Skip to main content
PMC home page
PLOS ONE logoLink to PLOS ONE
. 2021 Feb 26;16(2):e0247653. doi: 10.1371/journal.pone.0247653

Cervical transcutaneous vagal neuromodulation in chronic pancreatitis patients with chronic pain: A randomised sham controlled clinical trial

Janusiya Anajan Muthulingam 1,2, Søren Schou Olesen 2,3, Tine Maria Hansen 1,2, Christina Brock 2,4, Asbjørn Mohr Drewes 2,3, Jens Brøndum Frøkjær 1,2,*
Editor: Julia Robinson5
PMCID: PMC7909707  PMID: 33635894

Abstract

Background & aims

Chronic abdominal pain is the primary symptom of chronic pancreatitis, but unfortunately it is difficult to treat. Vagal nerve stimulation studies have provided evidence of anti-nociceptive effect in several chronic pain conditions. We investigated the pain-relieving effects of transcutaneous vagal nerve stimulation in comparison to sham treatment in chronic pancreatitis patients.

Methods

We conducted a randomised double-blinded, sham-controlled, crossover trial in patients with chronic pancreatitis. Patients were randomly assigned to receive a two-week period of cervical transcutaneous vagal nerve stimulation using the gammaCore device followed by a two-week sham stimulation, or vice versa. We measured clinical and experimental endpoints before and after each treatment. The primary clinical endpoint was pain relief, documented in a pain diary using a visual analogue scale. Secondary clinical endpoints included Patients’ Global Impression of Change score, quality of life and Brief Pain Inventory questionnaire. Secondary experimental endpoints included cardiac vagal tone and heart rate.

Results

No differences in pain scores were seen in response to two weeks transcutaneous vagal nerve stimulation as compared to sham treatment (difference in average pain score (visual analogue scale): 0.17, 95%CI (-0.86;1.20), P = 0.7). Similarly, no differences were seen for secondary clinical endpoints, except from an increase in the appetite loss score (13.9, 95%CI (0.5:27.3), P = 0.04). However, improvements in maximum pain scores were seen for transcutaneous vagal nerve stimulation and sham treatments as compared to their respective baselines: vagal nerve stimulation (-1.3±1.7, 95%CI (-2.21:-0.42), P = 0.007), sham (-1.3±1.9, 95%CI (-2.28:-0.25), P = 0.018). Finally, heart rate was decreased after two weeks transcutaneous vagal nerve stimulation in comparison to sham treatment (-3.7 beats/min, 95%CI (-6.7:-0.6), P = 0.02).

Conclusion

In this sham-controlled crossover study, we found no evidence that two weeks transcutaneous vagal nerve stimulation induces pain relief in patients with chronic pancreatitis.

Trial registration number

The study is registered at NCT03357029; www.clinicaltrials.gov.

Introduction

Chronic abdominal pain is the primary symptom of chronic pancreatitis (CP) and seen in the majority of patients during the course of disease [1]. Recent studies have provided evidence that CP patients have abnormal pain processing in the central nervous system [2]. Hence, sustained pancreatic nociceptive afferent inputs can over time result in sensitization of peripheral and central nociceptive pathways characterized by neuronal hyper-responsiveness. This can potentially lead to a chronic state of pain independent of peripheral nociceptive input, and patients typically respond inadequate to therapies directed against pathophysiological changes of the pancreas [1].

Vagal nerve stimulation (VNS) is a potential new non-pharmacological treatment targeting the autonomic and central nociceptive pathways. Previous studies have demonstrated that VNS provided analgesic effects in several painful conditions including fibromyalgia, pelvic pain and headache [3–5]. That has increased awareness of the potential of non-invasive vagal nerve stimulation (nVNS), and several medical devices are marketed. A study from our group, in healthy volunteers, showed that electrical and physiological modulation of parasympathetic tone resulted in pain attenuation [6], indicating that the parasympathetic nervous system is broadly anti-nociceptive [7]. nVNS has also shown its effect in clinical settings. Hence, two sham-controlled randomised controlled trials and six single-arm studies, demonstrated both rapid pain relief and a long term prophylactic effect of nVNS when used adjunctively with standard treatment in episodic and chronic migraine headaches [8]. The anatomical projections of the vagal nerve and its connections with many different brain areas involved in pain perception enables a potential analgesic effect [5]. The vagal nerve has widespread projections throughout the abdominal and pelvic viscera, thus a likely target of VNS could be abdominal pain in CP [9].

Based on the limited effectiveness of conventional treatments of CP pain, we hypothesized that two weeks nVNS could be effective as an adjuvant treatment to alleviate pain and improve quality of life in CP patients. The aims of this study were: 1) to primary evaluate the effects of two weeks nVNS on pain symptoms and quality of life by comparing nVNS treatment to sham treatment and 2) to investigate whether nVNS had an effect on modulating the autonomic nervous system. Finally, explorative analyses were provided comparing nVNS and sham treatment to their baselines.

Materials and methods

Study overview

This study was an investigator-initiated, randomised, double-blinded, sham-controlled, crossover clinical trial assessing the effect of cervical nVNS for pain treatment in patients with CP (Fig 1). The study was conducted at Aalborg University Hospital, Denmark from January 2018 to April 2019. As active nVNS treatment, an FDA approved medical device (gammaCore-S, ElectroCore LLC, Basking Ridge, New Jersey, USA) was used, and the sham devices were identical in appearance (ElectroCore LLC, Basking Ridge, New Jersey, USA). The North Denmark Region Committee on Health Research Ethics (N-20170023) and the Danish Medical Agency (2017023686) approved the study. Furthermore, the study was registered at Clinicaltrials.gov (NCT03357029). This study was conducted in accordance with the Declaration of Helsinki, and consistent with Good Clinical Practice. For more information on the study design, please refer to the study protocol which has been published in BMJ Open [10] (S1 Appendix CONSORT checklist and protocol).

Fig 1. CONSORT diagram.

Fig 1

Open in a new tab

Study enrolment and randomisation.

Inclusion of CP patients

Patients from the age of 18 years with a diagnosis of CP, based on the Mayo Clinic diagnostic criteria [11], were included regardless of the aetiology of CP. The included patients suffered from chronic abdominal pain more than 3 days per week lasting longer than 3 months, and they should consider their pain as insufficiently treated with usual analgesic treatments. All patients were required to read and understand Danish and to be able to comply with the scheduled visits and procedures. All participants signed and dated the informed consent and the Power of attorney document from the Danish Medical Agency [10]. Exclusion criteria were ongoing alcohol dependence, illegal drug dependencies, participating in other studies where an investigational drug was used. Patients with any clinically significant abnormalities that in the opinion of the investigator could increase the risk associated with trial participation were excluded. Moreover, patients with cardiovascular diseases, low blood pressure (<100/60), elevated intracranial pressure, female patients who were pregnant or lactating, contraindications for magnetic resonance imaging, previous surgery on the vagal nerve, and patients with known neuropathy were excluded. Patients were also excluded during the study, if they did not maintain the eligibility criteria during the study period [10]. All patients were asked to maintain their usual pain medication and dosage during the entire study, and they were only allowed to take extra pain medication in case of severe pain attacks. Any changes needed in pain medication during the study period was noted by the patient in the pain diary. Patients were allowed to continue the following analgesics during the study period: paracetamol, nonsteroidal anti-inflammatory drug (NSAID), opioids, tricyclic antidepressant (TCA), serotonin-norepinephrine reuptake inhibitor (SNRI) and gabapentoids.

Randomisation and blinding

Included patients were randomly assigned in two sequences (order of treatments: nVNS/sham or sham/nVNS) using block randomisation, allowing seven patients at the time to be randomised. The randomisation list was generated by an automatic web-based randomisation program. Patients and those administrating medical devices and assessing the outcomes were blinded to the group assignment.

Interventions

The active treatment (nVNS) was performed using the medical device gammaCore-S, which is commercially available and FDA-approved for the treatment of primary headache [8]. The sham treatment was performed using a sham device. The two devices were identical in appearance, weight, visual and audible feedback, but the sham device did not deliver electrical stimuli (only a vibration). Patients were instructed to self-administer the treatment bilaterally on the neck three times per day for two weeks at home. Two weeks stimulation was chosen, since previous prospective, double-blind, sham-controlled, randomized studies with transcutaneous VNS have demonstrated that stimulations with gammaCore induced a pain reliving effect at 30, 60 and 120 minutes in migraine patients [12]. Therefore, we expected that two weeks would be sufficient to induce a long-term pain-relieving effect. The treatment was delivered via two stainless steel contact surfaces that were covered with conductive gel before each treatment (Fig 2). Next, the device was positioned in parallel with the carotid artery on the neck. One stimulation had a duration of two minutes [13]. On the first day before treatment, patients were trained on correct positioning and instructed to adjust the stimulation intensity using +/–buttons on the device to achieve a comfortable and sufficient sensation. The sufficient sensation was gained, when the patient step-by-step increased the intensity of the stimulation, which induced mild pain and unpleasantness. Then, the intensity was decreased one step immediately. The intensity of the stimulation varied from 1 to 40 units, and patients registered the intensity of the stimulation after each treatment in the pain diary. The treatment periods were separated by a two-week wash-out period. The patients continued their usual analgesic medication during the wash-out period. The two week wash-out period has been used in previous studies of neuromodulation and shown to be sufficient to reset the effects of neuromodulation [10].

Fig 2. Study design & treatment procedure: Patients were instructed to self-administer the treatment bilaterally on the neck three times per day for two weeks at home.

Fig 2

Open in a new tab

The treatment was delivered via two stainless steel contact surfaces that were covered with conductive gel before each treatment.

Compliance was assessed by reading the remaining stimulations on the device display after each treatment period. Throughout both treatment periods, all patients were asked to register any adverse events using an adverse events questionnaire.

Clinical assessment of pain and quality of life

The primary clinical endpoint was to evaluate the effects of two weeks nVNS on pain symptoms and quality of life compared to two weeks of sham treatment. The daily experience of pain (the maximum pain intensity and the average pain intensity) was measured using a pain diary based on a visual analogue scale (VAS), with 0: no pain, and 10: worst pain imaginable [14]. All patients were instructed to fulfil the pain diary every day during the entire 8-week study period, starting with baseline measurements for seven consecutive days before the first treatment. One baseline measurement and the effect of two weeks treatment on pain symptoms were measured as an average of the pain scores for the last four days of the baseline and treatment periods, respectively.

The secondary clinical endpoints included three questionnaires; 1) the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30) [15], 2) Brief Pain Inventory (BPI) short form questionnaire [16], and 3) Patient’s Global Impression of Change (PGIC) [17]. EORTC QLQ-C30 and BPI questionnaires were fulfilled before and after each treatment period, while the PGIC score was assessed at the end of each treatment period.

Assessment of parasympathetic modulation effects

To investigate whether nVNS was able to modulate the cardiac vagal tone (CVT) in CP patients, the parasympathetic modulation was assessed using the electrocardiography (ECG) recorder device eMotion Faros 180 (Mega Electronics Ltd, Kuopio, Finland) [18]. First, the patients were instructed to rest for five minutes. Next, ECG electrodes (Ambu Blue Sensor P, Denmark) were placed in the left and right subclavicular areas and cardiac apex and then attached to a non-invasive cardiac monitor. A biosignal acquisition system (ExtremeBiometrics, London, UK) was used to assess cardiac derived parameters from five minutes recordings. Recordings were conducted before and after two weeks of nVNS/sham treatment. Similarly, recordings were also conducted after two weeks of nVNS/sham treatment.

Based on the ECG recording, CVT was assessed as a measure of the influence of the parasympathetic nervous system on the heart through the vagal nerve, which originates in the brain stem. CVT was calculated beat-to-beat based on detecting positive phase shifts in the R-R interval and used as a putative measurement of heart rate variability [18]. Thus, low CVT values reflect low heart rate variability, whereas high values reflect high heart rate variability. CVT was measured on a linear vagal scale (LVS) [18]. Before analysis, data were cleaned for noise, meaning that the first 10 samples of data were removed. Also, heart rate data points deviating from beat-to-beat with >15 beats pr. minute were removed, due to non-physiological causative factors. Since the consecutive R-R intervals affect the CVT measure, the previous and following 7 samples were also removed. Finally, heart rate was measured before and after both treatments.

