AGING 2019, Vol. 11, No. 14
www.aging-us.com 4836 AGING
Ageing is associated with changes in autonomic
nervous system function and is characterized by
increases in sympathetic and decreases in
parasympathetic nervous activity [1, 2]. Such autonomic
changes can be detrimental to heart function, emotion,
mood and gut function, and may play a role in a range
of conditions that increase in prevalence with ageing,
including heart failure [3, 4], hypertension  and
depression [6, 7]. These conditions are typically
accompanied with increases in medication consumption
and decreases in quality of life (QoL). Preventing or
reducing age-related changes in autonomic balance may
therefore improve health in older individuals, as well as
increase their independence, QoL and mood
(particularly depression). Potential benefits include
reduced risk of mortality and reductions in the need for
medication and/or hospitalisation.
Interventions which aim to boost parasympathetic
activity and/or decrease sympathetic activity include
vagus nerve stimulation (VNS) and transcutaneous
vagal nerve stimulation (tVNS). VNS involves
surgically implanting an electrode around the cervical
vagus nerve and a generator unit in the thoracic wall
. However, due to its invasive nature, technical
complications and side-effects e.g. pain, coughing,
hoarseness of voice [8, 9], potentially simpler and safer
therapies are of interest.
Effects of transcutaneous vagus nerve stimulation in individuals aged
55 years or above: potential benefits of daily stimulation
1, Lucy Atkinson1, Aaron Murray1, Jennifer Clancy2, Susan Deuchars1
School of Biomedical Sciences, Faculty of Biological Sciences, University of Leeds, Leeds, LS2 9JT, UK
School of Life Sciences, University of Glasgow, Glasgow, G12 8QQ, UK
Jim Deuchars; email: J.Deuchars@leeds.ac.uk
: vagus nerve stimulation, autonomic nervous system, neuromodulation, quality of life, mood
May 9, 2019 Accepted: June 28, 2019 Published: July 30, 2019
Bretherton et al. This is an open-
access article distributed under the terms of the Creative Commons Attribution
), which permits unrestricted use, distribution, and reproduction in any medium, provided the original
author and source are credited
is associated with attenuated autonomic function. Transcutaneous vagal nerve stimulation
autonomic function in healthy young participants. We therefore investigated the effects of a
of tVNS (studies 1 and 2) and tVNS administered daily for two weeks (study 3) in volunteers aged ≥
tVNS was performed using modified surface electrodes on the tragus and connected to a
nerve stimulation (TENS) machine. Study 1: participants (n=14) received a single session of tVNS
Study 2: all participants (n=51) underwent a single session of tVNS. Study 3: participants (n=29)
tVNS for two weeks. Heart rate variability and baroreflex sensitivity were derived. Quality of life
and sleep were assessed in study 3. tVNS promoted increases in measures of vagal tone and
with greater increases in baroreflex sensitivity than sham. Two weeks of daily tVNS
of autonomic function, and some aspects of QoL, mood and sleep. Importantly, findings showed
in measures of autonomic balance were more pronounced in participants with greater
prevalence. This suggests it may be possible to identify individuals who are likely to
ignificant benefits from tVNS.
www.aging-us.com 4837 AGING
tVNS is a simple, non-invasive and inexpensive therapy
that involves stimulating the auricular branch of the
vagus nerve (ABVN) at outer parts of the ear,
conferring autonomic benefits in healthy volunteers
. For instance,  revealed that 15 minutes of
tVNS administered to the tragus significantly increased
heart rate variability (HRV), at least partly through
reductions in sympathetic nerve activity. An increase in
parasympathetic activity is also likely, since tVNS
increases spontaneous cardiac baroreflex sensitivity
(BRS) even in healthy young males . Baseline HRV
decreases with increasing age [13-16] and since tVNS
induces larger increases in HRV in participants with
lower starting HRV , tVNS could be particularly
effective in older compared to younger participants. We
therefore investigated if tVNS could ameliorate age-
related changes in autonomic function.
We report on three studies into the effects of tVNS in
participants aged ≥ 55 years. In the first study we
compared the effects of acute tVNS on cardiovascular
autonomic function with the effects of sham stimulation
by measuring HRV and BRS. Since not all participants
responded to tVNS, we examined if it was possible to
identify potential tVNS responders from baseline
parameters. In the second study, we explored the effects
of acute tVNS on autonomic function in the same age
group by expanding the sample. The final study aimed
to examine how daily tVNS (a 15-minute session
administered once daily for two weeks) impacted
measures of autonomic function, as well as health-
related QoL, mood and sleep.
14 participants aged ≥ 55 years with no previous
medical history of hypertension, cardiac disease,
diabetes mellitus or epilepsy were enrolled on the study
(see Supplementary Table 1).
tVNS vs. sham effects on cardiac baroreflex and
Change in baroreflex sensitivity (BRS) between
baseline and tVNS significantly differed between the
tVNS and sham visits (p = 0.028): there was a
significantly greater increase in BRS during the tVNS
visit (3.28 ± 0.59 ms/mmHg) compared to the sham
visit (0.81 ± 0.68 ms/mmHg). Baseline heart rate
variability (HRV), measured as ratio of LF/HF power,
significantly predicted response to tVNS (R2 = 0.772, p
< 0.001, see Figure 1), where higher resting LF/HF ratio
was associated with greater decreases during tVNS.
Removing the potential outlier with a baseline LF/HF
ratio > 5 had a slight impact on the regression (R2 =
0.480, p = 0.009). This HRV analysis revealed that the
LF/HF ratio response of four participants was greater
than a 20% increase, corresponding to a definition of
responders previously applied .
Fifty-one healthy participants aged ≥ 55 years were
recruited. Two participants were excluded due to the
presence of > 3 ectopic heartbeats in any given five-
minute period (n = 1 female). A further participant was
excluded due to a respiration rate < 10 breaths per
minute (n = 1 male). Characteristics of the final sample
for study 2 are presented in Supplementary Table 2.
tVNS significantly increased heart rate variability in
≥ 55-year-old participants
Analysis of the whole cohort in study 2 revealed that
measures considered to reflect vagal activity were
significantly higher during tVNS (RMSSD: p = 0.007;
pRR50: p = 0.005; SD1: p = 0.007; BRS: p = 0.001) and
recovery (HF power: p = 0.008) compared to baseline
(see Supplementary Table 3).
In addition, LF power, SD2 and nSD2, representing
combined sympathetic and parasympathetic influences
on the heart, were significantly higher during tVNS (p =
0.007, p < 0.001 and p = 0.005 respectively) and
recovery (p = 0.001, p < 0.001 and p = 0.002
Figure 1. Baseline LF/HF ratio significantly predicted response
(change) to tVNS.
www.aging-us.com 4838 AGING
Measures of overall variability were significantly
impacted by tVNS. Total power, mean RR interval, Δ
RR, SDRR and S were significantly higher during tVNS
(all p < 0.001) and recovery (all p < 0.001) compared to
Response to tVNS was predicted by baseline LF/HF
Linear regression revealed a significant prediction
between baseline LF/HF ratio and Δ LF/HF ratio such
that a higher LF/HF ratio predicted a greater decrease to
tVNS (R2 = 0.498, p < 0.001, Figure 2) in the cohort of
tVNS responders showed shifts towards
parasympathetic prevalence during tVNS
Using the response definition employed in study 1 (20%
decrease in LF/HF ratio), 16 responders and 32 non-
responders were identified. Further analysis revealed
that baseline measures reflecting parasympathetic
activity (HF power: p = 0.013; nuHF: p = 0.001;
RMSSD: p = 0.029; SD1: p = 0.029; nSD1: p = 0.029
and BRS: p = 0.011) and overall variability in HR (S: p
= 0.034) were significantly lower in responders
compared to non-responders (see Supplementary Table
4). In addition, measures of baseline sympathovagal
outflow (nuLF: p = 0.001) and sympathovagal balance
(LF/HF ratio: p < 0.001) were significantly higher in
responders compared to non-responders. This suggests
that at rest, responders had significantly lower vagal
tone and greater sympathetic prevalence compared to
non-responders. There were no statistically significant
differences in demographic information between
responders and non-responders (see Supplementary
The characteristics of the final sample of 26 participants
are presented in Table 1.
