Abnormal task driven neural oscillations in multiple sclerosis: A visuomotor MEG study

Abstract

Multiple sclerosis (MS) is a debilitating disease commonly attributed to degradation of white matter myelin. Symptoms include fatigue, as well as problems associated with vision and movement. Although areas of demyelination in white matter are observed routinely in patients undergoing MRI scans, such measures are often a poor predictor of disease severity. For this reason, it is instructive to measure associated changes in brain function. Widespread white-matter demyelination may lead to delays of propagation of neuronal activity, and with its excellent temporal resolution, magnetoencephalography can be used to probe such delays in controlled conditions (e.g., during a task). In healthy subjects, responses to visuomotor tasks are well documented: in motor cortex, movement elicits a localised decrease in the power of beta band oscillations (event-related beta desynchronisation) followed by an increase above baseline on movement cessation (post-movement beta rebound (PMBR)). In visual cortex, visual stimulation generates increased gamma oscillations. In this study, we use a visuomotor paradigm to measure these responses in MS patients and compare them to age- and gender-matched healthy controls. We show a significant increase in the time-to-peak of the PMBR in patients which correlates significantly with the symbol digit modalities test: a measure of information processing speed. A significant decrease in the amplitude of visual gamma oscillations in patients is also seen. These findings highlight the potential value of electrophysiological imaging in generating a new understanding of visual disturbances and abnormal motor control in MS patients. Hum Brain Mapp 38:2441-2453, 2017. © 2017 Wiley Periodicals, Inc.

Keywords: MEG; multiple sclerosis; neuronal oscillations; post-movement beta rebound; visual gamma; visuomotor abnormalities.

Figure 1

Schematic diagram summarising the stimulus and analysis methods for our visuomotor experiment. [Color figure can be viewed at http://wileyonlinelibrary.com]

Figure 2

Raw cognitive scores for SDMT, CVLT‐II, and BVMT‐R tests for healthy controls (blue) and MS patients (green) with SE shown in red. Controls were found to score higher (significantly in SDMT and BVMT‐R) on all cognitive tests. [Color figure can be viewed at http://wileyonlinelibrary.com]

Figure 3

Beta band motor response. (a) Average t‐stat images for controls (top) and patients (bottom) contrasting 0–1 s to 1–2 s from button press in the beta (13–30 Hz) band. Images are shown on the same scale. (b) Average peak locations for healthy controls (top) and MS patients (bottom). The peak location for beta change is shown in blue, the location of maximum ERBD is shown in red, and the location of maximum PMBR in green. (c) TFSs for healthy controls (left) and MS patients (right) showing the mean percentage change in oscillatory power compared to baseline in response to a single button press. (d) The averaged beta response across all MS patients (red) and healthy controls (blue) with the SE shaded. The solid lines show the mean of the original responses, the dashed lines show the mean of the Weibull fits for each group. (e) A paired permutation test run on the rebound time for 20,000 iterations yielded a P value of 0.03 (two‐tailed test). The vertical red line shows the ‘true’ value in the difference between controls and patients, and blue shows the null distribution. Inset: the mean time‐to‐peak of the PMBR, measured using the Weibull fitting, with the SE shown on the error bars. (f) Correlation for MS patients between the time‐to‐peak of the rebound (x‐axis) and their corrected SDMT score (y‐axis). Plot shows a significant negative correlation (P = 0.008). (g) The amplitude spectrum for the resting state data (MS patients in red and healthy controls in blue) extracted from a region in left somatosensory cortex. Error bars show SE. [Color figure can be viewed at http://wileyonlinelibrary.com]

Figure 4

Visual gamma change. (a) TFS for controls (left) and patients (right) showing the mean percentage change in oscillatory power compared to baseline in visual cortex. (b) Average t‐stat images for controls (top) and patients (bottom) contrasting 0–2 s to 5–7 s from stimulus onset in the gamma (30–70 Hz) band. Images are shown on same scale. (c) Mean gamma (30–70 Hz) timecourses for healthy controls (blue) and MS patients (red) with the SE shown shaded. (d) A paired permutation test, yielding a P value of 0.04. The red line shows the ‘true’ mean difference between controls and patients, and blue shows the null distribution. Inset: mean percentage change in the gamma band (30–70 Hz) with SE shown in red. (e) The amplitude spectrum for resting state data (MS patients in red and healthy controls in blue) over a range of frequencies (for a location in the visual cortex) with SE shown on error bars. [Color figure can be viewed at http://wileyonlinelibrary.com]