Statistical analyses

The study was powered to detect a minimal difference in average daily pain scores of 30% between nVNS and sham treatment. Based on a standard deviation (SD) of 40%, we determined that a study with 16 patients in a crossover design was needed to provide a power of 80% with the use of a 2-sided significance level of 0.05 [10].

All data are presented as mean and SD unless otherwise stated. The continues variables were checked for normality using Shapiro-Wilk test. Before the primary analysis was performed, we defined the intensity of pain as an average of the scores for pain in the patient’s diary for the last 4 days in each treatment period [19]. For analysis of the primary endpoint, the difference in pain intensities between the two treatment sequences (nVNS/sham and sham/nVNS) were calculated for each patient and compared using a t-test for independent samples. The assumption that the wash-out phase was long enough to rule out a carry-over effect was checked in a pretest by calculating the sum of pain intensity scores for each subject and comparing these across the two sequence groups by means of a t-test for independent samples [20]. Changes in secondary continues endpoints (BPI, EORTC QLQ-C30, CVT data) were analyzed using a similar approach. Data on compliance was transformed to a relative scale (%) and analysed using a paired sample t-test. Categorical data (PGIC and adverse events) were analyzed using a McNemar test. Finally, one-sample t-tests were used to further analyse the difference for 1) nVNS vs. baseline, and 2) sham vs. baseline (within group analysis). All the statistical analyses were performed using statistical software package STATA V.16.1 (StataCorp LP, College Station). P-values <0.05 were considered statistically significant.

Results

Patient enrolment and clinical characteristics

Fifty-three patients were screened and 28 underwent randomisation (Fig 1). Twenty-two patients completed the baseline assessment. 17 patients completed the first treatment period and 16 patients completed both treatment periods. Demographic and clinical characteristics of the enrolled patients (n = 22) and for patients completing the study (n = 16) are summarized in Table 1. Characteristics for patients in each sequence (nVNS/sham, n = 4 and sham/nVNS, n = 12) are provided in Supporting information (S1 Table: Demographics by sequences). Five out of six dropouts were assigned to the nVNS/sham sequence. No patients dropped out due to adverse effects or non-tolerance of the interventions. In the nVNS treatment, 98%±3% of all stimulations were applied compared with 94%±11% in the sham treatment (P = 0.24). Furthermore, the average intensity of the stimulations was 31.8±6.9 units during nVNS treatment and 36.0±7.9 units for sham treatment (P = 0.02).

Table 1. Demographic and clinical characteristics of chronic pancreatitis patients.

Variables Chronic pancreatitis patients
Included in study (n = 22) Completed study (n = 16)
Age (years) 55.2±8.9 56.6±9.4
Male, n (%) 18 (81.8) 14 (88)
Body mass index (kg/m2) 22.5±4.6 21.9±3.6
Aetiology of chronic pancreatitis, n (%)
Alcohol 13 (35) 11 (42)
Nicotine 15 (41) 10 (38)
Nutritional 0 (0) 0 (0)
Hereditary factors 3 (8) 2 (8)
Efferent duct factors 5 (13) 3 (12)
Immunological factors 0 (0) 0 (0)
Miscellaneous and rare metabolic factors 1 (3) 0 (0)
Duration of chronic pancreatitis (years) 8.1±7.6 9.7±8.4
Diabetes, n (%) 11 (50) 9 (56)
Analgesics, n (%)
Paracetamol 14 (64) 11 (36)
NSAID 2 (9) 1 (3)
Opioids 19 (86) 13 (42)
Adjuvant analgesics, n (%) 1 (5) 1 (3)
TCA 0 (0) 0 (0)
SNRI Gabapentoids 7 (32) 5 (16)
Current alcohol use, n (%)
0 units 12 (55) 8 (50)
1–5 units 5 (23) 3 (19)
5–10 units 5 (23) 5 (31)
10–15 units 0 (0) 0 (0)
>20 units 0 (0) 0 (0)
Current smoking pattern, n (%)
non-smoker 5 (23) 4 (25)
1–5 cigarettes per day 3 (14) 2 (13)
5–10 cigarettes per day 2 (9) 2 (13)
10–15 cigarettes per day 3 (14) 2 (13)
15–20 cigarettes per day 6 (27) 3 (19)
>20 cigarettes per day 3 (14) 3 (19)
Open in a new tab

Note: Values are means ± SD. Percentages may not total 100 due to rounding.

Abbreviations: n = number of patients. NSAID = Nonsteroidal anti-inflammatory drug. TCA = Tricyclic antidepressant. SNRI = Serotonin-norepinephrine reuptake inhibitor.

Primary and secondary clinical outcomes

nVNS vs. sham

Primary and secondary clinical endpoints (after two weeks intervention) are summarized in Table 2. No significant differences were observed for the maximal or average pain score between two weeks nVNS treatment vs. sham treatment (difference in average pain score (visual analogue scale): 0.17, 95%CI (-0.86;1.20), P = 0.7) (Fig 3 and Supporting information (S1 Fig: Individual plots of the maximal and average pain scores)). Likewise, no differences were seen for secondary clinical endpoints, except from a significant increase in the appetite loss score with higher scores observed after nVNS (13.9, 95%CI (0.5:27.3), P = 0.04). PGIC showed no significant changes for nVNS treatment compared with sham treatment (Table 3, P = 0.63). The primary and secondary clinical endpoints with correction for the baseline assessments are given in Supporting information (S2 Table: The primary and secondary clinical endpoints corrected for baseline assessments), where no changes in pain scores were observed; however, a treatment effect of nVNS was shown with improvements in physical functioning (14.6, 95%CI (1.3:28.0), P = 0.03) and dyspnoea (-19.4, 95%CI (-37.3:1.6), P = 0.04).

Table 2. Primary and secondary clinical endpoints after two weeks of nVNS and sham treatment.
Primary clinical endpoints Pain diary
Variables Sequence Treatment period Mean±SD Within-individual differences for nVNS vs. sham Treatment effect for nVNS vs. sham (95% CI) P (effect) P (carry-over)
1 2
Average pain score (VAS) nVNS/sham 2.9±1.8 2.2±2.1 0.34±0.61 0.17 (-0.86:1.20) 0.72 0.49
sham/nVNS 3.6±2.5 3.3±2.4 -0.17±0.88
Maximal pain score (VAS) nVNS/sham 4.1±2.3 3.4±2.7 0.34±0.77 0.20 (-1.14:1.54) 0.76 0.56
sham/nVNS 4.6±2.6 4.4±2.5 -0.15±1.15
Secondary clinical endpoints Brief Pain Inventory questionnaire
BPI Pain nVNS/sham 3.1±2.1 2.8±2.5 0.1±0.4 0.19 (-1.11:1.48) 0.76 0.40
sham/nVNS 3.9±2.2 4.0±2.4 0.1±1.2
BPI Interference nVNS/sham 2.0±1.2 2.1±1.1 -0.1±0.2 -0.006 (-1.17:1.15) 0.99 0.13
sham/nVNS 3.8±2.1 3.9±2.6 0.0±1.0
Quality of life
Global health status nVNS/sham 64.6±19.7 66.7±22.6 -1.0±4.0 -1.04 (-13.1:11.0) 0.86 0.13
sham/nVNS 50.7±15.7 50.7±20.2 0.0±10.8
Physical functioning nVNS/sham 86.7±34.9 78.3±14.8 4.2±17.3 4.44 (-7.3:16.2) 0.43 0.04
sham/nVNS 59.4±18.1 60.0±17.3 0.3±5.8
Role functioning nVNS/sham 83.3±16.7 79.2±16.0 0.0±8.3 3.5 (-7.2:14.1) 0.49 0.08
sham/nVNS 50.0±24.6 56.9±28.8 3.5±7.5
Emotional functioning nVNS/sham 81.3±14.2 81.3±14.2 0.0±3.4 -1.7 (-13.5:10.0) 0.76 0.09
sham/nVNS 65.3±20.4 61.8±21.2 -1.7±10.6
Cognitive functioning nVNS/sham 83.3±19.2 87.5±16.0 -2.1±4.2 -2.1 (-12.7:8.5) 0.68 0.02
sham/nVNS 55.6±25.9 55.6±17.9 0.0±9.4
Social functioning nVNS/sham 83.3±19.2 87.5±16.0 -2.1±4.2 -4.2 (-16.2:7.9) 0.47 0.15
sham/nVNS 66.7±23.6 62.5±30.3 -2.1±10.7
Fatigue nVNS/sham 29.6±17.0 22.2±18.1 7.4±3.2 6.5 (-7.5:20.5) 0.34 0.16
sham/nVNS 45.4±20.9 43.5±30.9 -0.9±10.8
Nausea and vomiting nVNS/sham 0.0±0.0 0.0±0.0 0.0±0.0 1.4 (-8.0:10.8) 0.76 0.19
sham/nVNS 5.6±8.2 8.3±16.7 1.4±8.6
Pain nVNS/sham 45.8±25.0 33.3±13.6 6.2±8.0 4.2 (-13.1:21.4) 0.61 0.12
sham/nVNS 59.7±21.9 55.6±25.9 -2.1±15.1
Dyspnoea nVNS/sham 0.0±0.0 16.7±19.2 -5.6±9.6 -6.9 (-24.7:10.8) 0.41 0.28
sham/nVNS 22.2±25.9 19.4±26.4 -1.4±13.2
Insomnia nVNS/sham 11.1±19.2 16.7±33.3 5.6±9.6 6.9 (-25.7:39.5) 0.65 0.06
sham/nVNS 38.9±37.2 41.7±37.9 1.4±25.1
Appetite loss nVNS/sham 22.2±19.2 8.3±16.7 11.1±9.6 13.9 (0.5:27.3) 0.04 0.37
sham/nVNS 27.8±34.3 33.3±37.6 2.8±9.6
Constipation nVNS/sham 0.0±0.0 0.0±0.0 0.0±0.0 0.0 (-15.6:15.6) 1.00 0.29
sham/nVNS 13.9±30.0 13.9±26.4 0.0±14.2
Diarrhoea nVNS/sham 16.7±19.2 8.3±16.7 4.2±8.3 -2.8 (-15.9:10.4) 0.66 0.22
sham/nVNS 41.7±32.2 27.8±37.2 -6.9±11.1
Financial difficulties nVNS/sham 16.7±19.2 25.0±16.7 -4.2±8.3 0.0 (-12.3:12.3) 1.00 0.29
sham/nVNS 30.6±22.3 38.9±27.8 4.2±10.4
Secondary experimental endpoints Cardiac-derived parameters
Cardiac vagal tone (LVS) nVNS/sham 6.1±3.8 5.3±4.5 0.4±0.7 0.5 (-0.3:1.2) 0.18 0.48
sham/nVNS 4.6±1.9 4.7±1.9 0.1±0.5
Heart rate (beats/min) nVNS/sham 65.5±20.1 72.3±20.8 -3.4±2.8 -3.66 (-6.7:-0.6) 0.02 0.39
sham/nVNS 63.5±8.5 62.9±5.3 -0.3±2.4
Open in a new tab

Note: Comparisons between nVNS and sham treatments for the two treatment periods shown by treatment sequence (nVNS/sham, n = 4 and sham/nVNS, n = 12). The difference in pain intensities between the two treatment sequences were calculated for each patient and compared using a t-test for independent samples. One patient had missing data for role functioning, fatigue, dyspnoea, insomnia and appetite loss and one patient had missing data for cardiac vagal tone.

Abbreviations: VAS = visual analogue scale. SD = standard deviation. CI = confidence interval. nVNS = non-invasive vagal nerve stimulation. LVS = linear vagal scale. N = number of patients.

Fig 3.

Fig 3

Open in a new tab

Average pain score during the two study periods for the two sequences (nVNS/sham in black and sham/nVNS in grey). VAS = visual analogue scale.