Daily tVNS sessions improved baseline measures of
Baseline Δ RR, an indicator of cardiac vagal tone, (R2 =
0.330, p = 0.002) and BRS (R2 = 0.330, p = 0.002)
during visit 1 significantly predicted change during
baseline at visit 2: the lower the baseline Δ RR and BRS
in visit 1, the greater the increase in baseline Δ RR and
BRS at visit 2 (Figure 3).
Figure 2. Baseline LF/HF ratio significantly predicted change
in LF/HF during tVNS (Δ LF/HF ratio).
Table 1. Summary of characteristics of the final sample of study 3.
Final sample size (n)
Gender (frequency of males)
Baseline systolic blood pressure (SBP, mmHg)
Baseline diastolic blood pressure (DBP, mmHg)
Baseline mean arterial pressure (MAP, mmHg)
www.aging-us.com 4839 AGING
Measures of autonomic tone after two weeks of daily
Two-way repeated measure ANOVAs revealed that
measures of vagal tone: RMSSD, pRR50, SD1 and
nSD1, were significantly higher during visit 2 compared
to visit 1 when values during all three recordings
(baseline, tVNS, recovery) were combined across each
visit (main effect of visit: RMSSD, p = 0.016; pRR50, p
= 0.010; SD1, p = 0.016; nSD1, p = 0.022, see
Supplementary Table 6). Also, HF power tended to be
higher during visit 2 than during visit 1 (main effect of
visit: p = 0.051). Similarly, measures reflecting short
and long terms variations in HR, total power, SDRR,
SD2 and S were significantly greater during visit 2 than
visit 1 (main effect of visit: total power, p = 0.046;
SDRR, p = 0.024; SD2, p = 0.035 and S, p = 0.012).
Two-way repeated measure ANOVAs also showed that
total power (p = 0.010), SDRR (p = 0.001), SD2 (p =
0.001), S (p = 0.017), mean RR interval (p < 0.001) and
nSD2 (p = 0.001) were significantly higher during
tVNS compared to baseline (main effect of condition).
Some of these changes persisted into the recovery
period: total power (p = 0.034), SDRR (p = 0.008),
SD2 (p = 0.004), mean RR interval (p = 0.005) and
nSD2 (p = 0.008) were significantly higher during
recovery than in baseline. LF power was also
significantly higher during tVNS compared to baseline
(p = 0.033, main effect of recording: p = 0.039, see
Supplementary Table 7).
A significant interaction effect transpired for Δ RR (p =
0.036, see Figure 4). Further analysis revealed that Δ
Figure 3. Visit 1 baseline Δ RR (A) and BRS (B) significantly predicted change at visit 2 baseline, where lower baseline Δ RR (A) and
BRS (B) in visit 1 were associated with greater increases in baseline Δ RR (A) and BRS (B) in visit 2. In A, Δ Δ RR reflects the difference
between maximum and minimum RR intervals (Δ RR) between visit 1 baseline and visit 2 baseline. In B, Δ BRS reflects the difference
in BRS between visit 1 baseline and visit 2 baseline.
Figure 4. Δ RR significantly differed both within and between
the two visits (p = 0.036); * = significantly different to visit 2
www.aging-us.com 4840 AGING
RR was significantly higher during visit 2 recovery
compared to visit 1 recovery (p = 0.001). Furthermore,
in visit 2 only, Δ RR was significantly greater during
recovery than baseline (p = 0.024).
Response to tVNS in both visits was predicted by
baseline LF/HF ratio
Linear regressions revealed a statistically significant
prediction between baseline LF/HF ratio and change
during tVNS in visits 1 (R2 = 0.352, p = 0.001) and 2
(R2 = 0.697, p < 0.001). As illustrated in Figure 5,
higher baseline LF/HF ratio values were associated with
greater decreases in LF/HF ratio during tVNS in both
Responders had higher baseline sympathetic
prevalence than non-responders
Nine responders and 17 non-responders were identified
in visit 1. Responders had significantly lower indicators
of vagal tone at baseline in visit 1 than non-responders:
HF power (p = 0.034), nSD1 (p = 0.039) and BRS (p =
0.036, see Supplementary Table 8). In addition,
responders had significantly higher sympathetic
prevalence during visit 1 baseline than non-responders:
SD2 (p = 0.028) and nSD2 (p = 0.036), and
significantly lower overall variability in HR during visit
1 baseline compared to non-responders: SDRR (p =
0.038) and S (p = 0.039). There were no significant
differences in demographic, health-related QoL, mood
or sleep characteristics between responders and non-
responders (p > 0.05).
Interestingly, two-thirds (n = 6) of responders
encountered a lowering of their baseline LF/HF ratio
after two weeks of daily tVNS (see Figure 6, visit 1
Figure 5. Baseline LF/HF ratio in visits 1 (A) and 2 (B) significantly predicted change in LF/HF ratio between baseline and tVNS. Δ
refers to the differences between baseline and tVNS.
Figure 6. Baseline LF/HF ratio reduced after 2 weeks of daily
tVNS in six responders with three showing an increase in
baseline LF/HF ratio following the daily tVNS (indicated by the
www.aging-us.com 4841 AGING
mean: 2.68, SEM: 0.87; visit 2 mean: 0.79, SEM: 0.29, p
= 0.028). However, there were three responders whose
baseline LF/HF ratio increased between visit 1 (mean:
1.58, SEM: 0.54) and visit 2 (mean: 2.69, SEM: 0.95,
see Figure 6), although all still responded in visit 2.
For the group of responders in visit 1 (n = 9), LF/HF
ratio was significantly lower during tVNS (0.95 ± 0.18)
compared to baseline (2.32 ± 0.59, p < 0.001) and
recovery (1.46 ± 0.27, p = 0.029, Friedman’s test: p =
0.001, see Figure 7a). But, in visit 2, LF/HF ratio did
not significantly differ between baseline, tVNS and
recovery for the whole group of responders (p > 0.05,
see Figure 7b), possibly due to slight decreases in
baseline LF/HF ratio.
Some non-responders (n = 7) even encountered a
lowering of their baseline LF/HF ratio after two weeks
of daily tVNS (visit 1 mean: 1.26, SEM: 0.39; visit 2
mean: 0.63, SEM: 0.16, p = 0.018, see Figure 8). Seven
of these participants then became responders to tVNS in
visit 2. In contrast, some non-responders (n = 10)
showed increases in baseline LF/HF ratio after two
weeks of daily tVNS (visit 1 mean: 1.30, SEM: 0.35;
visit 2 mean: 2.61, SEM: 0.80, p = 0.009) all of whom
remained non-responders at visit 2.
