Concussion decreases brain signal variability

Sports-related concussion represents a major public health concern, with millions of cases diagnosed annually. Despite significant methodological advances in functional brain imaging, a clinically meaningful neuroimaging biomarker of concussion has yet to be established. This may be due in part to a reliance on traditional measures. For example, approaches examining mean activation across trials reduce responses to a single measure of central tendency and therefore fail to capture important individual differences. In contrast, blood oxygenation level dependent (BOLD) variability offers a promising new approach to measuring brain function, Specifically, it captures individual differences in the temporal variability in the BOLD signal over time. Further, it is sensitive to neural, vascular, and neurovascular coupling changes that accompany concussion. We collected baseline (prior to injury) and post-concussion (within 48 hours of injury) neuroimaging data in 56 collegiate athletes. Within this longitudinal dataset, we investigated changes in BOLD variability across multiple spatial scales. In addition, we examined their relationships with self-reported symptoms and cognitive performance. As expected, symptoms increased following concussion while computerized cognitive performance declined. BOLD variability decreased from baseline to post-concussion, particularly in the somatomotor and language networks. Moreover, changes in BOLD variability were related to changes in symptom reporting and cognitive performance. Together, these findings advance our understanding of how concussion affects brain function and highlight BOLD signal variability as a promising biomarker to aid ongoing efforts to improve the diagnosis and treatment of sports-related concussion.