Table 3. Patient’s Global Impression of Change (PGIC).
PGIC, n nVNS   P(exact)
  None Yes   0.63
Sham None 8 1 9
Yes 3 4 7
  11 5 16
Open in a new tab

Note: Distribution of improvement in PGIC induced by the nVNS and sham treatments (None: No change or worse; Yes: Improved). Abbreviations: n = number of patients.

nVNS and sham vs. baseline (within group analysis)

A significant reduction was found in the average pain scores (-0.7±1.2, 95%CI (-1.31:-0.003), P = 0.049) and the maximal pain score (-1.3±1.7, 95%CI (-2.21:-0.42), P = 0.007) comparing nVNS treatment with baseline. Also, significant reductions were detected in the maximal pain scores (-1.3±1.9, 95%CI (-2.28:-0.25), P = 0.018) and trend towards a reduction in the average pain scores (-0.6±1.2, 95%CI (-1.19:0.03), P = 0.06) for sham treatment as compared with baseline.

Secondary experimental outcomes

Cardiac-derived parameters are reported in Table 2. Compared to sham, nVNS treatment induced a decrease in heart rate (-3.7 BPM, 95% CI (-6.7:-0.6), P = 0.02), while no change in the CVT was observed (P = 0.18). Cardiac-derived parameters with correction for the baseline assessments showed no significant differences between the nVNS and sham treatments (Supporting information S2 Table: The primary and secondary clinical endpoints corrected for baseline assessments).

Adverse events

Five serious adverse events (SAEs) occurred during the study. One patient had a SAE with hospitalization due to acute pancreatitis in the nVNS period. Another patient had a SAE with death during the sham treatment, but this was not related to the intervention. Two patients had a SAE with hospitalization during the wash-out period (due to worsening of pain). During the follow-up period, a patient was admitted due to a skin rash. All SAEs were considered to be unrelated to the nVNS or sham treatment. There was no significant difference between the nVNS and sham treatment on any of the adverse events. Detailed information on adverse events are provided in Table 4.

Table 4. Adverse events.

Nausea, n nVNS   P(exact)
  None Yes   0.50
Sham None 1 0 1
Yes 2 13 15
  3 13 16
Constipation, n nVNS   P(exact)
  None Yes   1.00
Sham None 1 0 1
Yes 0 15 15
  1 15 16
Vomit, n   nVNS   P(exact)
  None Yes   0.50
Sham None 0 0 0
Yes 2 14 16
  2 14 16
Dizziness, n nVNS   P(exact)
  None Yes   0.63
Sham None 1 1 2
Yes 3 11 14
  4 12 16
Headache, n nVNS   P(exact)
  None Yes   1.00
Sham None 0 2 2
Yes 2 12 14
  2 14 16
Fatigue, n   nVNS   P(exact)
  None Yes   1.00
Sham None 3 1 4
Yes 2 10 12
  5 11 16
Abdominal pain, n nVNS   P(exact)
  None Yes   0.38
Sham None 1 1 2
Yes 4 10 14
  5 11 16
Diarrhoea, n nVNS   P(exact)
  None Yes   1.00
Sham None 1 1 2
Yes 2 12 14
  3 13 16
Dry mouth, n nVNS   P(exact)
  None Yes   1.00
Sham None 2 0 2
Yes 1 13 14
  3 13 16
Itching, n   nVNS   P(exact)
  None Yes   1.00
Sham None 0 1 1
Yes 0 15 15
  0 16 16
Other, n   nVNS   P(exact)
  None Yes   0.25
Sham None 0 0 0
Yes 3 13 16
  3 13 16
Open in a new tab

Note: Distribution of specific adverse events after nVNS and sham treatment (None: No adverse events; Yes: Adverse events noted).

Abbreviations: n = number of patients.

Discussion

To the best of our knowledge, this is the first clinical trial in CP patients evaluating the effects of nVNS on chronic abdominal pain. Overall, we found no evidence of significant difference in pain relief between the two treatments. A significant increase was demonstrated for appetite loss. However, both the nVNS and sham treatments induced significant improvements in pain scores when compared to their baselines. An enhanced parasympathetic tone was likely induced after two weeks nVNS, as reflected as a decrease in heart rate.

Effect of nVNS in patients with chronic pain

Previous clinical and pre-clinical studies of VNS provided strong evidence of analgesic effects in various clinical conditions including patients with fibromyalgia, pelvic pain and headaches [3]. In contrast to our hypothesis, we found no evidence that CP patients had a significant pain reduction comparing the nVNS and sham treatments. Interestingly, we found a reduction in pain when comparing nVNS treatment with baseline, and the same was found for the sham treatment. Previous clinical studies of pain treatment in CP patients have also shown an extensive placebo response [21] not only in pharmacological studies, but also in studies with surgical interventions, showing an improvement in subjective outcomes, including pain, disability and quality of life [22,23]. Similar pattern was observed in the current study, showing a potential sham effect in CP patients. A large pain reduction in the group receiving sham may mask the genuine efficacy of nVNS, and this may explain the non-significant results when comparing nVNS and sham treatment.

The lack of an anti-nociceptive effect following nVNS treatment, compared with sham treatment, could be explained by only partial enhancement of the parasympathetic nervous system to evoke anti-nociceptive effects as we only found an effect on the heart rate but not in CVT. The lack of pain improvement may also reflect that a longer-lasting treatment is needed, but the study could also be under-powered to detect a potential effect. Although, our findings are inconsistent with other VNS studies where sham control was not performed, it is important to emphasize that the influences of confounding covariates were reduced in our study as the patients serve as their own controls. As far as we are aware, the majority of the VNS studies were open labelled and non-crossover studies [24–27]. A study by Lange et al., provided evidence that VNS may be a useful adjunct treatment for patients with fibromyalgia [24], and Oshinsky et al. demonstrated that pain reduction was correlated with a reduction in the extracellular glutamate concentration in the brain [28]. Compared to these non-crossover studies, our current study found similar results, i.e. pain reduction comparing nVNS treatment with baseline. Our negative findings could therefore likely be explained by the strong study-design (double-blinded sham-controlled crossover study).

Recently, a functional MRI study showed that cervical vagal afferents can be accessed non-invasively via transcutaneous electrical stimulation of the neck, which overlies the course of the vagus nerve [29]. Also, the brain activation elicited by nVNS was analogous to that elicited by invasive VNS, providing validation of cervical nVNS effects [27].

Modulation of the parasympathetic tone with nVNS in CP patients

A previous study by Brock et al., demonstrated that nVNS induced an increase in CVT and reduced serum TNF-α in healthy volunteers, suggesting that nVNS exerts an autonomic and anti-inflammatory effect [30]. Thus, we expected that CVT would significantly increase in CP patients after two-week nVNS treatment as compared with sham treatment. However, we only found a decrease in heart rate after two weeks nVNS treatment as compared with sham, but no change in CVT. This indicates that nVNS might only be partially modulating the parasympathetic tone in CP patients. This could be explained by low sample size, as the study was powered for the primary endpoint (pain scores) and not for CVT. A study by Juel et al. in CP patients, showed that nVNS and deep slow breathing increased CVT compared to sham stimulation [6]. In this study, also no effects on pain were found. However, caution must be applied when comparing studies, as Juel et al. measured the acute effect of auricular VNS in combination with deep slow breathing implying that nVNS alone may not be adequate to modulate the vagal tone in CP patients.

Despite unsuccessful enhancement of mean CVT in our study, nVNS could potentially be able to module the parasympathetic tone in a sub-group of patients. A study conducted by Farmer et al., demonstrated that the 95% confidence interval of CVT was 1.9–17.8 LVS for healthy persons [18]. The average CVT in our study was in the lower part of the normal range, indicating that it is likely not possible to enhance the CVT in all patients. Despite one of the exclusion criteria was known peripheral neuropathy, some patients could have autonomic neuropathy. Hence, some patients could have vagal neuropathy, which can make it more difficult to modulate the parasympathetic tone and potentially mediating an anti-hyperalgesic effect [7]. A study by Liu et al. found that epilepsy patients having reduced heart rate viability compared to healthy controls, had significantly lower effect of VNS treatment (non-responders) [31]. Thus, the effect of nVNS could be related to the degree of autonomic dysfunction, and CVT may be useful as a screening tool in future studies. Accumulating evidence suggests that the autonomic nervous system plays an important role in the modulation of pain [7]. Considering that the parasympathetic nervous system acts broadly anti-nociceptive, while the sympathetic nervous system acts mainly pro-nociceptive, it is implicated that a balance between them is required for normal pain perception [32]. Thus, lack of an increase in the parasympathetic tone may lead to lack of prevention of visceral pain hypersensitivity in CP patients [7]. Additionally, an imbalance between the parasympathetic and sympathetic nervous systems may represent an important pathophysiological factor in CP patients with visceral pain. However, the latter should be investigated in detail in future studies.

Limitations

A number of study limitations need to be considered when evaluating our results. First, even though a sample size calculation was performed prior to this study, such calculations in clinical studies with experimental interventions are often based on some uncertain assumptions. Hence, we could have a too small sample with a risk of false negative results regarding our primary outcomes. Second, the paradigm and parameters of nVNS may influence the effect of the treatment including treatment duration, number of daily stimulations, stimulation intensity, unilateral vs. bilateral stimulation, etc. In this present study, the duration of the treatment was two weeks. There is no consensus regarding the duration of the treatment, and previous studies have used an intervention duration ranging from two weeks to six months in order to induce pain relieving effects [3,26–28,33]. The duration of treatment seems to differ depending on the type of disease, type of intervention (invasive VNS, auricular or cervical nVNS), and time frame (short-term vs. long-term effect) [3]. For example, in chronic pain patients, an auricular nVNS study applied the treatment for six weeks [34], while a cervical nVNS study applied the treatment for two weeks [12]. Given that CP is characterized by long-lasting ailment, future studies should possibly include a longer treatment duration to recover the functional plasticity in neural circuits of pain processing. In addition, the duration of the wash-out period of two weeks was chosen since it has been shown to be sufficient to reset the effects of neuromodulation [10], but long-term effects have also been demonstrated in patients with epilepsy [35]. However, if we extended the study duration (treatment and wash-out periods), a high risk of dropout would potentially occur, and study completion would become more difficult. Third, even though the sham device did not deliver electrical stimulation, it may not be physiologically inert as it provides some vibration and tactile sensation that potentially can stimulate the vagal nerve and induce an effect similar to the active treatment. This could also explain why there was no differences between the active and sham treatment, but a positive effect when comparing both the active and sham treatments to their respective baselines. Additionally, the order of treatment sequence was, despite block randomisation, unbalanced with more dropouts in the nVNS/sham sequence. It is unclear how this could have affected our findings, even though the statistical analysis approach takes the treatment sequence into account. Forth, in the current study, we did not analyse blood samples measuring inflammation. Although, no pain reductions were observed with two weeks nVNS, it would be interesting to assess whether the pancreatic inflammation was decreased. Fifth, even though the study was based on a cross-over design, where patients served as their own control, bias could be present due to the inclusion/exclusion criteria. For instance, ongoing alcohol dependence was an exclusion criteria whereas ongoing smoking was not, but nicotine could potentially impact the inflammatory-system [36]. Finally, various confounding factors are present in this patient cohort. For instance, the patients used several types of analgesics which may influence the effect of nVNS and as already discussed it may be difficult to modulate the parasympathetic tone for some patients. Furthermore, the impact of concomitant medication, the duration of CP or other disease characteristics (including stress, anxiety, and depression) could also be relevant to investigate. In future studies, a larger sample size is also required in order to investigate sub-groups of patients, who possible could have an effect of nVNS.

Conclusion

Our sham-controlled crossover study provided no evidence that adjuvant treatment with two weeks nVNS induces pain relief as compared to sham treatment in patients with painful chronic pancreatitis. Similarly, pain interference and quality of life scores were not improved in response to nVNS as compared with sham treatment. However, attention should be directed towards a potential pronounced sham effect of neuromodulation, since improvements were seen for both active and sham treatments as compared to baseline.

Supporting information

S1 Fig. Individual plots of the maximal and average pain scores.

(DOCX)

Click here for additional data file. (60.9KB, docx)
S1 Table. Demographics by sequences.

(DOCX)

Click here for additional data file. (17.6KB, docx)
S2 Table. The primary and secondary clinical endpoints corrected for baseline assessments.