For the group of visit 1 non-responders (n = 17), LF/HF
ratio did not significantly differ between baseline, tVNS
or recovery in visits 1 or 2 (p > 0.05, see Figure 9),
probably due to low initial values.
Daily tVNS also improved health-related QoL and
Aspects of health-related QoL and mood improved
following two weeks of daily tVNS (see Supplementary
Table 9). For health-related QoL, scores for role
Figure 7. LF/HF ratio values during visit 1 (A) and visit 2 (B) for each responder during baseline, tVNS and recovery. Dashed black line
indicates the group mean.
Figure 8. Baseline LF/HF ratio reduced after 2 weeks of daily
tVNS in seven non-responders with ten showing an increase in
baseline LF/HF ratio following the daily tVNS (indicated by the
www.aging-us.com 4842 AGING
limitations due to physical health significantly
decreased between visits 1 and 2 (p = 0.026).
Additionally, there was a trend for energy scores to
increase following the two weeks of daily tVNS (p =
0.058). For mood, self-reported tension (p = 0.015),
depression (p = 0.035), vigour (p = 0.030) and mood
disturbance (p = 0.006) all significantly improved
between the two visits. There was also a trend for self-
reported confusion to be lower at visit 2 than at visit 1
(p = 0.059).
Linear regressions revealed that SF-36 energy scores
during visit 1 significantly predicted change during visit
2 (R2 = 0.280, p = 0.029): those with low energy scores
at visit 1 reported greater increases at visit 2 (see Figure
Furthermore, visit 1 POMS scores for tension (R2 =
0.307, p = 0.014), depression (R2 = 0.475, p = 0.001),
anger (R2 = 0.490, p = 0.001) and confusion (R2 =
0.560, p < 0.001) significantly predicted scores at visit
2: participants with high tension, depression, anger and
confusion scores at visit 1 reported greater
improvements at visit 2 than those with low scores (see
Ease of falling asleep (R2 = 0.265, p = 0.024), how
quickly it took participants to fall asleep (R2 = 0.380, p
= 0.005), quality of sleep (R2 = 0.223, p = 0.041) and
ease of waking up (R2 = 0.264, p = 0.024) in visit 1,
significantly predicted change between the two visits.
As depicted in Figure 12, those who experienced greater
improvements in ease of falling asleep, time taken to
fall asleep, quality of sleep and ease of waking up at
visit 2 had lower values at visit 1.
Transcutaneous vagal nerve stimulation (tVNS) acutely
administered to the tragus in healthy volunteers aged ≥
Figure 10. LF/HF ratio values during visit 1 (A) and visit 2 (B) for each non-responder during baseline, tVNS and recovery. Dashed
black line indicates the group mean.
Figure 9. Visit 1 SF-36 energy score significantly predicted
change at visit 2.
www.aging-us.com 4843 AGING
Figure 12. Visit 1 tension (A), depression (B), anger (C) and confusion (D) scores significantly predicted change at visit 2.
Figure 11. Visit 1 ease of falling asleep (A), time taken to fall sleep (B), sleep quality (C) and ease of waking up (D) significantly
predicted change at visit 2.
www.aging-us.com 4844 AGING
55 years was associated with improvements in
spontaneous cardiac baroreflex sensitivity and HRV.
Significantly, tVNS administered for 15 minutes every
day for two weeks improved autonomic function and
may improve some aspects of health-related QoL, mood
and sleep. Intriguingly, findings consistently illustrated
that individuals who have greater baseline sympathetic
prevalence showed more pronounced shifts towards
parasympathetic prevalence with tVNS, with daily
tVNS for 2 weeks improving autonomic balance in
some individuals. This means that tVNS administered
daily for 2 weeks may help attenuate some of the
autonomic and psychological changes that occur in
ageing, expanding the period of healthy ageing.
tVNS effects on autonomic function
Positive effects of tVNS on autonomic function have
been reported in non-patient groups. For instance, tVNS
administered to the tragus significantly reduced
sympathetic nerve activity in healthy participants 
and boosted measures of parasympathetic activity, for
example, spontaneous cardiac BRS , RMSSD ,
respiratory sinus arrhythmia (coupling between
respiration and RR interval, ) and high frequency
power spectrum . Despite this evidence, there is
little work examining the autonomic implications of
administering tVNS in healthy older individuals who
are undergoing age-associated shifts towards
sympathetic prevalence. This study therefore examined
the autonomic implications of tVNS in individuals aged
≥ 55 years. We anticipated that a single session of tVNS
would temporarily improve measures of autonomic
function. We also hypothesized that daily
administration of tVNS for two weeks would further
enhance autonomic tone.
Consistent with , a single session of tVNS was
associated with greater increases in BRS compared to
sham. Furthermore, tVNS promoted shifts towards
parasympathetic prevalence in a larger cohort of
individuals aged ≥ 55 years with daily tVNS improving
resting vagal tone in some participants. BRS is a
predictor of all-cause mortality : depressed BRS (<
3 ms/mmHg) was associated with reduced survival rates
from cardiovascular death and all-cause mortality. This
means that daily tVNS may be an effective tool for
strengthening the relationship between changes in RR
interval and SBP in individuals who have attenuated
BRS, potentially reducing mortality and increasing life
Baseline LF/HF ratio (considered a measure of
sympathovagal, or autonomic, balance) was a
significant predictor of response to tVNS in all three
studies: higher LF/HF ratio (greater sympathetic
prevalence) was associated with greater decreases in
LF/HF ratio (shifts towards parasympathetic
prevalence). These patterns emerged in previous work
, providing further support to potentially using
LF/HF ratio as a tool for screening individuals who are
likely to encounter greater autonomic benefits from
tVNS. Such predictions could enable selection of
optimal individuals for tVNS and may be particularly
important considering the number of conditions that are
characterized with enhanced sympathetic
prevalence/autonomic imbalance, e.g. cardiovascular,
pain, inflammatory and mental health conditions.
Indeed, this would aid with developing a valid inclusion
criteria for tVNS studies. Such criteria will need further
development - in this study, baseline LF/HF ratios
around 1.5 were associated with minimal changes (Δ =
0) whereas those with baseline values > 1.5 were
associated with decreases following tVNS.
Interestingly, those who had low baseline LF/HF ratios
(< 1.5) showed increases following tVNS, presumably
due to already low sympathetic and high
parasympathetic prevalence. It may therefore be
possible to determine baseline LF/HF ratio thresholds
which differentiate between response types. However,
such thresholds may need to be ascertained with regard
to recording and analysis conditions, since HRV can be
determined using different algorithms with potentially
variable outcomes regarding exact values.
It is particularly interesting that the LF/HF ratio linear
regressions reached statistical significance in study 3
where the regression was stronger following two weeks
of daily tVNS. Indeed, given the interaction effect
between visit and condition for ∆ RR and a stronger
linear regression for visit 2, perhaps daily tVNS for two
weeks induced some kind of training effect.
