Abstracts

Rationale: Vagal nerve stimulation (VNS) is an adjunctive treatment for refractory epilepsy. Currently there is a lack of predictors of efficacy1. Transcutaneous VNS (tVNS) is a non-invasive device stimulating the auricular branch of the vagus nerve on the cymba conchae. fMRI studies showed similar activation patterns between cervical VNS and tVNS2. Moreover, tVNS was shown to affect pain perception3,4. This study aimed to characterize the effects of tVNS on the behavioral and EEG brain responses elicited by nociceptive heat stimuli vs. non-nociceptive vibrotactile stimuli delivered during tVNS stimulation, to explore whether these responses could be used in the future as biomarkers of VNS effects. Methods: Twelve healthy subjects were tested during a 2-day randomized protocol. Innocuous vibrotactile stimuli (selectively activating Aδ-fiber low-threshold mechanoreceptors) and nociceptive laser heat stimuli (selectively activating Aδ- and C-fiber thermonociceptors) were delivered to the right hand. Auricular tVNS (25Hz, 250µs, 30s ON/ 30s OFF; Cerbomed, Germany) was administered on the cymba conchae. Sham stimulation only differed by location (earlobe). Four blocks of somatosensory stimuli were delivered, each consisting of an alternating sequence of 20 vibrotactile (300Hz, 100ms) and 20 laser (60°C, 120ms) stimuli, delivered during both ON and OFF phases of auricular stimulation. For each stimulus, intensity of perception was assessed using a numerical rating scale. These different measures were compared using a 2-way repeated-measures ANOVA with the factors ‘stimulation’ (tVNS vs. sham) and ‘phase’ (ON vs. OFF). Sensory detection thresholds were assessed before and during auricular stimulation, and compared using a 2-way repeated-measures ANOVA with the factors ‘stimulation’ (tVNS vs. sham) and ‘time’ (before vs. during stimulation). Vibrotactile and laser event-related brain potentials (ERPs) were recorded using 32 channels. Results: For both somatosensory modalities, there was no significant interaction between the factors ‘stimulation’ and ‘time’ on sensory detection thresholds (vibrotactile: F=0.871, p=.357; laser C-fiber threshold: F=0.024, p=.879; laser Aδ-fiber threshold: F=0.185, p=.669), suggesting that tVNS did not significantly alter the ability to detect vibrotactile or laser stimuli. Similarly, for both somatosensory modalities, there was no main effect of ‘stimulation’ (vibrotactile: F=0.036, p=.851; laser: F=0.167, p=.685) and no significant interaction between the factors ‘stimulation’ and ‘phase’ on the reported intensities of perception, suggesting that the perception of vibrotactile and laser stimuli during tVNS and sham stimulations, as well as the intensity of perception during the ON and OFF phases of tVNS and sham stimulations, did not differ significantly. Clear vibrotactile- and laser-evoked ERPs were recorded both during the ON and the OFF phases of tVNS and sham stimulation. Similarly, there was no main effect of ‘stimulation’ (vibrotactile: F=0.051, p=.822; laser: F=0.005, p=.944) and no interaction between the factors ‘stimulation’ and ‘phase’ (vibrotactile: F=0.088; p=.769; laser: F=0.012; p=.912) on magnitude of vibrotactile- and laser-evoked ERPs, suggesting that these brain responses were not significantly affected by tVNS. Conclusions: This preliminary study shows that somatosensory-evoked EEG responses can be recorded during both the ON and the OFF phases of t-VNS. However, with a current sample size of 12 subjects, we did not find any significant effects of t-VNS on the elicited behavioral and brain responses. Funding: FSR (Fond Spécial de Recherche), from the Wallonia-Brussels federation