(DOCX)

Click here for additional data file. (28.8KB, docx)
S1 Appendix. CONSORT checklist and protocol.

(PDF)

Click here for additional data file. (1.1MB, pdf)

Acknowledgments

Electrocore LLC provided the gammaCore devices but did not influence the design or reporting of the study. Statistician Niels Henrik Bruun, MSc, at Unit of Clinical Biostatistics, Aalborg University Hospital, Aalborg, Denmark is acknowledged for support with the statistical analyses of the primary and secondary endpoints. We thank radiographer Kenneth Krogh Jensen at Department of Radiology, Aalborg University Hospital, Aalborg Denmark for his assistance in data collection.

Abbreviations

BPI

Brief pain inventory

CP

Chronic pancreatitis

CVT

Cardiac vagal tone

LVS

Linear vagal scale

PGIC

Patients’ Global Impression of Change score

nVNS

non-invasive vagal nerve stimulation

VAS

Visual analogue scale

VNS

Vagal nerve stimulation

Data Availability

Data is available at https://doi.org/10.6084/m9.figshare.14009030.

Funding Statement

This study was supported by Independent Research Fund Denmark (DFF – 7016-00073), Axel-Muusfeldts Foundation, and Læge Sofus Carl Emil Friis og Hustru Olga Doris Friis’ Legat. Jens Brøndum Frøkjær received all the grants. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

  • 1.Singh VK, Yadav D, Garg PK. Diagnosis and Management of Chronic Pancreatitis: A Review. JAMA. 2019;322: 2422–2434. 10.1001/jama.2019.19411 [DOI] [PubMed] [Google Scholar]
  • 2.Kleeff J, Whitcomb DC, Shimosegawa T, Esposito I, Lerch MM, Gress T, et al. Chronic pancreatitis. Nat Rev Dis Prim. 2017;3: 1–18. 10.1038/nrdp.2017.60 [DOI] [PubMed] [Google Scholar]
  • 3.Chakravarthy K, Chaudhry H, Williams K, Christo PJ. Review of the Uses of Vagal Nerve Stimulation in Chronic Pain Management. Current Pain and Headache Reports. 2015. 10.1007/s11916-015-0528-6 [DOI] [PubMed] [Google Scholar]
  • 4.Garcia RG, Lin RL, Lee J, Kim J, Barbieri R, Sclocco R, et al. Modulation of brainstem activity and connectivity by respiratory-gated auricular vagal afferent nerve stimulation in migraine patients. Pain. 2017;158: 1461–1472. 10.1097/j.pain.0000000000000930 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Napadow V, Edwards RR, Cahalan CM, Mensing G, Greenbaum S, Valovska A, et al. Evoked Pain Analgesia in Chronic Pelvic Pain Patients Using Respiratory-Gated Auricular Vagal Afferent Nerve Stimulation. Pain Med (United States). 2012;13: 777–789. 10.1111/j.1526-4637.2012.01385.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Juel J, Brock C, Olesen SS, Madzak A, Farmer AD, Aziz Q, et al. Acute physiological and electrical accentuation of vagal tone has no effect on pain or gastrointestinal motility in chronic pancreatitis. J Pain Res. 2017;10: 1347–1355. 10.2147/JPR.S133438 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Botha C, Farmer AD, Nilsson M, Brock C, Gavrila AD, Drewes AM, et al. Preliminary report: modulation of parasympathetic nervous system tone influences oesophageal pain hypersensitivity. Gut. 2015;64: 611–7. 10.1136/gutjnl-2013-306698 [DOI] [PubMed] [Google Scholar]
  • 8.Mwamburi M, Tenaglia AT, Leibler EJ, Staats PS. Review of evidence on noninvasive vagus nerve stimulation for treatment of migraine: efficacy, safety, and implications. Am J Manag Care. 2018;24: S507–S516. [PubMed] [Google Scholar]
  • 9.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 Stimul. 2013;6: 202–209. 10.1016/j.brs.2012.04.006 [DOI] [PubMed] [Google Scholar]
  • 10.Muthulingam JA, Olesen SS, Hansen TM, Brock C, Drewes AM, Frøkjær JB. Study protocol for a randomised double-blinded, sham-controlled, prospective, cross-over clinical trial of vagal neuromodulation for pain treatment in patients with chronic pancreatitis. BMJ Open. 2019;9: e029546. 10.1136/bmjopen-2019-029546 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Layer P, Yamamoto H, Kalthoff L, Clain J, Bakken L, DiMagno E. The different courses of early- and late-onset idiopathic and alcoholic chronic pancreatitis. Gastroenterology. 1994;107: 1481–7. 10.1016/0016-5085(94)90553-3 [DOI] [PubMed] [Google Scholar]
  • 12.Tassorelli C, Grazzi L, Tommaso M De, Pierangeli G. Noninvasive vagus nerve stimulation as acute therapy for migraine The randomized PRESTO study. 2018;0. 10.1212/WNL.0000000000005857 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Use I, Risks P. Instructions for Use for gammaCore ® -S 1: 1–28. [Google Scholar]
  • 14.Jensen MP, Turner JA, Romano JM, Fisher LD. Comparative reliability and validity of chronic pain intensity measures. Pain. 1999;83: 157–162. 10.1016/s0304-3959(99)00101-3 [DOI] [PubMed] [Google Scholar]
  • 15.Aaronson NK, Ahmedzai S, Bergman B, et al. The European organization for research and treatment of cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. Journal of the National Cancer Institute. 1993. pp. 365–376. 10.1093/jnci/85.5.365 [DOI] [PubMed] [Google Scholar]
  • 16.Cleeland CS. Brief Pain Inventory (BPI). Cleel CS MD Anderson Cancer Cent. 1982;1100: 6. 10.1007/978-1-4419-9893-4_13 [DOI] [Google Scholar]
  • 17.Farrar JT, Young JP, LaMoreaux L, Werth JL, Poole RM. Clinical importance of changes in chronic pain intensity measured on an 11-point numerical pain rating scale. Pain. 2001;94: 149–158. 10.1016/S0304-3959(01)00349-9 [DOI] [PubMed] [Google Scholar]
  • 18.Farmer AD, Coen SJ, Kano M, Weltens N, Ly HG, Botha C, et al. Normal values and reproducibility of the real-time index of vagal tone in healthy humans: a multi-center study. Ann Gastroenterol. 2014;27: 362–368. [PMC free article] [PubMed] [Google Scholar]
  • 19.Gilron I, Bailey JM, Tu D, Holden RR, Weaver DF, Houlden RL. Morphine, Gabapentin, or Their Combination for Neuropathic Pain. N Engl J Med. 2005;352: 1324–1334. 10.1056/NEJMoa042580 [DOI] [PubMed] [Google Scholar]
  • 20.Wellek S, Blettner M. Vom richtigen Umgang mit dem Crossover-Design in klinischen Studien: Teil 18 der Serie zur Bewertung wissenschaftlicher Publikationen. Dtsch Arztebl Int. 2012;109: 276–281. 10.3238/arztebl.2012.0276 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Olesen SS, Bouwense S a W, Wilder-Smith OHG, van Goor H, Drewes AM. Pregabalin reduces pain in patients with chronic pancreatitis in a randomized, controlled trial. Gastroenterology. 2011;141: 536–43. 10.1053/j.gastro.2011.04.003 [DOI] [PubMed] [Google Scholar]
  • 22.Pham A, Forsmark C. Chronic pancreatitis: review and update of etiology, risk factors, and management. F1000Research. 2018;7: 607. 10.12688/f1000research.12852.1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Drewes AM, Kempeneers MA, Andersen DK, Arendt-Nielsen L, Besselink MG, Boermeester MA, et al. Controversies on the endoscopic and surgical management of pain in patients with chronic pancreatitis: pros and cons! Gut. 2019;68: 1343–1351. 10.1136/gutjnl-2019-318742 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Lange G, Janal MN, Maniker A, Fitzgibbons J, Fobler M, Cook D, et al. Safety and efficacy of vagus nerve stimulation in fibromyalgia: a phase I/II proof of concept trial. Pain Med. 2011;12: 1406–1413. 10.1111/j.1526-4637.2011.01203.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Gottfried-Blackmore A, Adler EP, Fernandez-Becker N, Clarke J, Habtezion A, Nguyen L. Open-label pilot study: Non-invasive vagal nerve stimulation improves symptoms and gastric emptying in patients with idiopathic gastroparesis. Neurogastroenterol Motil. 2019;32: e13769. 10.1111/nmo.13769 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Paulon E, Nastou D, Jaboli F, Marin J, Liebler E, Epstein O. Proof of concept: short-term non-invasive cervical vagus nerve stimulation in patients with drug-refractory gastroparesis. Frontline Gastroenterol. 2017;8: 325–330. 10.1136/flgastro-2017-100809 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Barbanti P, Grazzi L, Egeo G, Padovan AM, Liebler E, Bussone G. Non-invasive vagus nerve stimulation for acute treatment of high-frequency and chronic migraine: an open-label study. J Headache Pain. 2015;16. 10.1186/s10194-015-0542-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Oshinsky ML, Murphy AL, Hekierski HJ, Cooper M, Simon BJ. Noninvasive vagus nerve stimulation as treatment for trigeminal allodynia. Pain. 2014;155: 1037–1042. 10.1016/j.pain.2014.02.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Frangos E, Komisaruk BR. Access to Vagal Projections via Cutaneous Electrical Stimulation of the Neck: fMRI Evidence in Healthy Humans. Brain Stimul. 2017;10: 19–27. 10.1016/j.brs.2016.10.008 [DOI] [PubMed] [Google Scholar]
  • 30.Brock C, Brock B, Aziz Q, Møller HJ, Pfeiffer Jensen M, Drewes AM, et al. Transcutaneous cervical vagal nerve stimulation modulates cardiac vagal tone and tumor necrosis factor-alpha. Neurogastroenterol Motil. 2017;29: 4–7. 10.1111/nmo.12999 [DOI] [PubMed] [Google Scholar]
  • 31.Liu H, Yang Z, Huang L, Qu W, Hao H, Li L. Heart-rate variability indices as predictors of the response to vagus nerve stimulation in patients with drug-resistant epilepsy. Epilepsia. 2017;58: 1015–1022. 10.1111/epi.13738 [DOI] [PubMed] [Google Scholar]
  • 32.Matteoli G, Boeckxstaens GE. The vagal innervation of the gut and immune homeostasis. Gut. 2013;62: 1214–22. 10.1136/gutjnl-2012-302550 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Silberstein SD, Calhoun AH, Lipton RB, Grosberg BM, Cady RK, Dorlas S, et al. Chronic migraine headache prevention with noninvasive vagus nerve stimulation: The EVENT study. Neurology. 2016;87: 529–538. 10.1212/WNL.0000000000002918 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Sator-Katzenschlager SM, Scharbert G, Kozek-Langenecker SA, Szeles JC, Finster G, Schiesser AW, et al. The Short- and Long-Term Benefit in Chronic Low Back Pain Through Adjuvant Electrical Versus Manual Auricular Acupuncture. Anesth Analg. 2004. 10.1213/01.ane.0000107941.16173.f7 [DOI] [PubMed] [Google Scholar]
  • 35.Grazzi L, Tassorelli C, De Tommaso M, Pierangeli G, Martelletti P, Rainero I, et al. Practical and clinical utility of non-invasive vagus nerve stimulation (nVNS) for the acute treatment of migraine: A post hoc analysis of the randomized, sham-controlled, double-blind PRESTO trial. J Headache Pain. 2018;19: 1–9. 10.1186/s10194-017-0831-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Piao WH, Campagnolo D, Dayao C, Lukas RJ, Wu J, Shi FD. Nicotine and inflammatory neurological disorders. Acta Pharmacol Sin. 2009;30: 715–722. 10.1038/aps.2009.67 [DOI] [PMC free article] [PubMed] [Google Scholar]

Decision Letter 0

Susan Hepp

21 Jul 2020

PONE-D-20-13163

The effects of two weeks transcutaneous vagal neuromodulation in chronic pancreatitis patients: a randomised sham controlled clinical trial

PLOS ONE

Dear Dr. Frøkjær,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Specifically, the reviewers had multiple comments and concerns about methods reporting, methods justification, and would like several points addressed in the discussion.