Furthermore, as some participants showed reductions in
baseline LF/HF ratio between the two visits, it seems
that daily use of tVNS confers autonomic benefits for
some individuals. Indeed, benefits of tVNS delivered
over long-time periods have been illustrated elsewhere,
although not with regards to autonomic function. For
instance, electrical stimulation of ABVN termination
sites for 15 minutes per day for two weeks in patients
with coronary artery disease decreased the need for
vasodilator medication and improved exercise tolerance
[21, 22]. In addition, in patients with paroxysmal atrial
fibrillation (PAF) who received tragus stimulation
(active) or ear lobe stimulation (sham) for 1 hour/day
for 6 months, there was a 75-85% decrease in AF
burden in the active compared to the sham stimulation
. Therefore, study 3 provides novel and timely data
showcasing that daily tVNS can have profound
autonomic benefits in individuals aged ≥ 55 years. Of
course, future studies should explore the reproducibility
of using baseline LF/HF ratio to determine response
www.aging-us.com 4845 AGING
type to tVNS. Additionally, it would be of interest to
explore the autonomic effects of more long-term use of
tVNS and optimum stimulation dosage.
tVNS effects on quality of life, mood and sleep
As well as conferring autonomic benefits, tVNS also
appears to have positive effects on psychological health
in patient groups. For instance, tVNS applied for two
weeks (once or twice for 15 minutes per day, ) or
for one month (twice for 30 minutes per day, ) in
patients with depression significantly reduced
depression scores. Furthermore, tVNS administered for
four weeks significantly improved QoL and depression
scores in patients with persistent postural-perceptual
Consistent with the above studies on psychological
health, we found that daily tVNS for two weeks
significantly improved measures of QoL, in particular,
role limitations due to physical health. Furthermore,
dimensions of mood including depression, tension,
vigour and mood disturbance were all improved
following two weeks of daily tVNS. These findings
therefore suggest that daily tVNS may be an effective
means of improving aspects of everyday life in this age
group. However, it should be acknowledged that since
this was a single-arm study, it is possible that these self-
report measures could have been influenced by a
placebo effect. Therefore, to further explore the effect
of daily tVNS on subjective measures (such as QoL and
mood), future studies should embrace designs and
protocols which allow for an assessment of the
contribution of placebo effects to improvements in such
measures, perhaps via a significantly enlarged sample
size. This could also be combined with other measures
which assess performance on cognitive tasks. This
would be particularly interesting, given that cognitive
function declines in older age and studies have shown
that tVNS can boost divergent thinking , associative
memory in older individuals  and the recognition of
emotions in faces when presented with whole faces 
and just the eye region .
Sleep has also been shown to deteriorate with age 
and could therefore potentially benefit from daily tVNS.
Indeed, on returning for their second visit, some
participants commented on improvements in aspects of
their sleep. Indeed, participants who found that falling
asleep took a long time and was difficult, had low sleep
quality and had difficulties waking up in the morning,
showed great improvements following two weeks of
daily tVNS. This is an important finding and warrants
further investigation employing a control group along
with validated sleep questionnaires, such as the
Pittsburgh Sleep Quality Index and polysomnography.
Age-related conditions which could benefit from
Normal ageing is associated with increases in
sympathetic prevalence and/or decreases in vagal tone
and overall variability [1, 2]. These ageing effects
interact with gender, whereby the greater sympathetic
prevalence in young males compared to young females
disappears in older age . In addition to normal
ageing, shifts towards sympathetic prevalence may
contribute to age-related conditions, for example,
hypertension, heart failure and atrial fibrillation.
Evidence suggests that tVNS could play a role in
ameliorating these conditions.
Electrical stimulation of ABVN termination sites for
only 15 minutes per day for 14 days in patients with
coronary artery disease decreased the need for
vasodilator medication and improved exercise tolerance
[21, 22]. Furthermore, tVNS reduced atrial fibrillation
in patients with paroxysmal atrial fibrillation (PAF) and
reduced plasma levels of the inflammatory cytokine
TNFα , which is associated with chronic increases
in inflammation with ageing . Similar positive
effects were observed when low-level tragus stimulation
(LL-TS) was administered to canines . Indeed, LL-
TS successfully reversed an increase in neural activity
(in the anterior right ganglionated plexi) and a decrease
in the window of vulnerability and effective refractory
period that accompanied induced atrial fibrillation .
Therefore, tVNS administered to ABVN sites appears to
temporarily improve symptoms associated with age-
related conditions, such as PAF and coronary artery
In addition to age-associated cardiovascular
pathophysiology, tVNS may confer benefits in other
age-related conditions. Also, tVNS reduced pain ratings
in chronic pelvic pain due to endometriosis in females
 and during sustained application of painful heat
. These findings are noteworthy given that ageing is
associated with increases in pain . tVNS may even
be used as a non-invasive functional technique for the
early diagnosis of Alzheimer’s disease and other
neurodegenerative conditions  and aid with the
progression of obesity and type 2 diabetes [39-42].
Considering the ease of application and affordability of
tVNS there is significant potential in prolonging the
period of healthy ageing and attenuating symptoms
associated with age-related conditions. Of course,
considerable work exploring the optimum tVNS
stimulation parameters (current, pulse width, pulse
frequency), tVNS session duration (e.g. 15 minutes) and
chronic paradigm (e.g. once daily for two weeks) for
specific conditions/patient groups may be required.
www.aging-us.com 4846 AGING
Future studies should also carefully consider their study
design, especially with respect to control groups.
Indeed, it should be acknowledged that the single-arm
design of studies 2 and 3 is a limitation. However, this
design was adopted for two main reasons. Firstly, 
and study 1 revealed tVNS was associated with greater
autonomic balance compared to sham. Secondly, there
are challenges with establishing a valid control for
chronically administered tVNS. For instance, if the
sham arm comprised of altering the internal workings of
the machine so that it administered no electrical
impulses but the machine still ‘looked’ active,
participants might have realized they were receiving no
stimulation (given that tVNS is perceptible). In which
case, the sham would not have been able to explore the
contribution of a placebo effect to changes in measures
of autonomic function, health-related QoL, mood and
sleep. An alternative sham could have entailed
administering the stimulation to the earlobe. However,
the assertion that the earlobe is not vagally innervated is
based on one study  and there has recently been
controversy about the anatomical location of the
auricular vagus [44-46]. Due to these challenges, we
decided to implement a single-arm design for studies 2
and 3 where each participant acted as their own control.
Future work could implement a wash-out period, which
would also aid with ascertaining the time required for
the effect of daily tVNS to diminish. This future work
should also explore the extent to which the changes in
the self-report measures were due to tVNS, a placebo
effect or a participant bias.
For the first time, we have shown that age-related
autonomic, QoL, mood and sleep changes may be
improved with tVNS administered every day for two
weeks. Importantly, the findings point to the influence
of initial values in determining magnitude and direction
of change following tVNS: high initial sympathetic
prevalence, tension, depression, anger and confusion
and low energy and sleep quality were associated with
greater improvements. With further work, it may
therefore be possible to identify which individuals will
most benefit from daily tVNS in terms of their
autonomic function and overall well-being.
University of Leeds ethical approval was secured
(Ethics Reference: BIOSCI 13-025) and the study
conformed to the standards outlined in the Declaration
of Helsinki. Informed written consent was obtained
voluntarily by all research participants and their data
were anonymised and stored securely according to the
UK Data Protection Act (1998). Participants were
informed that they could withdraw from the experiment
at any time.
Participants were excluded if they had a history of
cardiovascular disease, epilepsy or episodes of frequent
fainting (syncope). Participants were asked to abstain
from caffeine, alcohol, nicotine and strenuous exercise
for a minimum of 12 hours prior to their visit. No
female participants who participated were taking
hormone replacement therapy (HRT).
For study 1, 14 healthy volunteers aged 55 years or over
were recruited (n = 9 males, mean age: 69.11 ± 1.52
years). An additional 37 healthy participants aged 55
years or over were recruited for study 2, resulting in a
total sample size of 51 participants (n = 24 males, mean
age: 65.20 ± 0.79 years). The daily tVNS study was
conducted on 29 participants (study 3, n = 11 males,
mean age range: 64.14 ± 0.89 years).
