Subconcussive head impacts in sport: A systematic review of the evidence

Purpose: To identify and evaluate the evidence that examines subconcussive impacts in sport-specific settings, and address two objectives: a) to determine how 'subconcussion' is characterized in the current literature, and b) to identify directions for future research. Research design: Systematic review. Methods and procedures: CINAHL, EMBASE, MedLine, PsycINFO, SportDiscus, and Web of Science were searched for articles that sought to assess subconcussive impacts or outcomes related to non-concussive head impact exposure. Eligible articles were reviewed and evaluated with three quality assessment tools by rotating pairs of reviewers. Results: A total of 1966 articles were screened. Fifty-six studies met the inclusion criteria. Studies were classified into three main categories based on primary focus: neurobiological, neuropsychological, and impact exposure metrics. The neurobiological studies suggested that in male athletes, functional and microstructural deterioration was associated with repetitive head impacts. There was insufficient to weak evidence for the relationship between repetitive hits to the head and deterioration in neurocognitive performance. Studies of impact exposure metrics examined various indices, including linear acceleration, rotational acceleration, and location and frequency of hits. Insufficient evidence was presented to determine a minimal injury threshold for repetitive hits to the head. Across all categories of studies there was a lack of consistency and clarity in defining and measuring variables related to the concept of 'subconcussion'. Conclusions: Evidence reviewed predominantly from studies of male athletes in contact and collision sports identifies that repetitive hits to the head are associated with microstructural and functional changes in the brain. Whether these changes represent injury is unclear. We determined the term 'subconcussion' to be inconsistently used, poorly defined, and misleading. Future research is needed to characterize the phenomenon in question.