Abstract
OBJECTIVES: Mild traumatic brain injury (mTBI) is a heterogeneous injury, and it is difficult to predict those that will go on to experience sustained post-concussion syndrome (PCS, >3mths). As such, data from a wide variety of sources would be useful in investigating long term outcome, preferably within the same population. This study explored the interplay between functional, structural and metabolic changes after mTBI and their relationship with persistent PCS and cognitive performance. METHODS: fMRI, DTI and MRS data were acquired from participants with chronic (>1 year) mTBI and persistent PCS (n=8), participants with mTBI but no on-going PCS (n=8) and non-head injured controls (n=9). Functional data was acquired whilst participants undertook an n-Back and Paced Serial Visual Addition Task (PVSAT). Conventional analysis was undertaken to investigate areas of difference in BOLD response and fractional anisotropy (FA) between groups. These regions of interest were then used to extract individual BOLD contrast values or FA for each participant. MRS acquired from right DLPFC was analysed. Metabolites which differed between groups were used in further analysis. The relationship between functional, structural and metabolic indices was investigated using partial correlation, controlling the effect of age. RESULTS: Participants with mTBI and PCS displayed less of an increase in BOLD response in prefrontal (left inferior/middle frontal gyrus; PVSAT) and temporal (right medial/inferior temporal lobe; n-Back) areas when performing the most difficult task compared to controls, despite similar task performance in both groups. Greater post-concussion symptom report correlated with reduced temporal (right medial/inferior temporal; n-Back) and posterior cingulate/precuneus (PVSAT) BOLD response, as well as increased anterior cingulate (n-Back) BOLD response. Correlations revealed that reduced BOLD response in the left inferior/middle frontal gyrus also was associated with reduced FA in posterior corpus callosum (r=0.4, p<0.05) and reduced creatine concentration in rDLPFC (r=0.5 p<0.05) across all participants. This correlation approached significance when only mTBI participants were included in the analysis (FA: p=0.08; Creatine: p=0.06). CONCLUSIONS: Participants who sustained an mTBI a year previously and have persistent PCS did not exhibit increased activity in working memory related areas (prefrontal/temporal) with task difficulty to the same extent as controls, but did show an increase in an area related to attention and error monitoring (anterior cingulate). This suggests that increased attention in this group is compensating for reduced working memory capacity in the task to achieve the same level of performance. Furthermore, the functional differences in prefrontal cortex correlate with both structural changes indicative of impaired white matter tract integrity and metabolic changes indicative of an ongoing energy crisis. The symptoms experienced by these participants seem to be caused by an interaction of these modalities, and it is likely that each brain injury will result in a different pattern of change.