All study visits were carried out in a quiet, temperature
controlled (21 ± 2°C) human physiology study room at
the University of Leeds between the hours of 09.00 and
12.00. Participants reclined semi-supine on a couch for
the duration of each experiment.
Transcutaneous vagus nerve stimulation (tVNS)
tVNS was performed using a TENS machine (V-TENS
Plus, Body Clock Health Care Ltd, United Kingdom in
studies 1 and 2 and EMS7500 Roscoe Medical in study
3) with customised auricular electrode clips attached on
the inner and outer surface of the tragus of the ear
(Auricular Clips, Body Clock Health Care Ltd, UK).
Participants wore the electrode clips throughout all
three recordings (baseline, stimulation, recovery). tVNS
was applied continuously for 15 minutes with a pulse
width of 200 µs and pulse frequency of 30 Hz.
Amplitude was adjusted to the level of sensory
threshold (usually 2-4 mA) until the participants
reported a ‘pin-prick’ or ‘tingling’ sensation. The
stimulus was then turned down until the stimulus was
borderline perceptible and comfortable.
Study 1 procedure
Study 1 examined the extent to which tVNS and sham
stimulation impacted measures of autonomic function.
Participants attended a ‘tVNS visit’ and a ‘sham visit’.
Sham stimulation entailed positioning the surface
www.aging-us.com 4847 AGING
electrodes as per tVNS and informing participants that a
different set of stimulation parameters would be tested
involving a reduction in the current below the
participant’s level of sensory perception. The electrode
leads were then disconnected from the TENS machine
without the participant’s knowledge. The order of the
visits was randomized between participants. At the
beginning of each visit, they completed a basic health
questionnaire, physical activity level was assessed using
the Godin Leisure Time Exercise Questionnaire  and
height and weight obtained. Physiological equipment
that continuously recorded HR, blood pressure (BP) and
respiration were then attached and participants rested for
10 minutes before recordings commenced. Three sets of
recordings were obtained: 10 minute baseline, 15 minute
stimulation and 10 minute recovery. The order of the
recordings was identical for all participants. Figure 13
summarises the procedure employed in study 1.
Study 2 procedure
Study 2 explored the effects of acute tVNS (i.e. a single
15-minute session) on autonomic function in a larger
group of individuals aged ≥ 55 years. The procedure
was identical to that employed in study 1, with the
exception that participants were not required to attend
for a sham visit (rationale based on study 1 results).
Figure 14 summarises the study 2 procedure.
Figure 13. Procedure for study 1.
Figure 14. Procedure for study 2.
www.aging-us.com 4848 AGING
Study 3 procedure
Study 3 examined the effects of daily tVNS on
autonomic function, QoL, mood and sleep in 29 healthy
volunteers aged ≥ 55 years (Figure 15). The procedure
was similar to that employed in study 1, with the
exception that participants attended on a second
occasion, exactly two weeks following their first visit
for a second tVNS visit. In addition, participants
completed the SF-36, POMS questionnaire (Profile of
Mood States) and a sleep questionnaire at the beginning
of each visit. At the end of their first visit, participants
were provided with a tVNS machine to take home. They
were trained in how to use the device and instructed to
do a 15-minute session once daily for two weeks. A log
sheet was provided so that participants could record
device usage and any comments/observations following
In all three studies, HR, non-invasive BP and respiration
were continuously recorded during the three recordings
(baseline, tVNS, recovery). Non-invasive BP was
recorded for the primary purpose of deriving BRS. In
study 3, the SF-36 (measuring health-related QoL),
Profile of Mood States (POMS) questionnaire
(measuring mood) and sleep questionnaire (measuring
sleep quality, ease of falling asleep and waking up, and
sleep interruption) were completed by participants.
Frequency-domain, time-domain and non-linear HRV
and cardiac BRS were derived for the final five minutes
of each recording and health-related QoL, mood and
sleep scores were also calculated for study 3. For further
information on the measurements obtained, please see
the Methods Supplementary Section.
The final five minutes of each recording were analysed
offline in LabChart 8 (AD Instruments). Frequency-
domain, time-domain and non-linear HRV were derived
along with BRS. Change (Δ) between baseline and
stimulation was calculated in all three studies.
Frequency-domain HRV parameters included: the low
frequency (LF) component, detected at 0.04-0.15 Hz;
the high frequency (HF) component, detected at 0.15-
0.40 Hz; total power (0.04-0.40 Hz) and normalised
values for LF and HF power (nuLF and nuHF). The HF
component has been associated with parasympathetic
modulation of heart rate , whereas LF power
reflects both parasympathetic and sympathetic
Figure 15. Procedure for study 3.
www.aging-us.com 4849 AGING
modulation of heart rate . The ratio of LF to HF
power (LF/HF) was also calculated along with
normalised LF/HF (nuLF/HF) where baseline values
were set to 1. A decrease in LF/HF ratio suggests a shift
in cardiac autonomic input towards either reduced
sympathetic activity and/or increased vagal
Five time domain HRV variables were analysed: mean
RR interval, Δ RR (difference between the longest and
shortest RR intervals), SDRR (standard deviation of all
RR intervals), RMSSD (the square root of the squared
of differences between adjacent RR intervals) and
pRR50 (percentage of number of pairs of adjacent RR
intervals differing by more than 50 ms). SDRR reflects
global autonomic regulation and is an estimate of all
HRV . RMSSD and pRR50, being mainly related to
beat-to-beat variations, reflect parasympathetic output
The Poincaré plot is a two-dimensional graphic
representation of the correlation between consecutive
RR intervals, in which each interval is plotted against
the following interval. For quantitative analysis, an
ellipse is fitted to the shape formed by the plot with the
center determined by the average RR intervals. Two
parameters were derived: SD1 and SD2. SD1 measures
the standard deviation of the distances of the points to
the diagonal y=x and SD2 measures the standard
deviation of the distances of points to the line y=-x +
RRm, where RRm is the average of RR intervals. SD1
is an index of instantaneous recording of the variability
of beat-to-beat and represents parasympathetic activity,
while SD2 represents the variability of long and short
term variability. SD1 and SD2 were normalised relative
to heart rate by computing: (SD1 or SD2/RR
interval)*1000. SD2/SD1 was also derived and
represented the ratio between short and long term
variations in RR intervals. The area of the ellipse, S,
representing total variability in RR intervals was
ascertained by calculating π*SD1*SD2.
Cardiac baroreflex sensitivity (BRS)
Cardiac BRS was assessed using the sequence method,
in which ‘up’ and ‘down’ sequences were identified
. ‘Up’ sequences consisted of three or more
consecutive cardiac cycles for which there was a
sequential rise in both SBP (≥ 1 mmHg) and RR
interval (≥ 2 ms). ‘Down’ sequences consisted of three
or more cardiac cycles for which there was a sequential
fall in SBP and RR interval.
In Excel 2013, the RR interval was plotted against SBP
for each sequence (r ≥ 0.85 acceptance level) and the
average slope values for the ‘up’ and ‘down’ sequences
were combined to get an average cardiac baroreflex
slope. Values of cardiac BRS were accepted when the
number of sequences was ≥3 for both up and down
Statistical analysis was performed using SPSS (version
25). Normality of distribution was tested using Shapiro-
Wilk. Two-tailed statistical tests were used in all
instances with an alpha level of 0.05. All data are
presented as group mean ± standard error of the mean
(SEM) unless otherwise stated.
