Children with cerebral palsy have uncharacteristic somatosensory cortical oscillations after stimulation of the hand mechanoreceptors

Abstract

Numerous clinical investigations have reported that children with cerebral palsy (CP) have tactile discrimination deficits that likely limit their ability to plan and manipulate objects. Despite this clinical awareness, we still have a substantial knowledge gap in our understanding of the neurological basis for these tactile discrimination deficits. Previously, we have shown that children with CP have aberrant theta-alpha (4-14 Hz) oscillations in the somatosensory cortices following tactile stimulation of the foot. In this investigation, we evaluated if these aberrant theta-alpha oscillations also extend to the hand. Magnetoencephalography was used to evaluate event-related changes in the theta-alpha and beta (18-34 Hz) somatosensory cortical oscillations in groups of children with CP and typically developing (TD) children following tactile stimulation of their hands. Our results showed that the somatosensory theta-alpha oscillations were relatively intact in children with CP, which is in contrast to our previous results for foot tactile stimulations. We suspect that these inter-study differences may be related to the higher probability that the neural tracts serving the lower extremities are damaged in children with CP, compared to those serving the upper extremities. This inference is plausible since the participating children with CP had Manual Ability Classification System (MACS) levels between I and II. In contrast to the theta-alpha results, children with CP did exhibit a sharp increase in beta activity during the same time period, which was not observed in TD children. This suggests that children with CP still have deficits in the computational aspect of somatosensory processing.

Keywords: alpha; beta; magnetoencephalography; sensory; tactile; theta.

Figure 1

Time-frequency spectrograms for a representative typically developing child (A) and a child with cerebral palsy (B). Frequency (Hz) is shown on the y-axis and time (ms) is denoted on the x-axis (with 0 ms defined as stimulation onset). The event-related spectral changes during the hand stimulation task are expressed as percent difference from baseline (−500 to 0 ms). In each case, the MEG sensor with the greatest response amplitude that was located near the sensorimotor cortices, contralateral to the hand stimulated, was chosen. As shown, the TD child had a strong increase in 4–14 Hz alpha-theta band but also a decrease in 18–34 Hz beta band during the same time period. The child with cerebral palsy exhibited an increase in 4–14 Hz alpha-theta band and an increase in the 18–34 Hz beta band during the 25–225 ms time window. The color scale bar for both plots is shown to the far right.

Figure 2

Centroids of the group-mean beamformer images for each frequency bin (alpha-theta/beta), stimulated hand (left/right), and group (cerebral palsy/typically-developing). The amplitude of the peak voxel was extracted from each centroid of the alpha-theta and beta responses for the typically developing (TD) children and children with cerebral palsy (CP) for the respective hand that was stimulated. All statistics were conducted using these peak voxel amplitude values. For this experiment, the side of the body with the greatest impairment was stimulated for the children with CP, while the non-dominate hand was stimulated for the TD children.

Figure 3

Noise-normalized beamformer output image of the alpha-theta (4–14 Hz) response from 25 to 225 ms post-stimulation for a representative typically-developing (TD) child (left) and a child with cerebral palsy (CP; right). Both maps show voxels with pseudo-z values > 4.0. As can be discerned, both the children with CP and the TD children exhibited activity clustered in the central sulcus and postcentral gyrus of the hemisphere contralateral to stimulation. In both cases, this activity was maximal just posterior to the motor hand-knob feature of the precentral gyrus. As shown, both children had a strongly synchronized response (orange) to the external tactile stimulus. The images are displayed following the radiological convention (R=L).

Figure 4

Noise-normalized beamformer out image of the beta (18–34 Hz) response from 25 to 225 ms post-stimulation for a representative typically-developing (TD) child (left) and a child with cerebral palsy (CP; right). The map on the left shows voxels with pseudo-z values < −4.0, whereas as that on the right shows pseudo-z values > 4.0. As can be discerned, both the children with CP and the TD children exhibited activity clustered around the central sulcus and the postcentral gyrus, near the motor hand-knob feature in the hemisphere contralateral to stimulation. As shown, the TD children had a desynchronized response (blue) to the external tactile stimulation, while the children with CP had a synchronized response (orange). The images are displayed following the radiological convention (R=L).