Fatigue following mild traumatic brain injury relates to visual processing and effort perception in the context of motor performance

Abstract

Introduction: Following mild traumatic brain injury (mTBI), a substantial number of patients experience disabling fatigue for months after the initial injury. To date, the underlying mechanisms of fatigue remain unclear. Recently, it was shown that mTBI patients with persistent fatigue do not demonstrate increased performance fatigability (i.e., objective performance decline) during a sustained motor task. However, it is not known whether the neural activation required to sustain this performance is altered after mTBI.

Methods: Blood oxygen level-dependent (BOLD) fMRI data were acquired from 19 mTBI patients (>3 months post-injury) and 19 control participants during two motor tasks. Force was recorded from the index finger abductors of both hands during submaximal contractions and a 2-minute maximal voluntary contraction (MVC) with the right hand. Voluntary muscle activation (i.e., CNS drive) was indexed during the sustained MVC using peripheral nerve stimulation. Fatigue was quantified using the Fatigue Severity Scale (FSS) and Modified Fatigue Impact Scale (MFIS). Questionnaire, task, and BOLD data were compared across groups, and linear regression was used to evaluate the relationship between BOLD-activity and fatigue in the mTBI group.

Results: The mTBI patients reported significantly higher levels of fatigue (FSS: 5.3 vs. 2.6, p < 0.001). Both mTBI- and control groups demonstrated significant performance fatigability during the sustained MVC, but no significant differences in task performance or BOLD-activity were observed between groups. However, mTBI patients reporting higher FSS scores showed increased BOLD-activity in the bilateral visual cortices (mainly extrastriate) and the left midcingulate gyrus. Furthermore, across all participants mean voluntary muscle activation during the sustained MVC correlated with long lasting post-contraction BOLD-activation in the right insula and midcingulate cortex.

Conclusion: The fMRI findings suggest that self-reported fatigue in mTBI may relate to visual processing and effort perception. Long lasting activation associated with high levels of CNS drive might be related to changes in cortical homeostasis in the context of high effort.

Keywords: BOLD fMRI; FSS; Fatigability; MFIS; mTBI.

Fig. 1

Experimental paradigm and raw data.Schematic illustration of the experimental paradigm for (A) the submaximal contraction task and (B) the sustained MVC. (C) Visual feedback of the task and the produced force data provided to participants, note the horizontal cursor which was used to indicate the target force during the submaximal contractions. (D) Illustration of the force transducer. (E) Raw force data recorded from a mTBI participant during the sustained MVC task.

Fig. 2

BOLD-activation during the submaximal muscle contractions.Axial slices showing BOLD-activation during the submaximal contractions for the effect of ‘task’ with (A) the left hand and (B) the right hand. The effect of ‘force’ is shown in the lower panels for (C) the left- and (B) the right hand, and shows activation in the contralateral sensorimotor cortex and ipsilateral cerebellum. BOLD data are shown for both groups combined. The colour bar indicates voxel T-values, all voxels shown are significant at p < 0.05 family-wise error corrected. Z coordinates (MNI space) are shown for each slice, left is left according to neurological convention.

Fig. 3

Task and post-contraction activation for the sustained MVC.Axial slices showing BOLD-activation (A) during the sustained contractions and (B) for the post-contraction activation. Data are shown for both groups combined. The colour bar indicates voxel T-values, all voxels shown are significant at p < 0.05 family-wise error corrected. Z coordinates (MNI space) are shown for each slice, left is left according to neurological convention.

Fig. 4

Associations between voluntary muscle activation and BOLD-activity.Significant associations were observed between mean voluntary muscle activation during the sustained contraction and BOLD-activity in (A) the left middle temporal gyrus and (B) left intraparietal sulcus. For the post-contraction activation, associations were found in (C) the right insula and (D) the left middle cingulate cortex. For each of the clusters, the relationship between the mean beta’s (entire cluster) and voluntary muscle activation are provided (mTBI = red, control = blue; all p < 0.001). Peak coordinates (MNI space) of the clusters are provided above each plot. A cluster forming threshold of p < 0.001 (uncorrected) was used and all clusters are significant at p < 0.05 (family-wise error corrected), The colour bar indicates voxel T-values, left is left according to neurological convention. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 5

Associations between self-reported fatigue and BOLD-activity.In mTBI participants, significant associations were observed between FSS scores (corrected for HADS depression) and BOLD-activity during the submaximal contractions (A) with the left- and (E) right hand. For each of the clusters labelled in A and E, the relationship between the mean beta’s (entire cluster) and FSS scores are provided for the left (B-D) and right hand (F-H). Peak coordinates of the clusters are provided alongside each plot. A cluster forming threshold of p < 0.001 (uncorrected) was used and all clusters are significant at p < 0.05 (family-wise error corrected). All coordinates are shown in MNI space, the colour bar indicates voxel T-values, left is left according to neurological convention.