Multimodal Analysis of Secondary Cerebellar Alterations after Pediatric Traumatic Brain Injury

Abstract

While traditionally ignored as a region purely responsible for motor function, the cerebellum is increasingly being appreciated for its contributions to higher-order functions through cerebro-cerebellar networks. Traumatic brain injury (TBI) research generally focuses on the cerebrum, in part because of the frequency of acute pathology. Acute pathology is an important predictor of outcome, but neural disruption evolves over time in ways that influence daily-life functioning. We examine these changes in a multi-modal, multi-cohort study. Combining 12 datasets from the Enhancing NeuroImaging Genetics through Meta-Analysis (ENIGMA) Pediatric msTBI (moderate-severe TBI) working group, we measured volume of the total cerebellum and 17 subregions using a state-of-the-art, deep learning-based approach for automated parcellation in 598 children and adolescents with or without TBI (msTBI; n = 314 | non-TBI; n = 284; age M = 14.0 +/- 3.1 years). Further, we investigated brain-behavior relations between cerebellar volumes and a measure of executive functioning (Behavioral Rating Inventory of Executive Function). In a subsample with longitudinal data (n = 80), we assessed associations between late changes in cerebellar volume and early white matter (WM) microstructure using diffusion tensor imaging (DTI). Significantly smaller total cerebellar volume was observed in the msTBI group (Cohens d = -0.37). In addition, smaller volume was found in posterior lobe regions including crus II, lobule VIIB and VIIIB, and vermis VII and IX (Cohens d range = -0.22 to -0.43). These alterations were primarily driven by participants in the chronic phase of injury (> 6 months). Smaller cerebellum volumes were associated with more parent-reported executive function problems. In a subset of participants with longitudinal data, we found evidence of altered growth in total cerebellum volume, with younger msTBI participants showing secondary degeneration (volume reductions), and older participants showing disrupted development (slower growth rates). Changes in total cerebellum volume over time were also associated with WM microstructure in the first months post-injury, such that poorer WM organization in the first months post-injury was associated with decreases in volume longitudinally. Pediatric msTBI was characterized by smaller cerebellar volumes, primarily in the posterior lobe and vermis. The course of these alterations, along with group differences in longitudinal volume changes and injury-specific associations between DTI measures and volume changes, suggests secondary cerebellar atrophy, possibly related to supratentorial lesions, and/or disruption in cerebellar structural and functional circuits. Moreover, evidence for brain-behavior relationships underscore the potential cognitive and behavioral consequences of cerebellar disruption during a critical period of brain development.

ENIGMA’s Simple Seven: Recommendations to enhance the reproducibility of resting-state fMRI in traumatic brain injury

“…Moreover, a large field of view (FOV) (e.g., 250 mm x 250 mm) and many slices (50-70) are ideal, as they reduce the spatial aliasing and increase the SNR, although the additional collection time might be considered for patient tolerance. Several rsfMRI studies in TBI cropped the cerebellum; however, this subcortical region is very important in TBI impairment and recovery, therefore this practice is best avoided (Keleher et al, 2022;Caeyenberghs et al, 2009). It is also recommended to acquire fieldmaps or additional (two or more) reverse phase encoded spin echo images for distortion correction (Raimondo et al, 2021).…”