The supplementary methods and materials section
provides further information about the statistical tests
performed. In brief, paired sample t-tests (or Wilcoxon
signed-rank tests) explored differences between tVNS
and sham (study 1) and following 14 days of tVNS
(study 3). Repeated measure ANOVAs examined
differences between recordings (study 2) and visits
(study 3, two-way repeated measure ANOVAs). Linear
regressions explored the extent to which baseline
autonomic function significantly predicted response to a
single session and daily use of tVNS.
ABVN: auricular branch of the vagus nerve; BMI: body
mass index; BP: blood pressure; BRS: baroreflex
sensitivity; ECG: electrocardiogram; HF: high
frequency power; HR: heart rate; HRT: hormone
replacement therapy; HRV: heart rate variability; LF:
low frequency power; LF/HF: ratio of low frequency to
high frequency power; LL-TS: low-level tragus
stimulation; nSD1: SD1 normalised relative to heart
rate; nSD2: SD2 normalised relative to heart rate;
nuHF: normalised high frequency power; nuLF:
normalised low frequency power; PAF: paroxysmal
atrial fibrillation; POMS: profile of mood states;
pRR50: percentage of number of pairs of adjacent RR
intervals differing by more than 50 ms; QoL: quality of
life; RR interval: interval between adjacent R peaks;
RMSSD: square root of the squared of differences
between adjacent RR intervals; S: area of the ellipse of
the Poincaré plot; SBP: systolic blood pressure; SD1:
standard deviation of points perpendicular to the axis of
line-of-identity; SD2: standard deviation of points along
the axis of line-of-identity; SD2/SD1: ratio of SD2 to
SD1; SDRR: standard deviation of all RR intervals;
SEM: standard error of the mean; SF-36: questionnaire
providing a measure of health-related quality of life;
TENS: transcutaneous electrical nerve stimulation;
tVNS: transcutaneous vagal nerve stimulation; VNS:
vagal nerve stimulation.
www.aging-us.com 4850 AGING
SD, JD and JC conceived the study and the whole team
contributed to securing ethical approval. AM conducted
study 1, BB, LA and AM conducted study 2 and BB
and LA completed study 3. BB performed the statistical
analysis with support from SD and JD. BB wrote the
initial manuscript draft with input from JD and SD,
followed by feedback from other authors.
We would like to thank all of the volunteers who took
part in this project.
CONFLICTS OF INTEREST
The authors report no conflict of interest.
We would like to thank the Dunhill Medical Trust for
funding this project, grant number DMT R469/0216.
1. Kuo TB, Lin T, Yang CC, Li CL, Chen CF, Chou P. Effect
of aging on gender differences in neural control of
heart rate. Am J Physiol. 1999; 277:H2233–39.
2. Stein PK, Freedland KE, Skala JA, Carney RM, Davila-
Roman V, Rich MW, Kleiger RE. Heart rate variability
is independent of age, gender, and race in
congestive heart failure with a recent acute
exacerbation. Am J Cardiol. 1997; 79:511–12.
3. Bibevski S, Dunlap ME. Evidence for impaired vagus
nerve activity in heart failure. Heart Fail Rev. 2011;
4. Kearney MT, Zaman A, Eckberg DL, Lee AJ, Fox KA,
Shah AM, Prescott RJ, Shell WE, Charuvastra E,
Callahan TS, Brooksby WP, Wright DJ, Gall NP, Nolan
J. Cardiac size, autonomic function, and 5-year
follow-up of chronic heart failure patients with
severe prolongation of ventricular activation. J Card
Fail. 2003; 9:93–99.
5. Gerritsen J, Dekker JM, TenVoorde BJ, Kostense PJ,
Heine RJ, Bouter LM, Heethaar RM, Stehouwer CD.
Impaired autonomic function is associated with
increased mortality, especially in subjects with
diabetes, hypertension, or a history of cardiovascular
disease: the Hoorn Study. Diabetes Care. 2001;
6. Taylor CB. Depression, heart rate related variables
and cardiovascular disease. Int J Psychophysiol.
7. Barton DA, Dawood T, Lambert EA, Esler MD,
Haikerwal D, Brenchley C, Socratous F, Kaye DM,
Schlaich MP, Hickie I, Lambert GW. Sympathetic
activity in major depressive disorder: identifying
those at increased cardiac risk? J Hypertens. 2007;
8. George MS, Sackeim HA, Rush AJ, Marangell LB,
Nahas Z, Husain MM, Lisanby S, Burt T, Goldman J,
Ballenger JC. Vagus nerve stimulation: a new tool for
brain research and therapy. Biol Psychiatry. 2000;
9. Kreuzer PM, Landgrebe M, Husser O, Resch M,
Schecklmann M, Geisreiter F, Poeppl TB, Prasser SJ,
Hajak G, Langguth B. Transcutaneous vagus nerve
stimulation: retrospective assessment of cardiac
safety in a pilot study. Front Psychiatry. 2012; 3:70.
10. Murray AR, Atkinson L, Mahadi MK, Deuchars SA,
Deuchars J. The strange case of the ear and the
heart: the auricular vagus nerve and its influence on
cardiac control. Auton Neurosci. 2016; 199:48–53.
11. Clancy JA, Mary DA, Witte KK, Greenwood JP,
Deuchars SA, Deuchars J. Non-invasive vagus nerve
stimulation in healthy humans reduces sympathetic
nerve activity. Brain Stimulat. 2014; 7:871–77.
12. Antonino D, Teixeira AL, Maia-Lopes PM, Souza MC,
Sabino-Carvalho JL, Murray AR, Deuchars J, Vianna
LC. Non-invasive vagus nerve stimulation acutely
improves spontaneous cardiac baroreflex sensitivity
in healthy young men: A randomized placebo-
controlled trial. Brain Stimulat. 2017; 10:875–81.
www.aging-us.com 4851 AGING
13. Barantke M, Krauss T, Ortak J, Lieb W, Reppel M,
Burgdorf C, Pramstaller PP, Schunkert H,
Bonnemeier H. Effects of gender and aging on
differential autonomic responses to orthostatic
maneuvers. J Cardiovasc Electrophysiol. 2008;
14. Abhishekh HA, Nisarga P, Kisan R, Meghana A,
Chandran S, Trichur Raju, Sathyaprabha TN.
Influence of age and gender on autonomic
regulation of heart. J Clin Monit Comput. 2013;
15. Felber Dietrich D, Schindler C, Schwartz J,
Barthélémy JC, Tschopp JM, Roche F, von
Eckardstein A, Brändli O, Leuenberger P, Gold DR,
Gaspoz JM, Ackermann-Liebrich U, and SAPALDIA
Team. Heart rate variability in an ageing population
and its association with lifestyle and cardiovascular
risk factors: results of the SAPALDIA study. Europace.
16. Umetani K, Singer DH, McCraty R, Atkinson M.
Twenty-four hour time domain heart rate variability
and heart rate: relations to age and gender over nine
decades. J Am Coll Cardiol. 1998; 31:593–601.