Please submit your revised manuscript by Sep 04 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

  • A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

  • A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

  • An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Susan Hepp

Academic Editor

PLOS ONE

Journal Requirements:

Additional Editor Comments (if provided):

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The study describes effects of the cervical transcutaneous vagus nerve stimulation in chronic pancreatitis patients. Irrespective of negative findings, this study design and results are of high value for the interested reader.

Ad Title

„Transcutaneous vagal neuromodulation“ is too general, leading potentially to confusion with the auricular transcutaneous vagal neuromodulation. I would recommend to integrate either “cervical” or “on the neck”

Ad Abstract

Please add to the methodology section “cervical” or “on the neck”. Currently the reader of the abstract has no clue where the stimulation was performed.

Add Abbreviation

The abbreviation t-VNS is a bit misleading since tVNS is usually used for the transcutaneous auricular stimulation, just in contrast to pVNS as used for the percutaneous auricular stimulation. The stimulation on the neck is usually abbreviated as nVNS. It is a suggestion to avoid confusion of the readership.

Ad Methods

I think that 2 weeks stimulation is a relatively short period, especially in view of a long-lasting chronic ailment with the associate changes in plasticity. For instance, the auricular VNS is usually applied for 6 weeks in chronic ailments such as the chronic back pain (https://doi.org/10.1213/01.ane.0000107941.16173.f7). Even though it is shortly discussed in the limitations section, the reader would welcome a more detailed discussion on the effective treatment durations of VNS and cervical transcutaneous VNS (optionally, including also auricular VNS) in chronic ailments. Please discuss and justify this issue in detail.

Please amend whether state-of-the-art individual treatment of patients was optimized before the treatment, interrupted or continued during the tVNS study. Address concomitant medication/treatment of patients which is very important since improvements were registered with respect to the baseline.

Quantitative description of CVT is missing, even though a reference is given. Please describe it in a few sentences so that the reader has a clear idea about it.

More information is needed on the wash-out period, how long, 1 week? With uninterrupted concomitant treatment?

Ad Discussion

p.15 Discussed comparison/justification of cervical tVNS results with respect to invasive VNS results is limited due to the basically different technologies applied. Therefore, please address potential validation of cervical tVNS effects with respect to invasive VNS effects to substantiate your conclusions here.

Reviewer #2: In the present study, the authors used the antinociceptive effect of the vagus nerve (VN) to perform a two weeks non-invasive transcutaneous cervical VN stimulation (VNS) in patient with chronic pancreatitis (CP), in a sham-controlled crossover study. Unfortunately, VNS did not induce pain relief in patients with CP.

This is an interesting and original study as stated by the authors, since there are no clinical trials published on VNS in CP. I already had the opportunity to review the study protocol published in BMJ Open in 2019 by the same authors (ref. 10).

There are some points which need to be addressed:

- Abstract: specify that it was a cervical transcutaneous VNS. Specify the device (GammaCore).

- The number of patients is rather small for me, as specified by the authors.

- VNS stimulation is generally performed on the left VN while in the present study the authors stimulated both VN. Is-there any advantage to perform such a bilateral stimulation?

- The duration of the study (2 weeks of stimulation) is most likely too short since VNS is a slow acting therapy as demonstrated for epilepsy. Indeed, VNS induces modifications of neuronal plasticity, which are progressive. CP is not comparable to “headache” which is FDA approved with the device GammaCore. As stated by the authors CP induces central modifications which can induce a central sensitivity syndrome. In addition, most of these patients use opioid compounds which are also able to induce central perturbation of pain. So, it is simplistic to think that a two-week VNS will be able to improve such patients. In addition, if the effect of t-VNS is delayed over time, when switching to sham stimulation, it could induce a bias of the results of sham stimulation.

- In addition, the sham stimulation is rather questionable since it induces a vibration which, as noted by the authors, is able to have an effect on the auricular branch of the VN.

- The authors did not discuss the parameters of VNS.

- The duration of CP is 8.1 ± 7.6 years, so very variable. Thus, the central integration of pain and central sensitization is supposed to be patient-dependent and different between patients.

- The analgesic treatments may interfere with VNS (see above)

- Pain is multidimensional and rather subjective, so the use of a VAS is questionable, although I agree that VAS is often used in such conditions. It seems to me that there are no questionnaires evaluating stress (perceived), depression, anxiety, coping strategies in this trial.

- Did the patients used or have used complementary medicines which are able to impact the pain factor?

- A limit for me of non-invasive VNS by comparison to invasive one is the compliance. How the authors controlled that patients had regularly performed VNS (and bilaterally)?

- Regarding vagal tone, did the authors include patients with only low vagal tone which should be more appropriate and is supposed to better beneficiate of VNS. It is not clear for me in the manuscript.

- The authors did not explore the anti-inflammatory effect of VNS which is well known in different chronic inflammatory conditions such as rheumatoid arthritis and Crohn’s disease. That would have been interesting and there is a relation between inflammation and pain. This point is only stated in the “Limitations” of the study.

- Ongoing alcohol dependence was an exclusion criterion but not smoking? Nicotine is able to act on alpha7 nicotinic acetylcholine receptor and to have an anti-inflammatory effect.

Reviewer #3: *** General remarks: ***

The study follows a two period two treatment crossover design.

However, the analysis which was performed is too simplistic and fails to account adequately for the actual trial design.

The text itself is marred somewhat by a variety of minor syntactical, grammatical, and English usage errors. Since the authors did not provide page and line numbers as requested by PLOS ONE submission guidelines, no specific corrections will be offered.

*** Specific remarks: ***

The usual modern analysis for a crossover design involves using a linear mixed effects model with fixed effects for period and treatment, and random effects for sequence and subject nested in sequence. The analysis of variance results should be presented, as well as effect estimates. Consult Senn (2000) for the details.

For endpoints that the authors wish to model in finer detail, such as those involving measurements taken over time during each period, the authors may consider either summarizing the endpoints or elaborating the linear mixed effects model with within-period measurements. Summarizing is a less satisfactory method, because inference is then only to the summary measures rather than to the actual assessments of interest. The dependence due to repeated measurements over time should be accounted using an appropriate R-side correlation structure with either an unstructured correlation or a simple autoregressive correlation. While this model can be fit using either SAS (via PROC MIXED) or R (via the nlme package), it can also probably be fit using SPSS and perhaps using STATA as well.

Any linear model analysis will require diagnostics. At a minimum this will require graphical evaluation of the residuals and random effects to evaluate the qualities of homoscedasticity, independence, and symmetry. Normality can be evaluated using graphical evaluation of a normal quantile-quantile plot or by the use of a Shapiro-Wilk test.

The authors should plot the raw data from the trial either as a single spaghetti plot from pre to post with sequence indicated in color (i.e., both sequences superimposed) or in a separate panel for each sequence (i.e., both sequences juxtaposed). This should be done for each endpoint. These can be included as supplementary figures.

For endpoints that do not allow a simple spaghetti plot, the authors can plot individual subject profiles. In fact, this is highly recommended for the authors' own use anyway. It allows not only identification of outlying observations, but also identification of outlying subjects or outlying time points.

Note that randomization in a crossover design is not to groups, but to sequences.

Note that "repeated measures mixed models" is not a clear description. Instead, the authors need to describe the fixed effects, the random effects and their G-side covariance, and the R-side correlation structure used. If these are different for different endpoints, then the endpoints should be grouped as needed, and the model structure then detailed for each set of endpoints.

Note that models are not "performed", but are instead "fit", "used to assess", or the like. The same goes for tests of association --- these are not "performed" on data.

Rather than describing the analysis as "Statistical test X was used for categorical data", please list the endpoints followed by their statistical assessment, viz. "The association of treatment with endpoints X, Y, and Z was assessed using Fisher's exact test", or similar language.

The authors note that reductions were shown on the "financial difficulties" scale. This really seems to underscore the lack of any interpretable effects. This is not a criticism of the lack of effects, as it is equally important to publish the results of studies that "fail" to show effects as it is to publish studies that show effects. But, it is hard to see how the treatment would relate to this particular endpoint.

In general, please do not summarize results as "(all P>0.21)" and the like. The p-value is not that important of a measure, and anyway summarizing it in this way is only very slightly more informative than not summarizing it, at best. The results are in the table for the reader to see.

The authors write "indicating that t-VNS could be capable to module the CVT in a well-defined sub-group". Leaving aside the syntax error, this statement is a wild over-reach by the authors, considering their results and the methodology used (evidently) to get to this statement.

The authors note that "several types of analgesics" were in use. Acknowledging that it is unethical to control this usage in this setting, this seems like a potentially highly confounding factor. The authors should attempt at to quantify or at least assess the impact of this factor.

Table 1 should be presented also by sequence. Alternatively, the by-sequence breakdown could be presented in a supplementary table.

Table 2 is really confusing. Why are the sample sizes different for the various analyses? Why are some analyses labelled "supplementary"? What is being averaged over in the various columns? Perhaps Table 2 should be broken up into different tables, where the endpoints are similarly structured (i.e., single measurements post treatment, post versus pre, or more extensive repeated measures within period). The authors may wish to look at industry tables for presenting results from crossover designs.

Table 2 statistical tests should be presented pursuant to a proper analysis taking into account the crossover design and potentially the within-period repeated measures.

Table 4 should be reformatted to industry standards. The current layout is very poor.

In addition to Figure 3 it would be useful to see the data broken out by sequence, to see if there is anything like a carryover effect. The authors may wish to consider the possibility that this figure represents regression to the mean, since patients were required to have long-lasting pain before enrollment.

*** References: ***

Senn, S. J. (2000). Crossover design. Marcel Dekker, New York.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Eugenijus Kaniusas

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2021 Feb 26;16(2):e0247653. doi: 10.1371/journal.pone.0247653.r002

Author response to Decision Letter 0

7 Nov 2020

Response to reviewer #1 comments:

The reviewer wrote #1: “The study describes effects of the cervical transcutaneous vagus nerve stimulation in chronic pancreatitis patients. Irrespective of negative findings, this study design and results are of high value for the interested reader.”

Our reply: Thank you very much for your kind remarks and acknowledgement of our study.

The reviewer wrote #2: „Transcutaneous vagal neuromodulation“ is too general, leading potentially to confusion with the auricular transcutaneous vagal neuromodulation. I would recommend to integrate either “cervical” or “on the neck”

Our reply: We agree that it is more precise to include “cervical. The title has now been changed to: “The effects of two weeks cervical transcutaneous vagal neuromodulation in chronic pancreatitis patients: A randomised sham controlled clinical trial”.

The reviewer wrote #3: “Please add to the methodology section “cervical” or “on the neck”. Currently the reader of the abstract has no clue where the stimulation was performed.”

Our reply: We have now specified where the stimulation was performed by adding “cervical” in the abstract and the method section. Abstract: “Patients were randomly assigned to receive a two-week period of cervical transcutaneous vagal nerve stimulation using the gammaCore device followed by a two-week sham stimulation, or vice versa.”

Method section: “This study was an investigator-initiated, randomised, double-blinded, sham-controlled, crossover clinical trial assessing the effect of cervical nVNS for pain treatment in patients with CP.”

The reviewer wrote #4: The abbreviation t-VNS is a bit misleading since tVNS is usually used for the transcutaneous auricular stimulation, just in contrast to pVNS as used for the percutaneous auricular stimulation. The stimulation on the neck is usually abbreviated as nVNS. It is a suggestion to avoid confusion of the readership.

Our reply: We have now changed the abbreviation t-VNS to nVNS as suggested throughout the entire manuscript.

The reviewer wrote #5: Ad Methods: I think that 2 weeks stimulation is a relatively short period, especially in view of a long-lasting chronic ailment with the associate changes in plasticity. For instance, the auricular VNS is usually applied for 6 weeks in chronic ailments such as the chronic back pain (https://doi.org/10.1213/01.ane.0000107941.16173.f7). Even though it is shortly discussed in the limitations section, the reader would welcome a more detailed discussion on the effective treatment durations of VNS and cervical transcutaneous VNS (optionally, including also auricular VNS) in chronic ailments. Please discuss and justify this issue in detail.

