MRI-based cortical morphometry in traumatic brain injury: a pilot longitudinal study

PURPOSE: Traumatic brain injury (TBI) can result in various changes to cortical morphometry which may negatively affect quality of life for individuals with TBI. Previous research has identified both increases and decreases in cortical volume, surface area and thickness across brain regions, suggesting that the impact of TBI on cortical morphometry remains unclear. This project used the "recon-all-clinical" pipeline on 4 mm Axial MRI scans to investigate cortical volume, surface area and thickness following TBI, which is important for improving understanding of post-injury neuroanatomical changes and refining neuroimaging methodologies. METHODS: The measured regions were grouped using area-weighted means into cortical lobes consisting of frontal, parietal, occipital, temporal, cingulate gyrus and insula for analysis. The changes in morphometry were analysed longitudinally in group with TBI and compared with age/sex matched control scans. RESULTS: The findings showed significant cortical thinning over time in the group with TBI in the left occipital cortex and right cingulate gyrus, and no significant changes in cortical surface area or cortical volume across any of the measured regions following corrections for multiple comparisons. When comparing individuals with TBI and neurotypical control, this revealed increases in cortical thickness in the left insula in the initial TBI scan compared to neurotypical control. CONCLUSIONS: The increased thickness at initial TBI scan should be interpreted cautiously, as they may reflect segmentation uncertainty in the presence of lesion heterogeneity, altered tissue contrast, and imaging artefacts. Additionally, potential early post-injury neuroplastic or compensatory changes may contribute to these findings, rather than indicating true morphometric differences. After applying corrections, significant longitudinal reductions in the left occipital cortex and right cingulate gyrus were found, suggesting that these regions may be vulnerable following TBI. These results suggest that the complexity of cortical morphometry changes immediately following TBI may reduce the validity of these measurements. Future studies should investigate this further with a larger sample, to further determine the usefulness of "recon-all-clinical" for short-term morphometric analysis post-TBI.