Fibre-specific white matter in chronic traumatic brain injury patients : Towards single-subject profiles

Aims and Background: Moderate-to-severe traumatic brain injury (m-sTBI) leads to poor functional outcomes due to chronic deficits in cognitive and motor functions. These long-term functional outcomes are often difficult to treat and predict. The overarching aim of this thesis was to develop a science-led and principled paradigm to help better understand and potentially improve functional outcomes in chronic m-sTBI patients. m-sTBI patients are highly heterogeneous due to the nature and location of injuries, which is an important predictor of functional outcomes. Structural magnetic resonance imaging (MRI) provides quantitative measurements of macroscopic (i.e., anatomical MRI) and microscopic (i.e., diffusion MRI; dMRI) characteristics of these injuries. Imaging-based characterisation when combined with training can help identify behaviourally relevant biomarkers (i.e., neurological metrics that are associated with aspects of behaviour), which are notable indicators of one’s response to training or therapy. Therefore, the current thesis attempted to address the overarching aim by incorporating relevant structural neuroimaging and behaviourally relevant biomarkers in response to training.

Method and Results: This overarching aim of this thesis was addressed across three studies. Study 1 was designed to summarise the recent findings on training-induced structural neuroplasticity research in acquired brain injury (ABI) populations following cognitive and/or motor training (i.e., targeting the most prevalent chronic symptoms) due to scarce research in m-sTBI. The critical review revealed that most studies have used (1) dMRI as the primary structural MRI modality, which may be more sensitive to training-related changes; (2) non-biologically specific dMRI tensor-based metrics; and (3) non-intensive single-modality training protocols (i.e., cognitive/motor training alone, not combined). It was also argued that developing robust longitudinal single-subject profiling designs (i.e., 1 patient vs X controls), shifting away from group-comparisons, may be necessary in m-sTBI patients to control for excessive heterogeneity in their injuries. Given these findings, the overarching aim of the thesis was addressed in the empirical studies through a two-part paradigm. This was to first develop behaviourally relevant biomarkers in a healthy cohort (i.e., as those developed in m-sTBI patients are limited by excessive heterogeneity); then develop and present a preliminary, longitudinal single-subject profiling framework utilising fibre-specific white matter which may potentially be interpreted with the developed behaviourally relevant biomarkers. This was conducted in order to understand our knowledge of neurocognitive function in m-sTBI and how functional outcomes change due to the effects of different training regimes, at the individual-subject level.

Two empirical studies were conducted to present (1) how to develop a behaviourally relevant biomarker using fibre-specific dMRI metrics in a healthy cohort (Study 2); and (2) a proof-of-concept longitudinal single-subject profiling approach for individual m-sTBI subjects (Study 3). To demonstrate the paradigm, Study 2 aimed to develop a robust behaviourally relevant biomarker for attentional lapses in a healthy cohort; a common yet under-studied symptom of m-sTBI which may precipitate other cognitive deficits and certain motor deficits. The novel findings were that decreased white matter fibre-density of the superior longitudinal fasciculus-I is associated with greater susceptibility to attentional lapses in healthy controls, and may be a behaviourally relevant biomarker that can be targeted in m-sTBI patients. The aim of Study 3 was to develop and present a clinically specific longitudinal single-subject profiling framework for chronic m-sTBI patients which incorporates (1) subject and tract specific characterisations of white matter microstructure; and (2) attention (e.g., attentional lapses) and motor (e.g., fine motor skills) behaviours following a combined cognitive and motor training. These novel longitudinal single-subject profiles may be compared to putative behaviourally relevant biomarkers for further interpretations, such as those discovered in Study 2. These novel profiles await further validation and extensions, but ultimately may assist with diagnostic and/or treatment decisions made to individual m-sTBI patients.

Conclusion: The current thesis presents a principled, case-based neuroimaging paradigm that may help better understand and potentially improve functional outcomes in chronic m-sTBI patients. This novel paradigm provides an important stepping stone for future research to further expand upon single-subject profiling and by developing comprehensive behaviourally relevant biomarkers targeting m-sTBI deficits for in-depth interpretations. With further validations and extensions of this approach, the case-based paradigm developed in this thesis may assist with conventional care options and continued work may lead to neuroimaging-guided training to help better understand, predict and assist recovery with outcomes in chronic m-sTBI patients.