Functional brain activity in persistent postural-perceptual dizziness (PPPD) during galvanic vestibular stimulation reveals sensitization in the multisensory vestibular cortical network

Abstract

Persistent postural-perceptual dizziness (PPPD) is often preceded by vestibular disorders. We applied galvanic vestibular stimulation (GVS) and related stimulus-evoked activity to individual ratings of perceived motion for each stimulus and to perceived egomotion thresholds by GVS and behavioural parameters outside the scanner: levels of functional disability by standardized questionnaires, visual motion coherence, passive egomotion perception by chair rotation and quantitative postural stability. We hypothesized that the preceding vestibular disorder predisposes to abnormal brain excitability by vestibular stimulation. All participants showed normal vestibular function tests on quantitative testing. GVS with different intensities was applied to 28 patients and 28 age- and gender-matched healthy participants (HC) in the scanner. After each stimulus, participants rated their perceived level of egomotion. GVS perception threshold was significantly lower in PPPD patients. Contrasting stimulus-identical GVS against a sham stimulus, group comparison revealed a stronger activation in the patient's supramarginal gyrus, insular cortex (operculum 3), and vermis. This stronger excitability was not related to the individual threshold of perceived egomotion by GVS. Patients rated GVS-evoked egomotion intensity by identical GVS intensities larger than HC but neural activity did not correlate with individual ratings of perceived egomotion by GVS. As GVS evoked larger egomotion and larger brain activation in patients, the ratio of brain activity to egomotion perception was not different between groups. GVS-evoked insular activity increased with the level of PPPD-related disability and postural imbalance. The larger activation in multisensory cortical vestibular network indicates a sensitization to vestibular stimuli eliciting egomotion perception which increases with levels of PPPD disability. It seems to reflect a sensory-neural amplification rather than an abnormal sensory-perceptual scaling.

Keywords: GVS; PPPD; Persistent postural-perceptual dizziness; Vestibular perception; fMRI.

Fig. 1

GVS-evoked brain activation (main effect, contrasting fixed GVS against the sham condition) in the bilateral parietal operculum (A), posterior insula (B) and lobule VII of the vermis (C) in PPPD patients and HC. Axial, coronal and sagittal view (from left to right), irrespective of the group (patient/HC). The coordinates (x, y, z) are in the Montreal Neurological Institute space. Whole brain analysis was performed using FWE-correction (p < 0.05). Tvalues are indicated along the colour bars. Abbreviations: HC = healthy control subject, PPPD = persistent postural-perceptual dizziness.

Fig. 2

Group comparisons (PPPD vs. HC) using ROI analysis for brain activations during fixed GVS. Fixed GVS elicited stronger activation in the patients’ supramarginal gyrus (A), the left operculum (OP4) (B), the right insula OP3 (C), the right IPL PF (D) and lobule VII of the vermis (E). The coordinates (x, y, z) are in the Montreal Neurological Institute space. Error bars indicate standard error of mean [ROI analysis, FWE p < 0.05 corrected]. * ≤ 0.05, *** < 0.001. Abbreviations: HC = healthy control subject, IPL PF = right inferior parietal lobe, OP3/4 = parietal operculum, PPPD = persistent postural-perceptual dizziness.

Fig. 3

Right IPL brain activation (A) and rating (B) for GVS intensities low, fix and high between PPPD patients and HC. Vestibular brain activation does not increase with enhanced GVS intensity, exemplarily shown in the right inferior parietal lobe. Patients perceived GVS-evoked egomotion (rating in %) significantly higher than HC for low and fix GVS with an overall increase of egomotion perception with increasing GVS intensity for both patients and HC. Patients are marked orange, HC grey. The coordinates (x, y, z) are given for the Montreal Neurological Institute space. Abbreviations: GVS = galvanic vestibular stimulation, HC = healthy control subject, IPL PF = right inferior parietal lobe.

Fig. 4

Insular cortex activation of OP3 by fixed GVS increases with level of PPPD-related functional disability (A: ALQ, B: NPQ) and postural sway speed in the simple platform condition (eyes open [EO], firm surface, no GVS, C). Patients are marked orange, HC grey. The coordinates (x, y, z) are in the Montreal Neurological Institute space. Abbreviations: ALQ = Athens-Lubeck-Questionnaire, EO = eyes open, GVS = galvanic vestibular stimulation, HC = healthy control subject, Ig2 = part 2 of the granular insula, OP3 = parietal operculum, NPQ = Niigata PPPD Questionnaire.