Functional Neural Correlates of Transcutaneous Vagal Nerve Stimulation: Effects of Anatomical Site and Waveform Parameters * An exploratory High Resolution fMRI Study in Healthy Volunteers

The autonomic nervous system (ANS) is a bidirectional brain-body interface that
serves to integrate changes in the external environment with the internal
milieu and maintain homeostasis. Comprising of two broadly opposing arms, the
sympathetic (SNS) and parasympathetic nervous systems (PNS), the functions of
the ANS are considerable, spanning from metabolic control to cardiorespiratory
regulation and pain processing. Whilst it is well documented that the brain is
the central hub for the regulation of autonomic function, the use of
neuroimaging to investigate the association between brain morphology and
autonomic function, not least with advanced neuro-analytical techniques, is far
from comprehensive.

The main branch of the PNS is the vagus nerve. Although the vagus is
predominantly composed of afferent nerves, the efferent arm is postulated to
play a pivotal role in modulating visceral nociception, gastrointestinal
motility cardioregulation , and inflammation. Similarly, the SNS, due to its
multiple interactions with the peripheral and central nervous system, may also
influence peripheral inflammation and nociception.

Neuroanatomy of the vagal nerve: Efferent vagal nerve (the tenth cranial nerve,
*CN X*) fibres arise in the dorsal motor nucleus (DMNx) and the nucleus
ambiguus (NA), innervating the foregut, midgut, and aspects of the hindgut.
Afferent vagal fibres, meanwhile, originate in the mucosal or muscle layers of
the GI tract and have cell bodies in the nodose ganglia, which relay sensory
information to the nucleus tractus solitarii (NTS) located in the area
postrema. This is in close proximity to the DMNx forming the *dorsal vagal
complex*, an area of key importance in autonomic and limbic responses to
interoceptive physiology. From the dorsal vagal complex, visceral information
ascends to subcortical areas, including the hypothalamus, thalamus and amygdala
via the parabrachial nucleus. In turn this relays on to higher cortical areas
such as the insula cortex, cingulate and prefrontal cortices * which is
frequently referred to as the *central autonomic network* (CAN). The CAN
modulates visceral function and perception through descending inhibition. Taken
together, this bi-directional neurophysiological pathway has proposed to be a
major constituent component of the *brain-gut axis*, and has been a focus of
more than three decades of animal and human research

Quantifying activity of the ANS: Advances within autonomic neuroscience have
facilitated the development of many techniques to quantitatively assess
autonomic function, such as the evaluation of heart rate variability (HRV).
Moreover, with advances in signal waveform analytics it is possible to derive
autonomic indices from both electrocardiogram (ECG) and respiratory waveform
data. Amongst such measurements to objectively quantify SNS and PNS activity
are the cardiometrically derived parameters, referred to as cardiac sympathetic
index (CSI) and cardiac vagal tone (CVT), respectively. In contrast to the
traditional derivation of autonomic activity, such as by HRV, both CSI and CVT
yield superior temporal resolutions and have been associated with multiple
neurophysiological functions, including pain processing and the
neuro-endocrine-immune axis.

Perturbation of autonomic tone in clinical disorders: Whilst the ANS* primary
function is to maintain bodily homeostasis, it plays an important role at both
a central and peripheral level in modulating the pain experience. Specifically,
the SNS and PNS are considered to be pro and anti-nociceptive respectively,
with the parasympathetic nervous system posited to have anti-nociceptive
activity due to vagal nerve-mediated activation of key brain areas implicated
in descending analgesia. Furthermore, an imbalance of the ANS occurs frequently
in chronic pain disorders, such that the parasympathetic tone is relatively
diminished. Moreover, abnormal resting SNS and PNS activity has been reported
in a number of clinical disorders such as functional chest pain, irritable
bowel syndrome (IBS), inflammatory bowel disease, fibromyalgia, Ehlers-Danlos
syndrome and diabetes mellitus.

