Turning and multitask gait unmask gait disturbance in mild-to-moderate multiple sclerosis: Underlying specific cortical thinning and connecting fibers damage

Abstract

Multiple sclerosis (MS) causes gait and cognitive impairments that are partially normalized by compensatory mechanisms. We aimed to identify the gait tasks that unmask gait disturbance and the underlying neural correlates in MS. We included 25 patients with MS (Expanded Disability Status Scale score: median 2.0, interquartile range 1.0-2.5) and 19 healthy controls. Fast-paced gait examinations with inertial measurement units were conducted, including straight or circular walking with or without cognitive/motor tasks, and the timed up and go test (TUG). Receiver operating characteristic curve analysis was performed to distinguish both groups by the gait parameters. The correlation between gait parameters and cortical thickness or fractional anisotropy values was examined by using three-dimensional T1-weighted imaging and diffusion tensor imaging, respectively (corrected p < .05). Total TUG duration (>6.0 s, sensitivity 88.0%, specificity 84.2%) and stride velocity during cognitive dual-task circular walking (<1.12 m/s, 84.0%, 84.2%) had the highest discriminative power of the two groups. Deterioration of these gait parameters was correlated with thinner cortical thickness in regional areas, including the left precuneus and left temporoparietal junction, overlapped with parts of the default mode network, ventral attention network, and frontoparietal network. Total TUG duration was negatively correlated with fractional anisotropy values in the deep cerebral white matter areas. Turning and multitask gait may be optimal to unveil partially compensated gait disturbance in patients with mild-to-moderate MS through dynamic balance control and multitask processing, based on the structural damage in functional networks.

Keywords: cognitive-motor interference; dynamic balance control; gait disturbance; multiple sclerosis; the timed up and go test.

FIGURE 1

Structural alterations between both groups and correlated areas with EDSS score in patients with MS. (a) Areas where cortical thickness tends to be larger (red) or smaller (blue) in patients with MS than controls are shown (uncorrected p < .05). (b) Areas where FA values are significantly lower (blue) in patients with MS than controls are shown (corrected p < .05). (c) Regions where lower FA values are correlated with higher EDSS score in patients with MS (corrected p < .05). CC, corpus callosum; EDSS, expanded disability status scale; FA, fractional anisotropy; MS, multiple sclerosis; SFG, superior frontal gyrus; SLF, superior longitudinal fasciculus

FIGURE 2

Correlated areas with total TUG duration in patients with MS. (a) Areas where thinner cortical thickness is correlated with longer total TUG duration (corrected p < .05, shown by blue). TPJ includes the inferior parietal lobule and banks of the superior temporal sulcus. (b) Outline of the correlated areas is shown by a white line and projected on the functional networks atlas 7Networks. The correlated areas are located on parts of the DMN (red) and VAN (violet). (c) Areas where lower FA values are associated with longer total TUG duration (corrected p < .05, shown by blue). DMN, default mode network; TPJ, temporoparietal junction; TUG, timed up and go test; VAN, ventral attention network. See other abbreviations in Figure 1 legend

FIGURE 3

Tractographies reconstructed from correlated areas with total TUG duration. (a) Tractography that projects from left TPJ (Figure 2a) and passes through the white matter areas where TBSS analysis identified (Figure 2c) (blue). Display threshold is 1000. (b) Tractography that projects from the left precuneus and adjacent areas (Figure 2a) and passes through the white matter areas where TBSS analysis identified (Figure 2c) (green). Display threshold is 1000. (c) Tractography which connects left TPJ (Figure 2a) with the left precuneus and adjacent areas (Figure 2a) (black). Display threshold is 100. (d) Outlines of reconstructed tractographies (blue, green, and black, defined in A–C) are shown on the functional network atlas 7Networks. Reconstructed tractographies overlap the DMN (red), FPN (orange) and VAN (violet). FPN, frontoparietal network; TBSS, tract‐based spatial statistics. See other abbreviations in Figures 1 and 2 legends

FIGURE 4

Correlated areas for stride velocity during cognitive‐DT circular walking and reconstructed tractographies. (a) Areas where smaller cortical thickness is correlated to slower stride velocity during cognitive‐DT circular walking (corrected p < .05, shown by red/yellow). (b) Outlines of the identified areas are projected on the functional network atlas 7Networks (shown by white line). (c) Tractographies connecting those cortical areas are reconstructed. They connect between the left precuneus and the left medial SFG and left cingulate cortex (green), between the left precuneus and the right postcentral gyrus (yellow) and between the left precuneus and the left TPJ (black). Display thresholds are 500 (green), 200 (yellow) and 20 (black) for each tract. (d) Outlines of reconstructed tractographies (green, yellow and black), defined in (d) are shown on the functional network atlas 7Networks. Reconstructed tractographies overlap the DMN (red), FPN (orange), and VAN (violet). DT, dual‐task; SMN, sensorimotor network. See other abbreviations in Figures 1, 2, 3 legends