17. Ylikoski J, Lehtimäki J, Pirvola U, Mäkitie A, Aarnisalo
A, Hyvärinen P, Ylikoski M. Non-invasive vagus nerve
stimulation reduces sympathetic preponderance in
patients with tinnitus. Acta Otolaryngol. 2017;
18. La Marca R, Nedeljkovic M, Yuan L, Maercker A,
Ehlert U. Effects of auricular electrical stimulation on
vagal activity in healthy men: evidence from a three-
armed randomized trial. Clin Sci (Lond). 2010;
19. Sclocco R, Garcia RG, Gabriel A, Kettner NW,
Napadow V, Barbieri R. Respiratory-gated auricular
vagal afferent nerve stimulation (RAVANS) effects on
autonomic outflow in hypertension. Conference
proceedings: Annual International Conference of the
IEEE Engineering in Medicine and Biology Society
IEEE Engineering in Medicine and Biology Society
Annual Conference. 2017; 2017:3130-3133.
20. De Ferrari GM, Sanzo A, Bertoletti A, Specchia G,
Vanoli E, Schwartz PJ. Baroreflex sensitivity predicts
long-term cardiovascular mortality after myocardial
infarction even in patients with preserved left
ventricular function. J Am Coll Cardiol. 2007;
21. Zamotrinsky A, Afanasiev S, Karpov RS, Cherniavsky
A. Effects of electrostimulation of the vagus afferent
endings in patients with coronary artery disease.
Coron Artery Dis. 1997; 8:551–57. PMID:9431484
22. Zamotrinsky AV, Kondratiev B, de Jong JW. Vagal
neurostimulation in patients with coronary artery
disease. Auton Neurosci. 2001; 88:109–16.
23. Stavrakis S, Stoner J, Humphrey MB, Scherlag BJ,
Jackman WM, Po SS. Transcutaneous electrical vagus
nerve stimulation to suppress atrial fibrillation
(TREAT AF): A randomized clinical trial. Conference
proceedings: Heart Rhythm Scientific Sessions. 2019.
24. Hein E, Nowak M, Kiess O, Biermann T, Bayerlein K,
Kornhuber J, Kraus T. Auricular transcutaneous
electrical nerve stimulation in depressed patients: a
randomized controlled pilot study. J Neural Transm
(Vienna). 2013; 120:821–27.
25. Fang J, Rong P, Hong Y, Fan Y, Liu J, Wang H, Zhang
G, Chen X, Shi S, Wang L, Liu R, Hwang J, Li Z, et al.
Transcutaneous vagus nerve stimulation modulates
default mode network in major depressive disorder.
Biol Psychiatry. 2016; 79:266–73.
26. Eren OE, Filippopulos F, Sönmez K, Möhwald K,
Straube A, Schöberl F. Non-invasive vagus nerve
stimulation significantly improves quality of life in
patients with persistent postural-perceptual
dizziness. J Neurol. 2018 (Suppl 1); 265:63–69.
27. Colzato LS, Ritter SM, Steenbergen L.
Transcutaneous vagus nerve stimulation (tVNS)
enhances divergent thinking. Neuropsychologia.
28. Jacobs HI, Riphagen JM, Razat CM, Wiese S, Sack AT.
Transcutaneous vagus nerve stimulation boosts
associative memory in older individuals. Neurobiol
Aging. 2015; 36:1860–67.
www.aging-us.com 4852 AGING
29. Sellaro R, de Gelder B, Finisguerra A, Colzato LS.
Transcutaneous vagus nerve stimulation (tVNS)
enhances recognition of emotions in faces but not
bodies. Cortex. 2018; 99:213–23.
30. Colzato LS, Sellaro R, Beste C. Darwin revisited: the
vagus nerve is a causal element in controlling
recognition of other’s emotions. Cortex. 2017;
31. Crowley K. Sleep and sleep disorders in older adults.
Neuropsychol Rev. 2011; 21:41–53.
32. Stavrakis S, Humphrey MB, Scherlag BJ, Hu Y,
Jackman WM, Nakagawa H, Lockwood D, Lazzara R,
Po SS. Low-level transcutaneous electrical vagus
nerve stimulation suppresses atrial fibrillation. J Am
Coll Cardiol. 2015; 65:867–75.
33. Guarner V, Rubio-Ruiz ME. Low-grade systemic
inflammation connects aging, metabolic syndrome
and cardiovascular disease. Interdiscip Top Gerontol.
34. Yu L, Scherlag BJ, Li S, Fan Y, Dyer J, Male S, Varma V,
Sha Y, Stavrakis S, Po SS. Low-level transcutaneous
electrical stimulation of the auricular branch of the
vagus nerve: a noninvasive approach to treat the
initial phase of atrial fibrillation. Heart Rhythm.
35. Napadow V, Edwards RR, Cahalan CM, Mensing G,
Greenbaum S, Valovska A, Li A, Kim J, Maeda Y, Park
K, Wasan AD. Evoked pain analgesia in chronic pelvic
pain patients using respiratory-gated auricular vagal
afferent nerve stimulation. Pain Med. 2012; 13:777–
36. Busch V, Zeman F, Heckel A, Menne F, Ellrich J,
Eichhammer P. The effect of transcutaneous vagus
nerve stimulation on pain perception--an
experimental study. Brain Stimulat. 2013; 6:202–09.
37. Plooij B, Scherder EJ, Eggermont LH. Physical
inactivity in aging and dementia: a review of its
relationship to pain. J Clin Nurs. 2012; 21:3002–08.
38. Fallgatter AJ, Ehlis AC, Ringel TM, Herrmann MJ. Age
effect on far field potentials from the brain stem
after transcutaneous vagus nerve stimulation. Int J
Psychophysiol. 2005; 56:37–43.
39. Shiraishi T, Onoe M, Kojima T, Sameshima Y,
Kageyama T. Effects of auricular stimulation on
feeding-related hypothalamic neuronal activity in
normal and obese rats. Brain Res Bull. 1995; 36:141–
40. Wang S, Zhai X, Li S, McCabe MF, Wang X, Rong P.
Transcutaneous vagus nerve stimulation induces
tidal melatonin secretion and has an antidiabetic
effect in Zucker fatty rats. PLoS One. 2015;
41. Li H, Zhang JB, Xu C, Tang QQ, Shen WX, Zhou JZ,
Chen JD, Wang YP. Effects and mechanisms of
auricular vagus nerve stimulation on high-fat-diet--
induced obese rats. Nutrition. 2015; 31:1416–22.
42. Ju Y, Zhang H, Chen M, Chi X, Lan W, Zhang H, Mo J,
Yung NY. Effects of auricular stimulation in the
cavum conchae on glucometabolism in patients with
type 2 diabetes mellitus. Complement Ther Med.
43. Peuker ET, Filler TJ. The nerve supply of the human
auricle. Clin Anat. 2002; 15:35–37.
44. Mahadi KM, Lall VK, Deuchars SA, Deuchars J.
Cardiovascular autonomic effects of transcutaneous
auricular nerve stimulation via the tragus in the rat
involve spinal cervical sensory afferent pathways.
Brain Stimulat. 2019S1935-861X(19)30208-6.
45. Badran BW, Brown JC, Dowdle LT, Mithoefer OJ,
LaBate NT, Coatsworth J, DeVries WH, Austelle CW,
McTeague LM, Yu A, Bikson M, Jenkins DD, George
MS. Tragus or cymba conchae? Investigating the
anatomical foundation of transcutaneous auricular
www.aging-us.com 4853 AGING
vagus nerve stimulation (taVNS). Brain Stimulat.
46. Burger AM, Verkuil B. Transcutaneous nerve
stimulation via the tragus: are we really stimulating
the vagus nerve? Brain Stimulat. 2018; 11:945–46.