Neuromodulation of the ANS: The above observations, illustrating that autonomic
tone is perturbed in a variety of clinical conditions provides a rationale for
suggesting that increasing vagal tone may influence disease processes, or
possibly even attenuate them.
Over the last decade or so, techniques that modulate autonomic tone have been
investigated as potential methods to modify ANS tone in clinical disorders, so
called *autonomic neuromodulation*. One such method that has received much
attention is vagal nerve stimulation, a form of *neurostimulation* to the tenth
cranial nerve, the vagus nerve. Electrical vagal nerve stimulation (VNS) was
first used in humans in 1988 and is an efficacious treatment for drug-resistant
epilepsy. Traditional VNS is undertaken in a procedure where a bipolar helical
electrode is placed around the cervical vagal nerve, which is connected to a
pulse generator placed in a subcutaneous pocket in the chest, not dissimilar to
a cardiac pacemaker. However, this method of VNS necessitates surgical
implantation with its attendant risks and complications. Recently, an external
transcutaneous VNS (tVNS) method, consisting of small electrodes to interface
with the concha of the outer ear has become available. The auricular branch of
the vagus nerve innervates the concha of the ear and is located directly under
the skin, making it a suitable target for transcutaneous stimulation. tVNS has
been demonstrated to be safe, well tolerated and have a high degree of
user-friendliness.

Gaps in our knowledge;
1) site of stimulation: In addition to auricular-targeted tVNS, the cervical
branch of the vagus nerve over the neck has been suggested an
additional/alternative stimulation site. However, there is presently minimal,
if any, evidence to suggest the utility of one stimulation site over another
from an efficacy perspective. Simply put, it is not known if one site offers
*superior* vagal nerve stimulation to the other. The vast majority of tVNS
studies do not concurrently measure autonomic indices in real-time, further
casting doubt on what tVNS is actually stimulating. In addition, there remains
minimal evidence whether both even achieve a neural signature of
parasympathetic activation with the use of fMRI.

2) waveform parameters of stimulation:
To date there is a paucity of set waveform parameters used in tVNS. Rather,
waveform frequency and intensities seem to vary drastically between studies.
Moreover, some groups have even gone so far as to patent individual waveform
characteristics (40). That being said, similar to the choice of auricular or
cervical sited tVNS, there is a lack of evidence to support the use of
particular electrical waveform parameters. Moreover, we question whether
in-fact there is a specificity of waveform parameters to achieve tVNS, or
rather variable/random frequencies achieve comparable results.

The advent of 7 Tesla MRI: The advantage of scanning at 7 Tesla (7T) over 1.5
Tesla and 3 Tesla is increase in the Signal to Noise Ratio (SNR), Contrast to
Noise Ratio (CNR), resolution and/or a decrease in scanning time. The benefit
of scanning at 7T is believed to outweigh the small risk of dizziness while
entering the bore (not during the actual scan itself). Note that the 7 Tesla
scanner is becoming more and more standard and is certified for research.
Worldwide, there are more than 50 scanners used for human basic and clinical
research and no negative effect have been reported. According to the guidelines
from the U.S. Food and Drug Administration (FDA), clinical MR systems using
static magnetic fields up to 8.0 Tesla are considered *non-significant risk*
for adult patients (level was set in 2003)
(http://www.mrisafety.com/safety_article.asp?subject=229). The human body is
non- magnetic and therefore the static magnetic field (up to 14 Tesla or more)
does not harm biological tissue. People, including elderly and patients,
tolerate the experience of a 7T scanner without many difficulties or long-term
side effects. In The Netherlands, there currently are three 7 Tesla instruments
(Leids Universitair Medisch Centrum, University Medical Center Utrecht, and
Donders Institute for Brain, Cognition and Behaviour, located in Nijmegen).
These groups have been carrying out patient research for several years, with
local ethical approval and no reported adverse effects or safety issues.
Moreover, the use of tVNS in a 7T MR environment is concurrently studied at the
Maastricht Scannexus MR site with nil procedural complications reported to date
[Protocol ID NL51297.068.14].

