric) analyses.

Results: Subjects divided between those who were more symptomatic (above-median overall Symptom Score, range 9-33, n=6) and less symptomatic (Score range 0-6, n=8), did not significantly differ with respect to mean age (ranges 28-31, 25-34), sex, years of education, or number of concussions. The spm analysis of early phase images, representing perfusion plus initial radiotracer distribution, found no significant differences between the more and less symptomatic subject groups. Late phase images, representing conventionally determined presence and frequency of amyloid plaques, revealed regions of significantly greater radiotracer accumulation in more symptomatic subjects, with the most significant and largest cluster located in right pregenual anterior cingulate cortex (t=7.20, p<0.0005). Conversely, no cluster of comparable significance was identified as having greater radiotracer accumulation in the less symptomatic subjects.

Conclusions: Symptomatic head trauma in young former college athletes was associated with regionally specific accumulation of cortical amyloid plaques, most significant in pregenual cingulate cortex, a brain region known to have high density of opioid receptors.