47. Godin G, Shephard RJ. A simple method to assess
exercise behavior in the community. Can J Appl
Sport Sci. 1985; 10:141–46. PMID:4053261
48. Chapleau MW, Sabharwal R. Methods of assessing
vagus nerve activity and reflexes. Heart Fail Rev.
2011; 16:109–27. https://doi.org/10.1007/s10741-
49. Akselrod S, Gordon D, Ubel FA, Shannon DC, Berger
AC, Cohen RJ. Power spectrum analysis of heart rate
fluctuation: a quantitative probe of beat-to-beat
cardiovascular control. Science. 1981; 213:220–22.
50. Camm AJ, Malik M, Bigger JT, Breithardt G, Cerutti S,
Cohen RJ, Coumel P, Fallen EL, Kennedy HL, Kleiger
RE, Lombardi F, Malliani A, Moss AJ, et al, and Task
Force of the European Society of Cardiology and the
North American Society of Pacing and
Electrophysiology. Heart rate variability: standards of
measurement, physiological interpretation and
clinical use. Circulation. 1996; 93:1043–65.
51. Parati G, Di Rienzo M, Bertinieri G, Pomidossi G,
Casadei R, Groppelli A, Pedotti A, Zanchetti A,
Mancia G. Evaluation of the baroreceptor-heart rate
reflex by 24-hour intra-arterial blood pressure
monitoring in humans. Hypertension. 1988; 12:214–
www.aging-us.com 4854 AGING
Heart rate was derived from electrocardiogram (ECG)
recorded by 3 Ag-AgCl surface electrodes on the chest
sampled at 2 kHz and continuous blood pressure was
recorded via finger pulse plethysmography (Finometer;
Finapres Medical System, Netherlands) sampled at 400
Hz in LabChart 8 (AD Instruments). Continuous blood
pressure recordings were obtained in order to derive
cardiac baroreflex sensitivity (BRS). Respiration was
recorded via a strain-gauge transducer (Pneumotrace,
UFI, CA, USA) placed around the chest and sampled at
0.4 kHz (in LabChart 8). Participants were allowed to
breathe spontaneously but their respiration rate monitored
to ensure there were no major deviations (> ±2 breaths
per minute from baseline value) and a minimum
respiration rate of ≥ 10 breaths per minute was met. One
participant in studies 2 and 3 had a respiration rate < 10
breaths per minute and was subsequently excluded from
all further analyses.
In study 3, two questionnaires were completed by
participants at the beginning of each visit: SF-36 and
Profile of Mood States (POMS). The SF-36 provides an
indication of health-related QoL from which scores on
different aspects of QoL can be derived e.g. physical
functioning, role limitations due to physical and
emotional health, energy, emotional well-being, social
functioning, pain, and general health scores. The POMS
questionnaire was used to gain an indication of how
specific aspects of mood may change as a result of the
fortnight of daily tVNS e.g. depression, tension, anger,
fatigue, confusion, vigour and mood disturbance).
To assess the extent to which autonomic function (HRV
and BRS) was significantly influenced by tVNS and sham
stimulation, paired sample t-tests (or Wilcoxon signed-
rank tests) were performed on change (Δ) between
baseline and stimulation. Linear regressions examined
whether baseline measures of autonomic function
significantly predicted response (change (Δ) between
baseline and stimulation).
To explore the extent to which tVNS significantly
influenced measures of autonomic function in a larger
sample of individuals aged ≥ 55 years, one-way repeated
measure ANOVAS with Bonferroni pairwise
comparisons were conducted. For non-normally
distributed data, Friedman tests with Wilcoxon signed-
rank post-hoc tests were performed. Linear regressions
explored the extent to which baseline measures of
autonomic function significantly predicted response (Δ)
to tVNS. To explore differences in baseline measures of
autonomic function, independent sample t-tests (or Mann-
Whitney U tests) were performed.
Paired sample t-tests (or Wilcoxon signed-rank tests)
examined the extent to which baseline autonomic tone,
health-related QoL, mood and sleep changed following
the fortnight of daily home-use tVNS. Two-way repeated
measure ANOVAs explored differences in measures of
autonomic function between the two visits and between
the three conditions. Linear regressions assessed whether
baseline autonomic tone, health-related QoL, mood and
sleep in visit 1 predicted change to visit 2 baseline to
tVNS. Independent sample t-tests (or Mann-Whitney U
tests) explored differences between responders and non-
www.aging-us.com 4855 AGING
Supplementary Table 1. Summary of characteristics of the final sample of study 1.
Final sample size (n)
Gender (frequency of males)
Baseline SBP (mmHg)
Baseline DBP (mmHg)
Baseline MAP (mmHg)
Supplementary Table 2. Summary of characteristics of the final sample of study 2.
Final sample size (n)
Gender (frequency of males)
Baseline SBP (mmHg)
Baseline DBP (mmHg)
Baseline MAP (mmHg)
Supplementary Table 3. Statistically significant differences between the three recordings (baseline, tVNS,
recovery) transpired. Reported p-values are from the one-way ANOVAs/Friedman tests. * = significantly
different to baseline (post-hoc tests). Data presented as the mean ± 1 SEM.
Total power (ms2)
* 1726 (255)
* 2011 (259)
LF power (ms2)
* 391 (75)
* 460 (61)
HF po wer (ms2)
* 448 (85)
Mean RR interval (ms)
* 1001 (21)
* 1009 (21)
Δ RR (ms)
* 249 (15)
* 42 (3)
* 42 (3)
* 31 (3)
* 10 (2)
* 22 (2)
* 54 (4)
* 55 (3)
* 53 (3)
* 55 (3)
* 4586 (794)
* 4429 (660)
* 8.83 (0.67)
www.aging-us.com 4856 AGING
Supplementary Table 4. Summary of statistically significant differences in baseline measures of autonomic
function between responders and non-responders. Data presented as the mean ± 1 SEM.
Responders (n = 16)
Non-responders (n = 32)
HF po wer (ms2)
Supplementary Table 5. Demographic characteristics of responders and non-responders: there were no
statistically significant differences between responders and non-responders.
Responders (n = 16)
Non-responders (n = 32)
Supplementary Table 6. Measures of vagal tone were significantly higher during visit 2 compared to visit 1.
Reported p-values are from the significant main effects of visit. Data presented as the mean ± 1 SEM.
Total power (ms2)
HF po wer (ms2)
Supplementary Table 7. Measures of overall variability significantly differed between the three recordings
(baseline, tVNS, recovery). Reported p-values are from the significant main effects of recording. * = significantly
different to baseline. Data presented as the mean ± 1 SEM.
Total power (ms2)
* 2216 (395)
* 2169 (400)
LF power (ms2)
* 529 (109)
* 47 (5)
* 44 (4)
* 61 (6)
* 57 (5)
* 5876 (1244)
Mean RR interval (ms)
* 1011 (33)
* 1010 (33)
* 59 (5)
* 56 (4)
www.aging-us.com 4857 AGING
Supplementary Table 8. Summary of demographic characteristics and visit 1 baseline autonomic measures for
responders and non-responders. Data are presented as the mean ± 1 SEM.
Responders (n = 9)
Non-responders (n = 17)
Visit 1 baseline
HF po wer (ms2)
Supplementary Table 9. Summary of changes in health-related QoL and mood between visits 1 and 2. Data
presented as the mean ± 1 SEM.
Role limitations due to physical health score
Mood disturbance score