Our reply: This is a valuable point, and we agree with the reviewer that two weeks stimulation may be a relatively short treatment period to improve chronic pain. However, previous prospective, double-blind, sham-controlled, randomized studies with transcutaneous VNS have demonstrated that cervical transcutaneous VNS (with gammaCore) improved pain relief (compared to sham treatment) at 30, 60 and 120 minutes in migraine patients [Tassorelli et al. Neurology. 2018; 91,4]. Therefore, we expected that two weeks could be enough to induce a long-term pain relieving effect in patients with chronic pancreatitis. The study conducted by Sator-Katzenschlager et al. [Sator-Katzenschlager et al. Anesth Analg. 2004; 98,5] was performed in chronic back pain (somatic pain and not visceral pain) which may require a different treatment duration than chronic visceral pain. Also, the study did use auricular electroacupuncture and not cervical transcutaneous vagal nerve stimulation. Finally, the treatment was performed for 6 weeks, however the electrical stimulation was only given once weekly, in order to assess both short term and long-term benefit of electrical acupuncture. In contrast, our study aimed to assess a short-term effect of transcutaneous vagal nerve stimulation. Therefore, caution must be applied when comparing to previous VNS studies.

As requested, we have now specified the rationale for choosing two weeks treatment in the method section: “Two weeks stimulation was chosen, since previous prospective, double-blind, sham-controlled, randomized studies with transcutaneous VNS have demonstrated that stimulations with gammaCore induced a pain reliving effect at 30, 60 and 120 minutes in migraine patients [13]. Therefore, we expected that two weeks would be sufficient to induce a long-term pain-relieving effect.”

Furthermore, we have now included a discussion regarding the treatment durations and type of VNS: “The duration of treatment seems to differ depending on the type of disease, type of intervention (invasive VNS, auricular or cervical nVNS), and time frame (short-term vs. long-term effect) [3]. For example, in chronic pain patients, an auricular nVNS study applied the treatment for six weeks [34], while a cervical nVNS study applied the treatment for two weeks [12]. Given that CP is characterized by long-lasting ailment, future studies may include a longer treatment duration to recover the functional plasticity in neural circuits of pain processing. In addition, the duration of the wash-out

period of two weeks was chosen since it has been shown to be sufficient to reset the effects of neuromodulation [10], but long-term effects have also been demonstrated in patients with epilepsy [35]. However, if we extended the study duration (treatment and wash-out periods), a high risk of dropout would potentially occur, and study completion would become more difficult.”

The reviewer wrote #6: Please amend whether state-of-the-art individual treatment of patients was optimized before the treatment, interrupted or continued during the tVNS study. Address concomitant medication/treatment of patients which is very important since improvements were registered with respect to the baseline.

Our reply: We have now specified that all patients continued their usual medication during the entire study, but they were allowed to take additional medication in case of severe pain attacks. The method section is updated: “All patients were asked to maintain their usual pain medication and dosage during the entire study, and they were only allowed to take extra pain medication in case of severe pain attacks. Any changes needed in pain medication during the study period was noted by the patient in the pain diary. Patients were allowed to continue the following analgesics during the study period: paracetamol, nonsteroidal anti-inflammatory drug (NSAID), opioids, tricyclic antidepressant (TCA), serotonin-norepinephrine reuptake inhibitor (SNRI) and gabapentoids.”

Patients served as their own controls due to the cross-over design and they were asked to continue their usual medication during the entire study period. The sample size of the current study did not allow for a detailed subgroup analysis of the impact of concomitant pain medication but we agree that it would be relevant to investigate the impact of concomitant medication in future studies. Thus, we added the following in the discussion: “Furthermore, the impact of concomitant medication, the duration of CP or other disease characteristics (including stress, anxiety, and depression) could also be relevant to investigate. In future studies, a larger sample size is required in order to investigate sub-groups of patients, who possible could have an effect of nVNS.”

The reviewer wrote #7: Ad Method: Quantitative description of CVT is missing, even though a reference is given. Please describe it in a few sentences so that the reader has a clear idea about it.

Our reply: We have now specified the methods for acquisition of cardiac derived parameters and a description of the method used for calculation of CVT in the method section: “Next, ECG electrodes (Ambu Blue Sensor P, Denmark) were placed in the left and right subclavicular areas and cardiac apex and then attached to a non-invasive cardiac monitor. A biosignal acquisition system (ExtremeBiometrics, London, UK) was used to assess cardiac derived parameters from five minutes recordings.” “CVT was calculated beat-to-beat based on detecting positive phase shifts in the R-R interval and used as a putative measurement of heart rate variability [18]. Thus, low CVT values reflect low heart rate variability, whereas high values reflect high heart rate variability.”

The reviewer wrote #8: Ad Method: More information is needed on the wash-out period, how long, 1 week? With uninterrupted concomitant treatment.

Our reply: We thank the reviewer for this comment and fully agree that this information should be included. A two-week wash-out period was used in the current study. This is now specified in the method section: “The treatment periods were separated by a two-week wash-out period. The patients continued their usual analgesic medication during the wash-out period. The two week wash-out period has been used in previous studies of neuromodulation and shown to be sufficient to reset the effects of neuromodulation [10].”

The reviewer wrote #9: Ad Discussion: p.15 Discussed comparison/justification of cervical tVNS results with respect to invasive VNS results is limited due to the basically different technologies applied. Therefore, please address potential validation of cervical tVNS effects with respect to invasive VNS effects to substantiate your conclusions here.

Our reply: We agree that cervical tVNS and invasive VNS are two different technologies, therefore caution must be applied when comparing those technologies. Hence, we have now addressed a potential validation of cervical tVNS effect with respect to invasive VNS in the discussion section: “Recently, a functional MRI study showed that cervical vagal afferents can be accessed non-invasively via transcutaneous electrical stimulation of the neck, which overlies the course of the vagus nerve [29]. Also, the brain activation elicited by nVNS was analogous to that elicited by invasive VNS, providing validation of cervical nVNS effects [27].”

 

Response to reviewer #2 comments:

The reviewer wrote #1: Abstract: specify that it was a cervical transcutaneous VNS. Specify the device (GammaCore). Our reply: We agree that this information should be presented in the abstract: “Patients were randomly assigned to receive a two-week period of cervical transcutaneous vagal nerve stimulation using the gammaCore device followed by a two-week sham stimulation, or vice versa.” Also, the title of the manuscript has been updated, specifying the use of cervical VNS.

The reviewer wrote #2: The number of patients is rather small for me, as specified by the authors.

Our reply: We agree with the reviewer that the sample size might be small, especially with regards to the secondary outcomes. We performed a sample size calculation based on the primary outcomes (VAS scores), revealing that 16 patients in a cross-sectional study should be sufficient to detect a clinical meaningful difference of 30% in the average daily pain score (primary endpoint). Thus, we consider that the number of patients was sufficient for the analysis of the primary outcome. Since no data exists on nVNS treatment in CP patients, the assumptions of the power calculation (standard deviation, etc.) was based on data from a study of patients with chronic pancreatitis patients, who received pregabalin treatment, which related to an improvement in clinical measures of the pain scores [Olesen et al. Gastroenterology. 2011;141,2]. However, we cannot exclude that our study was under-powered to detect an effect of VNS on the VAS score, since the true effect of VNS could be less than pregabalin. “The lack of pain improvement may reflect that a longer-lasting treatment is needed, but the study could also be under-powered to detect a potential effect.” We also acknowledge that it could be relevant to include more subjects in future studies in order to be able to investigate sub-groups of responders, impact of etiological factors, etc. We now added this to the discussion, now reading: “In future studies, a larger sample size is required order to investigate sub-groups of patients, who possible could have an effect of nVNS.”

The reviewer wrote #3: - VNS stimulation is generally performed on the left VN while in the present study the authors stimulated both VN. Is-there any advantage to perform such a bilateral stimulation? Our reply: This is a very good point. To our best knowledge, there are no clear evidence or recommendations with regards to unilateral or bilateral transcutaneous VNS. Hitherto, previous transcutaneous VNS stimulation studies using the gammaCore device have stimulated bilaterally (on both left and right vagal nerve) in chronic pain patients [Tassorelli et al. Neurology. 2018; 91,e364-e373] and in immune-mediated inflammatory diseases [Tarn et al. Neuromodulation 2019;22,5]. One argument could be that both left and right vagal nerves need to be stimulated to ensure that the input from the transcutaneous VNS to the nucleus tractus sollitarius in the brainstem is sufficient in order to alter the afferent input to other regions of the central nervous system. It is uncertain whether unilateral stimulation with transcutaneous VNS is enough to modulate the brainstem and higher centers in the central nervous system. Future studies will indeed be needed to evaluate the optimal stimulation approach and investigate whether there is any difference between unilateral and bilateral stimulation.

In our study, in order to stimulate efficiently with transcutaneous VNS, and since we treated for only 2 weeks, we have chosen to replicate the same bilateral stimulation approach as the previous gammaCore VNS studies. Accordingly, we added the following in the limitations section of the discussion: “Firstly, the paradigm and parameters of nVNS may influence the effect of the treatment, including the treatment duration, number of daily stimulations, stimulation intensity, unilateral vs. bilateral stimulation, etc.”

The reviewer wrote #4: The duration of the study (2 weeks of stimulation) is most likely too short since VNS is a slow acting therapy as demonstrated for epilepsy. Indeed, VNS induces modifications of neuronal plasticity, which are progressive. CP is not comparable to “headache” which is FDA approved with the device GammaCore. As stated by the authors CP induces central modifications which can induce a central sensitivity syndrome. In addition, most of these patients use opioid compounds which are also able to induce central perturbation of pain. So, it is simplistic to think that a two-week VNS will be able to improve such patients. In addition, if the effect of t-VNS is delayed over time, when switching to sham stimulation, it could induce a bias of the results of sham stimulation. Our reply: This is a valuable point. We already elaborated more on the treatment duration in response to reviewer 1, question #5, and the discussion is now reading: “The duration of treatment seems to differ depending on the type of disease, type of intervention (invasive VNS, auricular or cervical nVNS), and time frame (short-term vs. long-term effect) [3]. For example, in chronic pain patients, an auricular nVNS study applied the treatment for six weeks [34], while a cervical nVNS study applied the treatment for two weeks [12]. Given that CP is characterized by long-lasting ailment, future studies should possibly include a longer treatment duration to recover the functional plasticity in neural circuits of pain processing.”

We agree with the reviewer that a carry-over is a concern due to the cross-over design of our study. However, no carry-over effect was detected in the primary clinical endpoints in the revised statistical analysis (according to the reply to reviewer 3). Hence, it does not seem to be a significant bias of our study. To date, no other studies have conducted a cross-over study with the gammaCore device. Therefore, estimating the wash-out period was challenging. We agree that several VNS studies have shown long-term effect with VNS in patients with epilepsy. VNS treatment in epilepsy patients have been conducted with both invasive and non-invasive VNS, therefore it is important to differentiate between invasive VNS treatment and non-invasive transcutaneous VNS treatment. Most of the long-term VNS studies have been conducted with invasive VNS. The following is now added in the method section: “The treatment periods were separated by a two-week wash-out period. The patients continued their usual analgesic medication during the wash-out period. The two week wash-out period has been used in previous studies of neuromodulation and shown to be sufficient to reset the effects of neuromodulation [10].” The following was added in the discussion: “In addition, the duration of the wash-out period of two weeks was chosen since it has been shown to be sufficient to reset the effects of neuromodulation [10], but long-term effects have also been demonstrated in patients with epilepsy [35].”Finally, in the limitation section we addressed the risk of increasing the dropout rate if the study duration (treatment duration and wash-out period combined) was longer and the recruitment would become more difficult: “However, if we extended the study duration (treatment and wash-out periods), a high risk of dropout would potentially occur, and study completion would become more difficult.”

The reviewer wrote #5: In addition, the sham stimulation is rather questionable since it induces a vibration which, as noted by the authors, is able to have an effect on the auricular branch of the VN.

