FV 2. Modulation of the cortical vestibular network by means of rTMS and its neural correlate

Abstract

Repetitive transcranial magnetic stimulation (rTMS) has been applied to sensory research and for the treatment of disorders in psychiatry and neurology since 1993 ( Wassermann and Zimmermann, 2012 ). The widespread cortical vestibular system, however, has to date not been targeted. Aim of our study was to manipulate vestibular regions in the human cortex by temporary virtual lesioning through rTMS and analyse the changes in the perception of a galvanic vestibular stimulation (GVS) as well as the differing activation patterns in functional neuroimaging. The effects of rectangular and sinusoidal bimastoidal GVS were studied in 16 healthy volunteers (8 F; mean age 23 years) in a 3T scanner. Data was analysed with SPM12. Results of the baseline fMRI session were used as functional localizers for three rTMS target regions: (1) cingulate sulcus visual (CSv) bilaterally, (2) the right inferior parietal lobule (BA40), and (3) the right polysensory superior temporal sulcus area (STP) ( Smith, 2012 ). The vestibular fMRI sessions with GVS were repeated within minutes after virtual lesioning of the target regions with low-frequency rTMS (1 Hz, 1800 pulses, 90% active motor threshold) followed by a psychophysical debriefing and a clinical neurootological examination. Generally speaking virtual lesioning of vestibular cortex areas by means of rTMS resulted in a reduced, never in a more pronounced, activation pattern of the cortical vestibular regions. (1) RTMS of area CSv bilaterally led to a reduction in vestibular activation of area 6, area MST, and the inferior parietal lobule in both hemispheres. For unilateral galvanic stimuli this effect was constricted to the cerebral hemisphere ipsilateral to the side of excitation. More subjects perceived their head to be rotating in space during sinusoidal GVS instead of the sensation of the whole body twirling compared to all other sessions (72% vs 30%). (2) Virtual lesioning of the right inferior parietal lobule lead to reduced activation of the target region itself and area 6 in the same hemisphere during GVS. Here, subjects showed the largest reduction in grading the perceived amount of rotation induced by rectangular GVS (mean reduction of 45 ± 7° compared to the baseline session). The effect of a significantly reduced sensation of head rotation during GVS was seen in all target areas. (3) RTMS of area STP lead to a significant isolated deactivation of the left midbrain during GVS. There were no pathological findings in our healthy subjects after the virtual lesioning in the ensuing clinical neurootological examinations. The realignment parameters of the vestibular stimulations as a correlate for a potential vestibular motor response also gave no significant differences between the four sessions. This is the first demonstration by means of psychophysics, fMRI and rTMS that points at the potential for a manipulation of vestibular cortex regions. Global cerebral network effects within the vestibular system were only seen after rTMS lesioning of the vestibular midline structure CSv bilaterally. The resulting shift from body to isolated head rotation in space in the perception after GVS for the lesioned area CSv could reflect its role in egomotion processing. The quantitative perceptual findings for the right inferior parietal lobule as an rTMS lesion site on the other hand might suggest a leading role for this region in egocentric referencing within the confines of a three-dimensional environment.