Acute effects of auricular vagus nerve stimulation on the gut-brain axis in functional dyspepsia

81 

Abstract Submission 

Roberta Sclocco¹, Harrison Fisher², Andrew Bolender², Junhwan Lee³, Braden Kuo², Vitaly Napadow⁴

¹Spaulding Rehabilitation, Harvard Medical School, Charlestown, MA, ²Massachusetts General Hospital, Boston, MA, ³Korea Institute of Oriental Medicine, Daejeon, Korea, Republic of, ⁴Massachusetts General Hospital, Harvard Medical School, Charlestown, MA

Roberta Sclocco  
Spaulding Rehabilitation, Harvard Medical School
Charlestown, MA

Harrison Fisher  
Massachusetts General Hospital
Boston, MA

Andrew Bolender  
Massachusetts General Hospital
Boston, MA

Junhwan Lee  
Korea Institute of Oriental Medicine
Daejeon, Korea, Republic of

Braden Kuo  
Massachusetts General Hospital
Boston, MA

Vitaly Napadow  
Massachusetts General Hospital, Harvard Medical School
Charlestown, MA

Functional dyspepsia (FD) is a disorder of gut-brain interactions associated with upper gastrointestinal (GI) pain and discomfort, but lacking effective therapies. FD has been linked with dysregulation of the gut-brain axis, mediated by abnormal vagal afference and brainstem neurocircuitry. In fact, the vagus nerve is intimately involved in autonomic control of the upper GI tract, with afferent projections to medullary brainstem nuclei, mainly to the nucleus tractus solitarii (NTS). Our own functional Magnetic Resonance Imaging (MRI) data found altered NTS-cortical connectivity in FD patients compared to controls in a post-meal state, associated with slower propagation of gastric peristaltic waves (Sclocco et al., 2022), as measured by our recently developed abdominal cine-MRI whole-stomach dynamic imaging approach (Sclocco et al., 2021). We propose that such maladaptive plasticity across different aspects of gut-brain communication can be targeted by transcutaneous auricular vagal nerve stimulation (taVNS), which interacts with brainstem vagal circuitry.

15 FD patients (13F, 29.1±13.2y/o) and 15 healthy controls (HC; 10F, 32.1±7.7y/o) consumed their maximum tolerable amount of a 470ml high-calorie food-based contrast meal (pineapple-based for enhanced gastric MRI contrast). Post-meal, subjects were scanned with stomach MRI (+15, +45, +70min) and brain fMRI (+25, +55, +80min) at 3T, while experiencing active ("A", 1.5s stimulation trains delivered at 100Hz in left cymba concha during exhalation) or sham ("S", no current) taVNS on two separate visits (Fig.1A). During stomach scans, 4D cine-MRI were collected continuously for 5min (temporal resolution 7s). After gastric segmentation, peristaltic propagation velocity in the antrum was calculated by comparing cross-sectional area time series from two sections 10 antral slices apart to track the duration of time for a peristalsis contractile wave peak to propagate from one location to the next. Velocities were compared between conditions using a mixed effects model with fixed effects of Scan and Ingested Volume. During brain fMRI, resting-state data were acquired using an accelerated multiband sequence (2mm isotropic voxels, 1.27s TR, 288 volumes). Following preprocessing (FMRIPREP, AFNI), average BOLD timeseries extracted from a region previously localized as NTS were used to generate seed-to-voxel whole-brain functional connectivity maps. These maps were combined across scans for each subject in a fixed effects model, and then contrasted between conditions (z>2.3, pFWE<0.05).

taVNS did not modulate gastric function in HC. In FD, peristaltic propagation velocity was on average 0.7mm/s faster during active taVNS compared to sham (β ̂=0.67, SE=0.28, t=2.39; Fig.1B). Since there was no significant effect of time, data were averaged across post-meal time points, and follow-up comparison confirmed significantly higher velocity during active taVNS (A: 5.1±0.3mm/s (mean±SEM); S: 3.7±0.4mm/s; p=0.017). Relative to HC, FD patients demonstrated a significant reduction in NTS connectivity to anterior cingulate cortex (ACC) and medial and ventrolateral prefrontal cortices (m/vlPFC; Fig.1C) during active taVNS compared to sham. Interestingly, we had previously reported higher NTS connectivity in FD compared to HC in a similar set of brain regions.

Our analysis found that taVNS acutely modulates gut-brain communication in FD patients, potentially restoring impaired gastric motility by reversing altered cognitive processing of interoceptive (gastric) signaling. Further, our MRI approach allowed for a fully non-invasive evaluation of gut-brain interaction in a post-meal state. Future work focusing on longitudinal effects of taVNS will inform therapeutic applicability in disorders of gut-brain interaction such as FD.

Non-Invasive Stimulation Methods Other ¹

Perception: Pain and Visceral ²

Brainstem

Other - Vagus Nerve; Brain-Gut Axis; Vagal Nerve Stimulation