Our reply: We agree. As already mentioned in the limitation section, the sham device may provide some vibration and tactile sensation that potentially can stimulate the vagal nerve and induce an effect similar to the active treatment. This could also explain why there is no differences between the active and sham treatment, but a positive effect when comparing both the treatments to their respective baselines. The limitation section now reads: “Secondly, even though the sham device did not deliver electrical stimulation, it may not be physiologically inert as it provides some vibration and tactile sensation that potentially can stimulate the vagal nerve and induce an effect similar to the active treatment. This could also explain why there was no differences between the active and sham treatment, but a positive effect when comparing both the active and sham treatments to their respective baselines.”

The reviewer wrote #6: The authors did not discuss the parameters of VNS.

Our reply: There are many options for vagal nerve stimulation, such as frequency, amplitude, shape and duration of the electrical pulse. We used the fixed parameters of VNS related to the gammaCore device as specified in the method section. We agree that the parameters of VNS may influence the effect of treatment, as well as the treatment duration, number of daily stimulations, the stimulation intensity, unilateral vs. bilateral stimulation, etc. As this type of treatment is relatively new, the “optimal” study design and parameter settings are still not determined and need to be investigated further. We now added the following to the limitations section: “Firstly, the paradigm and parameters of nVNS may influence the effect of treatment, including the treatment duration, number of daily stimulations, stimulation intensity, unilateral vs. bilateral stimulation, etc.”

The reviewer wrote #7: The duration of CP is 8.1 ± 7.6 years, so very variable. Thus, the central integration of pain and central sensitization is supposed to be patient-dependent and different between patients. Our reply: This is correct. We are aware that the pain pattern and pain duration were not the same for all patients. Therefore, the degree of central sensitization, and also the potential benefit of VNS, could be patient-dependent as the reviewer pointed out. Indeed, future studies should take the degree of central sensitization into account (by sensory profiling, QST, brain MRI, etc.) to identify characteristics predicting sub-groups of patients that could specifically benefit from nVNS. We now added the following in the limitation section in the discussion: “Furthermore, the impact of concomitant medication, the duration of CP or other disease characteristics (including stress, anxiety, and depression) could also be relevant to investigate. In future studies, a larger sample size is required in order to investigate sub-groups of patients, who possible could have an effect of nVNS.”

The reviewer wrote #8: The analgesic treatments may interfere with VNS (see above)

Our reply: We agree. The use of analgesic drugs during the study could have an impact of degree of pain relief induced by nVNS. As also replied to reviewer 1, question #6, we have now specified that all patients continued their usual medication during the entire study, but they were allowed to take additional medication in case of severe pain attacks. The method section is updated: “All patients were asked to maintain their usual pain medication and dosage during the entire study, and they were only allowed to take extra pain medication in case of severe pain attacks. Any changes needed in pain medication during the study period was noted by the patient in the pain diary. Patients were allowed to continue the following analgesics during the study period: paracetamol, nonsteroidal anti-inflammatory drug (NSAID), opioids, tricyclic antidepressant (TCA), serotonin-norepinephrine reuptake inhibitor (SNRI) and gabapentoids.”

Patients served as their own controls due to the cross-over design and they were asked to continue their usual medication during the entire study as discussed above. This likely limits the effect of concomitant pain medication on the study results. However, we agree that it would be relevant to investigate the impact of concomitant medication in future studies powered for subgroup analyses. Thus, we added the following in the discussion: “Furthermore, the impact of concomitant medication, the duration of CP or other disease characteristics (including stress, anxiety, and depression) could also be relevant to investigate. In future studies, a larger sample size is required in order to investigate sub-groups of patients, who possible could have an effect of nVNS.”

The reviewer wrote #9: Pain is multidimensional and rather subjective, so the use of a VAS is questionable, although I agree that VAS is often used in such conditions. It seems to me that there are no questionnaires evaluating stress (perceived), depression, anxiety, coping strategies in this trial.

Our reply: We agree, and we think that this could be very relevant and interesting to investigate in future studies. In this study we focused on pain intensity and quality of life but we acknowledge that additional outcome measures focused on other pain dimensions would have been relevant. However, the number of outcome measures need to be balanced against feasibility concerns and by inclusion of numerous questionaries patients may be fatigued and thus provide less valid responses. Notwithstanding this limitation, the QOL questionnaire includes several components reflecting cognitive, emotional and social functioning to some degree. Future studies could be optimized based on our knowledge from this present study and include some of the mentioned components. We added the following in the limitations section: “Furthermore, the impact of concomitant medication, the duration of CP or other disease characteristics (including stress, anxiety, and depression) could also be relevant to investigate.”

The reviewer wrote #10: Did the patients used or have used complementary medicines which are able to impact the pain factor? Our reply: We have now, as stated in the response to your question #8, specified that all patients continued their usual medication during the entire study, and were allowed to take additional pain medication only in case of severe pain attacks. This is now addressed in the manuscript according to the reply to your question #8 and reviewer 1 question #6.

The reviewer wrote #11: A limit for me of non-invasive VNS by comparison to invasive one is the compliance. How the authors controlled that patients had regularly performed VNS (and bilaterally)? Our reply: We agree. The patients were instructed to register each stimulation intensity for the stimulation on both sides of the neck and as stated in the method section: “patients registered the intensity of the stimulation after each treatment in the pain diary”. Furthermore, as also stated in the methods section: “compliance was assessed by reading the remaining stimulations on the device display after each treatment period.” Thus, missing stimulations (non-compliance) could very likely be identified. In the end we have to rely on the patient’s willingness to provide the correct information. Information on compliance is provided in the result section, reading: “In the nVNS treatment, 98%±3% of all stimulations were applied compared with 94%±11% in the sham treatment (P=0.24). Furthermore, the average intensity of the stimulations was 31.8±6.9 units during nVNS treatment and 36.0±7.9 units for sham treatment (P=0.02).”

The reviewer wrote #12: Regarding vagal tone, did the authors include patients with only low vagal tone which should be more appropriate and is supposed to better beneficiate of VNS. It is not clear for me in the manuscript. Our reply: This is a very good point. We did not use CVT (assessment of vagal tone) as a screening tool for inclusion. An exclusion criterion was known peripheral neuropathy, but some patients could potentially have autonomic neuropathy which could make it more difficult to modulate the parasympathetic tone. A study by Liu et al. found that epilepsy patients having reduced heart rate viability compared to healthy controls, had significantly lower effect of VNS treatment (non-responders) [Liu et al. Epilepsia. 2017;48,6]. Thus, it seems likely that the potential effect of nVNS could be related to the degree of autonomic dysfunction. This could be relevant to investigate in future studies. We now clarified this in the discussion: “Despite one of the exclusion criteria was known peripheral neuropathy, some patients could have autonomic neuropathy. Hence, some patients could have vagal neuropathy, which can make it more difficult to modulate the parasympathetic tone and potentially mediating an anti-hyperalgesic effect [7]. A study by Liu et al. found that epilepsy patients having reduced heart rate viability compared to healthy controls, had significantly lower effect of VNS treatment (non-responders) [31]. Thus, the effect of nVNS could be related to the degree of autonomic dysfunction, and CVT may be useful as a screening tool in future studies.”

The reviewer wrote #13: The authors did not explore the anti-inflammatory effect of VNS which is well known in different chronic inflammatory conditions such as rheumatoid arthritis and Crohn’s disease. That would have been interesting and there is a relation between inflammation and pain. This point is only stated in the “Limitations” of the study.

Our reply: We agree. Exploration of the anti-inflammatory effect of VNS could be very interesting and could add more valuable information about the nVNS treatment. According to our published protocol in BMJ open [Muthulingam et al. BMJ Open. 2019;9,7], we have collected blood samples to assess the anti-inflammatory effect of this nVNS treatment, and we are very excited to look further into these data.

The reviewer wrote #14: Ongoing alcohol dependence was an exclusion criterion but not smoking? Nicotine is able to act on alpha7 nicotinic acetylcholine receptor and to have an anti-inflammatory effect. Our reply: We agree that smoking may have an impact on the anti-inflammatory effect. However, adding smoking as an exclusion criterion may lead to difficulties in recruiting CP patients in a 8-week clinical trial. In our study (see Table 1) almost 80% of patients were smokers and to enroll only non-smokers would have made it virtually impossible to recruit an adequate number of patients. We have included the following in the discussion: “Fourthly, even though the study was based on a cross-over design, where patients served as their own control, bias could be present due to the inclusion/exclusion criteria. For instance, ongoing alcohol dependence was an exclusion criteria whereas ongoing smoking was not, but nicotine could potentially impact the inflammatory-system [36].”

 

Response to reviewer #3 comments:

Reviewer #3 (General remarks): The study follows a two period two treatment crossover design. However, the analysis which was performed is too simplistic and fails to account adequately for the actual trial design.

The text itself is marred somewhat by a variety of minor syntactical, grammatical, and English usage errors. Since the authors did not provide page and line numbers as requested by PLOS ONE submission guidelines, no specific corrections will be offered.

Our reply: We thank the reviewer for these general remarks. We agree that our previous analyses did not adequately take the cross-over design into consideration in particularly regarding the analysis of carry-over effects. We have now consulted a senior statistician at our institution who agreed with the reviewer that a linear mixed effects model with fixed effects for period and treatment, and random effects for sequence and subject nested in sequence would be an appropriate statistical method. However, even thought our study was block randomized, the final allocation into the two sequences sham/VNS and VNS/sham was not balanced as most dropouts occurred in the latter sequence group. Consequently, a linear mixed effect model was not optimal for the structure of our data. As an alternative, the senior statistician advised us to use an analytic approach based on t-tests taking into account the cross-over design [Wellek et al. Dtsch Arztebl Ind. 2012;109,15]. This approach follows the guidelines recommended for analysis and reporting of crossover trials published in PLoS One [Li et al. PLoS One. 2015;10,8]. Overall, the results and conclusions obtained from the revised analysis are the same (i.e. a largely negative study). However, we are very grateful for the comments on the statistical analysis, which lead us to work more further with our data.

Since we have not utilized a mixed linear model for analysis, the specific questions related to this model (see below), are not applicable for our current analysis, and consequently not addressed in detail below.

We have revised the method and result sections and the tables in the manuscript according to the new analyses. We have addressed the remaining questions below and finally, we have now revised the entire manuscript thoroughly according to syntactical, grammatical and English usage errors.

The reviewer wrote #1:

The usual modern analysis for a crossover design involves using a linear mixed effects model with fixed effects for period and treatment, and random effects for sequence and subject nested in sequence. The analysis of variance results should be presented, as well as effect estimates. Consult Senn (2000) for the details.

Our reply: Please refer to the comments provided for the general remarks above. We have revised the method and result sections, and the tables, according to our new statistical analyses. Overall, the results and conclusions are the same.

The reviewer wrote #2:

For endpoints that the authors wish to model in finer detail, such as those involving measurements taken over time during each period, the authors may consider either summarizing the endpoints or elaborating the linear mixed effects model with within-period measurements. Summarizing is a less satisfactory method, because inference is then only to the summary measures rather than to the actual assessments of interest. The dependence due to repeated measurements over time should be accounted using an appropriate R-side correlation structure with either an unstructured correlation or a simple autoregressive correlation. While this model can be fit using either SAS (via PROC MIXED) or R (via the nlme package), it can also probably be fit using SPSS and perhaps using STATA as well.

Our reply: Please, see the reply to the general remarks. We have now provided estimates with 95% confidence intervals for the key endpoints in the abstract and result section.

The reviewer wrote #3: Any linear model analysis will require diagnostics. At a minimum this will require graphical evaluation of the residuals and random effects to evaluate the qualities of homoscedasticity, independence, and symmetry. Normality can be evaluated using graphical evaluation of a normal quantile-quantile plot or by the use of a Shapiro-Wilk test.

Our reply: We have checked the normality using Shapiro-Wilk test. All the variables are normally distributed. The following is added in the Method section: “The continues variables were checked for normality using Shapiro-Wilk test.”

The reviewer wrote #4: The authors should plot the raw data from the trial either as a single spaghetti plot from pre to post with sequence indicated in color (i.e., both sequences superimposed) or in a separate panel for each sequence (i.e., both sequences juxtaposed). This should be done for each endpoint. These can be included as supplementary figures. For endpoints that do not allow a simple spaghetti plot, the authors can plot individual subject profiles. In fact, this is highly recommended for the authors' own use anyway. It allows not only identification of outlying observations, but also identification of outlying subjects or outlying time points.