Summary of Study rationale:
There exist multiple knowledge gaps in the use of tVNS, which concurrently
hinder the advancement of this research niche, not least the development of
neuromodulatory devices for patient populations. In particular, to summarise,
these knowledge gaps are the following:
i) Do auricular or cervical tVNS differ in effect and/or efficacy in
neuromodulation, including at the brain level? and
ii) Do fixed or variable waveform parameters differ in effect and/or efficacy
in neuromodulation, including at the brain level?

The aim of this exploratory (pilot) study is 1) To study the effect of tVNS
with high-resolution 7-Tesla (7T) functional magnetic resonance imaging (fMRI)
and concomitant autonomic monitoring by cardiac and respiratory waveform
analytics; 2) To explore the functional brain differences (using functional
magnetic resonance imaging, fMRI) between cervical [neck] vagal nerve
stimulation and auricular [ear] vagal nerve stimulation; 3) To explore the
functional brain differences between various vagal nerve stimulation parameters.

Primary objective:
To explore the possible functional brain differences between auricular and
cervical tVNS.

Secondary objectives:
1. To explore the functional brain differences between various stimulation
parameters in cervical-tVNS.
2. To explore the functional brain differences between various stimulation
parameters in auricular-tVNS.
3. To compare brain effects between 1) and 2).
4. To explore the differences in auricular and cervical tVNS in augmenting PNS
tone.
5. To explore the differences in varying tVNS parameters in augmenting PNS tone.
6. To explore if stimulus epoch duration relates to demonstrable change in
functional brain activity and/or PNS tone.
7. To explore brain effects of subliminal (sub-threshold) auricular tVNS.

Exploratory cross-sectional study (pilot study)

Transcutaneous vagal nerve stimulation.

There are 4 main sub-interventions of this:
i) Auricular fixed-frequency tVNS;
ii) Cervical fixed-frequency tVNS;
iii) Auricular variable-frequency tVNS and
iv) Cervical variable-frequency tVNS.

Subjects will be randomised to receive either fixed or variable-frequency
first, however subjects will always receive both frequencies (in other words,
randomisation only determines the order of stimulation type. 2 visits will be
planned with a wash-out period of two weeks in between.
Visit 1 subjects will receive both forms of cervical stimulation, visit 2
subjects will receive both forms of auricular stimulation. In addition,
subjects will receive subliminal auricular stimulation at the end of visit 2.

Volunteers will not benefit from participating in this study. There are no
risks associated with the use of tVNS, including in magnetic resonance imaging
(MRI). Moreover, tVNS in the MR environment has been approved by the Scannexus
safety board and, thus far, 16 individuals (13 healthy older individuals and 3
patients with preclinical Alzheimer*s disease) underwent simultaneous tVNS-fMRI
at this prospective site, with this particular equipment before with nil
procedural complications/adverse events (NL51297.068.14). Given the nature of
tVNS (by definition, *neurostimulation*), it can induce a transient tingling
feeling but does not cause pain. Moreover, the equipment to be used in this
study, which are also used in the aforementioned study, has been closely
developed in line with the Respiratory-gated Auricular Vagal Afferent Nerve
Stimulation (RAVANS) system (Napadow lab * Boston). Ultra-high magnetic field
MRI is very safe and no adverse events are anticipated when taking into account
all contra-indications. Solely *Certified Users* will operate the MRI according
to approved guidelines and protocol. Subjects will be screened for
contraindications (metal implants etc.) prior to inclusion and again on the day
of scanning. Some participants may experience mild vertigo, nausea or a metal
taste when entering the MRI environment. In extremely rare cases, a small burn
may arise due to heating caused by radiofrequency. All participants will be
informed about any unexpected medical findings (MRI findings). In the rare
event the subject does not wish to be informed, they would not be permitted to
participate in this study.