This study was aimed to evaluate the effects of lateralized motor impairment on somatosensory perception and brain processing of non-painful tactile stimulation in individuals with bilateral cerebral palsy (CP). Our findings revealed that although individuals with CP exhibited lower pain sensitivity than healthy controls, there were no significant differences on touch sensitivity, pain sensitivity or proprioception between CP individuals with right- (RMI) and left-dominant motor impairments (LMI). Regarding somatosensory brain processing, we found that lip stimulation elicited higher beta power, but more similar SEP amplitudes over the contra- and the ipsilateral hemispheres in CP individuals than in healthy controls. Moreover, lip and thumb stimulations elicited smaller and more symmetrical SEP amplitudes, as well as reduced beta power (only for thumb stimulation) in CP individuals with LMI than in CP individuals with RMI.
Thus, our results revealed an altered somatosensory processing in individuals with CP, as measured by SEP amplitudes and frequency power. The pattern of brain activation displayed by our individuals with CP after stimulation of thumbs and lips seems to be different from that observed in healthy controls. Reduced beta power and enhanced SEP amplitudes over somatosensory cortices have been observed in healthy controls following somatosensory stimulation, suggesting that these changes might be interpreted as an activation of cortical networks involved in somatosensory processing . In the present study, healthy individuals showed a desynchronization in the beta frequency band in response to touch stimulation of lips and thumbs. In contrast, our individuals with CP appeared to display an increased beta power over the contralateral hemisphere, particularly in CP individuals with RMI. These findings are in agreement with previous studies showing that, although CP is mainly characterized by motor impairments, brain processing of incoming somatosensory information is also significantly altered in this pathology. Thus, for instance, neuroimaging research has found that children with periventricular leukomalacia show more severe injury in posterior white matter fibers connecting the thalamus to the sensory cortex than in descending corticospinal tracts . Moreover, it has been demonstrated that deficits on somatosensory processing (reduced touch sensitivity, proprioception and strength) in CP could be related to injury severity of diffuse thalamocortical projections to somatosensory and parietal cortices . In this sense, our results provide further empirical evidence for an abnormal brain processing of bodily information in CP individuals.
Motor function seems to be often asymmetric in CP, even in individuals with bilateral lesions [36–38]. This asymmetry has been reported by using physiological measures such as nerve conduction velocities or dichotic listening processing [39, 40]. In the present study, we found significant differences on the pattern of hemispheric brain activation elicited by somatosensory stimulation depending on the dominant side of motor impairment. In this sense, participants with RMI displayed enhanced responses to bodily stimulation over contralateral as compared to ipsilateral hemispheres, whereas individuals with LMI showed no hemispheric differences. Moreover, CP subjects with RMI showed higher SEP amplitudes when the affected side of the body was stimulated, while no differences on brain activation were found in LMI when the right or left body side was stimulated. In a previous study of our lab , we have shown that CP children and adults elicited higher SEP amplitudes in contralateral hemisphere when left dominant side of motor impairment was stimulated as compared to right dominant side of motor impairment. Thus, it seems that CP individuals with left- and right-dominant motor impairments differ on brain processing of lateralized bodily information. An unusual pattern of bilateral cortical activation and recruitment of ipsilateral tracts have been usually linked to widespread cortical reorganization after brain lesions [41–44]. Moreover, our findings are in agreement with previous studies showing that motor impairments are different depending on the paretic dominant side of motor impairment after unilateral CP lesions. Thus, for instance, Van Kampen and colleagues  reported that children with left hemiparesis had longer decision time when asked to intercept a ball located 4 meters away and started their reach movement earlier than healthy controls and children with right hemiparesis. In addition, Craje and colleagues  observed that participants with right hemiparesis had more difficulties in switching between different grip types than participants with left hemiparesis. Our findings are also in agreement with previous data showing that bilaterally impaired CP children with spastic diplegia displayed higher intrahemispheric coherence for delta, beta and theta EEG bands in the left than in the right hemisphere . In our opinion, all these results support the view that brain damage to right or to left hemisphere may have led to different plasticity mechanisms in cerebral palsy. In the present study, we observed that CP individuals with RMI displayed an asymmetrical pattern of brain activity more similar to that exhibited by healthy controls  than that of individuals with LMI. One possible explanation for these differences could be that damage of the left hemisphere releases the right hemisphere from its non-dominant role, while damage of the right hemisphere only emphasizes the usual role of the left hemisphere. Thus, for example, due to the dominant role of the left somatosensory cortex in sensorimotor integration for complex finger movements , damage of this hemisphere may have lead to a contralateral directed plasticity phenomenon in CP individuals with RMI. Nevertheless, further research should be necessary to elucidate the role of hemispheric dominance on somatosensory processing in CP individuals and the potential mechanisms of this differentiation.
Our study has some limitations, which should be taken into account for the interpretation of the results. Firstly, although our sample of persons with CP seems to be representative of a large population in the community, sample was small and heterogeneous. Thus, it could be that our sample size was not sufficient to reach an appropriate statistical power and to show differences between individuals with LMI and RMI on behavioral measures of somatosensory processing. In addition, the wide range of age, cognitive levels and underlying brain lesions as well as the presence of different subtypes of CP, have also limited the conclusions of the study. Moreover, it is possible that CP individuals may have been developing a left hemispheric dominance before their lesion. Nevertheless, the early onset of this pathology (in most cases, with a prenatal beginning) minimizes the influence of a possible left-hemispheric dominance development in comparison with several years of brain reorganization in the childhood and adolescence. On the other hand, the study should be replicated in hemiparetic CP individuals with clear and limited brain lesions. Finally, somatosensory-evoked potentials provide information from brain functioning arising from sensory cortices and, therefore, the influence of subcortical brain structures in somatosensory processing remains unexplored. Nevertheless, our study lays the scientific basis for implementation of further research on a scarcely investigated topic.