Our reply: We agree, that single spaghetti plots can provide important information and we now provided these plots for the primary endpoints in the supplementary materials (S3, Figure S1). The result section is reading: “No significant differences were observed in the maximal or average pain score between two weeks nVNS treatment and sham treatment (Figure 3 and supplementary material S3 for individual plots).”

The reviewer wrote #5: Note that randomization in a crossover design is not to groups, but to sequences.

Our reply: We thank the reviewer for noticing the error. We have now replaced “groups” with “sequences” in the manuscript.

The reviewer wrote #6: Note that "repeated measures mixed models" is not a clear description. Instead, the authors need to describe the fixed effects, the random effects and their G-side covariance, and the R-side correlation structure used. If these are different for different endpoints, then the endpoints should be grouped as needed, and the model structure then detailed for each set of endpoints.

Our reply: Please, see the reply to the general remarks. The statistical description has now been updated according to the new method used for statistical analysis.

The reviewer wrote #7: Note that models are not "performed", but are instead "fit", "used to assess", or the like. The same goes for tests of association --- these are not "performed" on data.

Our reply: Thank you for pointing this out. We have now revised the manuscript according to the proposed terminology.

The reviewer wrote #8: Rather than describing the analysis as "Statistical test X was used for categorical data", please list the endpoints followed by their statistical assessment, viz. "The association of treatment with endpoints X, Y, and Z was assessed using Fisher's exact test", or similar language.

Our reply: We have now updated the statistical section and specified the following: “Categorical data (PGIC and adverse events) were analyzed using a McNemar test.”

The reviewer wrote #9: The authors note that reductions were shown on the "financial difficulties" scale. This really seems to underscore the lack of any interpretable effects. This is not a criticism of the lack of effects, as it is equally important to publish the results of studies that "fail" to show effects as it is to publish studies that show effects. But, it is hard to see how the treatment would relate to this particular endpoint.

Our reply: We agree that it is hard to see how this short-term treatment would relate to “financial difficulties”. According to the updated manuscript, the revised statistical analysis showed no significant differences for “financial difficulties”.

The reviewer wrote #10: In general, please do not summarize results as "(all P>0.21)" and the like. The p-value is not that important of a measure, and anyway summarizing it in this way is only very slightly more informative than not summarizing it, at best. The results are in the table for the reader to see.

Our reply: We agree, the summarized results can now be seen in the tables and only few specific p-values are provided in the main text along with corresponding effect estimates.

The reviewer wrote #11: The authors write "indicating that t-VNS could be capable to module the CVT in a well-defined sub-group". Leaving aside the syntax error, this statement is a wild over-reach by the authors, considering their results and the methodology used (evidently) to get to this statement.

Our reply: Thank you for this comment. We agree that the statement is overreached and with our new analysis approach, only heart rate was significantly different between treatments. We have now rephrased the sentence in the Discussion section: “Despite unsuccessful enhancement of mean CVT in our study, nVNS could potentially be able to module the parasympathetic tone in a sub-group of patients.”

The reviewer wrote #12: The authors note that "several types of analgesics" were in use. Acknowledging that it is unethical to control this usage in this setting, this seems like a potentially highly confounding factor. The authors should attempt at to quantify or at least assess the impact of this factor.

Our reply: We agree, that the use of medication may influence the effect of nVNS. As the study was designed as a randomized cross-over study, and as patients were asked to continue their usual medication during the entire study (they were allowed to take additional medication in case of severe pain attacks), we intended to minimize the influence of medication between sessions. However, we agree that the potential interaction between different types of analgesics and the effect of nVNS should be investigated further in future studies, which of course need to be designed and powered for that purpose. We included the following statement in the limitations section: “Finally, various confounding factors are present in this patient cohort. For instance, the patients used several types of analgesics which may influence the effect of nVNS and as already discussed it may be difficult to modulate the parasympathetic tone for some patients. Furthermore, the impact of concomitant medication, the duration of CP or other disease characteristics (including stress, anxiety, and depression) could also be relevant to investigate. In future studies, a larger sample size is required in order to investigate sub-groups of patients, who possible could have an effect of nVNS.”

The reviewer wrote #13: Table 1 should be presented also by sequence. Alternatively, the by-sequence breakdown could be presented in a supplementary table.

Our reply: We agree, and this information is now implemented in the revised tables (Table 1 and Table S1 in supplementary material S2).

The reviewer wrote #14: Table 2 is really confusing. Why are the sample sizes different for the various analyses? Why are some analyses labelled "supplementary"? What is being averaged over in the various columns? Perhaps Table 2 should be broken up into different tables, where the endpoints are similarly structured (i.e., single measurements post treatment, post versus pre, or more extensive repeated measures within period). The authors may wish to look at industry tables for presenting results from crossover designs. Table 2 statistical tests should be presented pursuant to a proper analysis taking into account the crossover design and potentially the within-period repeated measures.

Our reply: Thank you for this comment. We have now simplified Table 2 and updated according to the revised statistical analyses.

The reviewer wrote #15: Table 4 should be reformatted to industry standards. The current layout is very poor.

Our reply: Thank you for this observation. We now improved the layout of Table 4.

The reviewer wrote #16: In addition to Figure 3 it would be useful to see the data broken out by sequence, to see if there is anything like a carryover effect. The authors may wish to consider the possibility that this figure represents regression to the mean, since patients were required to have long-lasting pain before enrollment.

Our reply: This is a great suggestion. We have now provided a new figure according to the reviewer’s suggestions, see Figure 3.

Attachment

Submitted filename: Reply letter PONE-D-20-13163.docx

Click here for additional data file. (64KB, docx)

Decision Letter 1

Susan Hepp

2 Feb 2021

PONE-D-20-13163R1

The effects of two weeks cervical transcutaneous vagal neuromodulation in chronic pancreatitis patients: a randomised sham controlled clinical trial

PLOS ONE

Dear Dr. Frøkjær,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Three experts in the field have reviewed your manuscript and have provided positive feedback.

The revisions that need to be made relate to your smaller sample size.

You state that your team found no difference between the groups in this study, however as your sample size is small, it is possible that your test is underpowered. You do state this limitation briefly in your manuscript. However it would be more correct to state throughout your manuscript that you found no evidence of a difference between groups (as opposed to finding evidence of no difference between groups). As such, please re-word some of your statements made in your manuscript as follows:

Lines 68-69:

Current statement:

"In this sham-controlled crossover study, two weeks transcutaneous vagal nerve stimulation did not induce pain relief in patients with chronic pancreatitis."

Suggested statement:

"In this sham-controlled crossover study, we found no evidence that two weeks transcutaneous vagal nerve stimulation induces pain relief in patients with chronic pancreatitis."

Lines 388-390:

Current statement:

"Our sham-controlled crossover study provided evidence that adjuvant treatment with two weeks nVNS did not induce pain relief as compared to sham treatment in patients with painful chronic pancreatitis"

Suggested statement:

"Our sham-controlled crossover study provided no evidence that adjuvant treatment with two weeks nVNS induces pain relief as compared to sham treatment in patients with painful chronic pancreatitis"

In addition, please add one to two sentences about your sample size limitation to the limitations section of your discussion.

Finally, we request that you revise the phrase "the effects of" in your manuscript title.

Please submit your revised manuscript by Mar 18 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

  • A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

  • A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

  • An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Susan Hepp, PhD

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: (No Response)

Reviewer #3: (No Response)

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: (No Response)

Reviewer #3: (No Response)

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: (No Response)

Reviewer #3: (No Response)

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: (No Response)

Reviewer #3: (No Response)

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #2: (No Response)

Reviewer #3: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Eugenijus Kaniusas

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2021 Feb 26;16(2):e0247653. doi: 10.1371/journal.pone.0247653.r004

Author response to Decision Letter 1

3 Feb 2021

Response to comments:

The editor wrote #1: “The revisions that need to be made relate to your smaller sample size.

You state that your team found no difference between the groups in this study, however as your sample size is small, it is possible that your test is underpowered. You do state this limitation briefly in your manuscript. However it would be more correct to state throughout your manuscript that you found no evidence of a difference between groups (as opposed to finding evidence of no difference between groups). As such, please re-word some of your statements made in your manuscript as follows:

Lines 68-69: Current statement: "In this sham-controlled crossover study, two weeks transcutaneous vagal nerve stimulation did not induce pain relief in patients with chronic pancreatitis."

Suggested statement: "In this sham-controlled crossover study, we found no evidence that two weeks transcutaneous vagal nerve stimulation induces pain relief in patients with chronic pancreatitis."

Lines 388-390: Current statement: "Our sham-controlled crossover study provided evidence that adjuvant treatment with two weeks nVNS did not induce pain relief as compared to sham treatment in patients with painful chronic pancreatitis"

Suggested statement: "Our sham-controlled crossover study provided no evidence that adjuvant treatment with two weeks nVNS induces pain relief as compared to sham treatment in patients with painful chronic pancreatitis"”

Our reply: Thanks for this helpful comment. We agree, due to the smaller sample size, that it is more correct to conclude that we found no evidence of difference between groups/treatments. We now reviewed the entire manuscript to adopt this. We agree that this concerns the two concluding statements (lines 68-69 and 388-390), which have now been corrected as suggested. In the beginning of the Conclusion section, we also adjusted lines 280-281 now reading “Overall, we found no evidence of significant difference in pain relief between the two treatments” and lines 288-290 reading “In contrast to our hypothesis, we found no evidence that CP patients had a significant pain reduction comparing the nVNS and sham treatments”.

The editor wrote #2: “In addition, please add one to two sentences about your sample size limitation to the limitations section of your discussion.”

Our reply: We now added the sample size discussion in the beginning of the limitations section, now reading: “First, even though a sample size calculation was performed prior to this study, such calculations in clinical studies with experimental interventions are often based on some uncertain assumptions. Hence, we could have a too small sample with a risk of false negative results regarding our primary outcomes”.

The editor wrote #3: “Finally, we request that you revise the phrase "the effects of" in your manuscript title.

Our reply: We agree. The title is now rephrased avoiding the “effect of” statement. It now simply reads: “Cervical transcutaneous vagal neuromodulation in chronic pancreatitis patients with chronic pain: a randomised sham controlled clinical trial”.

Attachment

Submitted filename: Response to reviewers PONE_D-20-13163-R1 030221.docx

Click here for additional data file. (15.9KB, docx)

Decision Letter 2

Julia Robinson

11 Feb 2021

Cervical transcutaneous vagal neuromodulation in chronic pancreatitis patients with chronic pain: a randomised sham controlled clinical trial

PONE-D-20-13163R2

Dear Dr. Frøkjær,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Julia Robinson

Senior Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Acceptance letter

Julia Robinson

17 Feb 2021

PONE-D-20-13163R2

Cervical transcutaneous vagal neuromodulation in chronic pancreatitis patients with chronic pain: a randomised sham controlled clinical trial

Dear Dr. Frøkjær:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Julia Robinson

Staff Editor

PLOS ONE

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    S1 Fig. Individual plots of the maximal and average pain scores.

    (DOCX)

    Click here for additional data file. (60.9KB, docx)
    S1 Table. Demographics by sequences.

    (DOCX)

    Click here for additional data file. (17.6KB, docx)
    S2 Table. The primary and secondary clinical endpoints corrected for baseline assessments.

    (DOCX)

    Click here for additional data file. (28.8KB, docx)
    S1 Appendix. CONSORT checklist and protocol.

    (PDF)

    Click here for additional data file. (1.1MB, pdf)
    Attachment

    Submitted filename: Reply letter PONE-D-20-13163.docx

    Click here for additional data file. (64KB, docx)
    Attachment

    Submitted filename: Response to reviewers PONE_D-20-13163-R1 030221.docx

    Click here for additional data file. (15.9KB, docx)

    Data Availability Statement

    Data is available at https://doi.org/10.6084/m9.figshare.14009030.

    Articles from PLoS ONE are provided here courtesy of PLOS

    RESOURCES