Music and low-frequency vibrations for the treatment of chronic musculoskeletal pain in elderly: A pilot study.

BACKGROUND: Transcutaneous vagal nerve stimulation has analgesic potential and might be elicited by abdominally administered low-frequency vibrations. The objective was to study the safety and effect of a combination of music and abdominally administered low-frequency vibrations on pain intensity in elderly patients with chronic musculoskeletal pain.
METHODS: This trial was an international multicenter, randomized controlled pilot study. Patients at age ≥ 65 years with musculoskeletal pain for ≥ 3 months and a daily pain score ≥ 4 out of 10 were recruited at three centers. They were randomized to receive either a combination of music and low-frequency (20-100 Hz) vibrations administered to the abdomen, or a combination with the same music but with higher frequency (200-300 Hz) vibrations administered to the abdomen. Low-frequency vibrations were expected to result in pain reduction measured with a numeric pain rating scale (NRS). Patients in both groups received eight treatments of the music combined with the vibrations in three weeks. Primary outcomes were safety (Serious Adverse Events) and pain intensity measured at baseline, after the last treatment and at six weeks follow-up. Multilevel linear model analyses were performed to study group and time effects.
RESULTS: A total of 45 patients were analyzed according to intention-to-treat principle. After 344 treatments, 1 Adverse Event was found related to the intervention, while 13 Adverse Events were possibly related. A multilevel linear model showed that the interaction effect of group by time did not predict pain intensity (F[1, 45.93] = 0.002, p = 0.97) when comparing pain intensity at baseline, after the last treatment and at follow-up.
CONCLUSIONS: The combination of music and abdominally administered vibrations was found to be safe and well tolerated by the elderly patients. However, over time, neither the low-frequency treatment group nor the high-frequency treatment group provided clinically meaningful pain relief. There is no evidence that the low-frequency treatment elicited vagal nerve stimulation. TRIAL REGISTRATION: The trial was prospectively registered in the Netherlands Trial Register (NTR: NL7606) on 21-03-2019.

Introduction

When looking into the impact of chronic musculoskeletal pain (CMP), it is reported that low back pain, neck pain and pain related to osteoarthritis are respectively the leading, 4th and 13th biggest causes of global years lived with disability, with an accumulated mean total of over 119 million in 2013 [1, 2]. Yet, the years lived with disability caused by these conditions are rising due to the aging of the global population, illustrating the enormous challenge that pain management is facing [3-7].

Most commonly reported treatments for CMP include the intake of drugs (e.g. paracetamol, NSAID’s and/or opioid analgesics [4]) and non-drug treatments such as physical- and exercise therapies, and transcutaneous electrical nerve stimulation [4, 8–10]. Interestingly, 31% of patients with CMP report not to be treated anymore for various reasons, including but not limited to not experiencing any effect of therapy on pain and suffering from side effects such as constipation and nausea [4]. The latter applies especially to elderly, as ageing is generally associated with higher prevalence of comorbidities and polypharmacy [11], which reduces appropriate pharmacological and invasive treatment options.

A relatively new approach in chronic pain management, is vagal nerve stimulation (VNS) [12]. Effects of VNS have been reported to alleviate pain, depression, anxiety, migraine and refractory epilepsy [12, 13]. Although it is unknown how VNS exactly results in relief in the aforementioned conditions, findings show that stimulation of vagal afferents elicits anti-inflammatory responses via activation of the cholinergic anti-inflammatory pathway [14] and potentially suppresses the transmission of pain signals both in the peripheral and central nervous system [15, 16]. VNS is mostly administered by a surgically implanted pulse generator that stimulates the left vagal nerve rostral of the aortic arch [12]. Downsides of this invasive method are the risk of side effects caused by efferent stimulation such as bradycardia, arrhythmia, cough and voice alteration and complications due to the surgery (e.g. infections and vocal cord paresis) [12, 13, 17, 18]. Non-invasive transcutaneous VNS (t-VNS) has already been proven safe and effective in the treatment of migraine, epilepsy, depression and cluster headache [19-22]. However, the results of studies examining the effect of t-VNS on the perception of pain have been inconsistent [23, 24]. Currently, t-VNS has mainly been attempted by electrically stimulating either auricular or cervical branches of the vagal nerve, but recent preliminary results suggest that VNS might also be induced by vibratory stimulation [21].

While vibratory stimulation has already proven to be effective in alleviating chronic pain [25-29], the effect has mainly been ascribed to the mechanisms of gate control theory [30]. According to this theory, the transmission of nociceptor signaling via small diameter C fibers can be inhibited in the dorsal horn by simultaneous stimulation of local mechanoreceptors with large diameter Aβ fibers [30, 31]. It is hypothesized that exposing the abdomen to vibratory stimulation could result in VNS; a considerable amount of vagal afferents innervate abdominal organs including the gastrointestinal tract, the liver and the pancreas [13, 32] and an abundance of mesentery nerves are thought to innervate Pacinian corpuscles [33]. These rapidly adapting corpuscles are known to be highly sensitive to pressure and vibrations with frequencies ranging from 20 to 1000 Hz [33], and send afferent stimuli through myelinated αβ fibers via the thalamus to SI and SII in the cortex [34]. Although the abdominal vagal afferents and the mesenterially innervated Pacinian corpuscles have not been linked in anatomical studies with one another, there are studies indicating a relationship between these two [35-39]. For instance, it is found that the combination of music and vibration that is used in vibro-acoustic therapy evokes changes in pain perception, heart rate variability, respiration rate, endocrine function [36, 37], which are all regulated by the vagal nerve [35, 40]. Moreover, it is observed that during vibro-acoustic therapy especially lower frequency vibrations (ranging from 30 to 120 Hz) were more effective in abdominal pain relief than higher frequency vibrations [37, 38]. This supports the idea that application of low-frequency vibration in the mesentery might result in VNS, as lower frequency vibrations have the ability to travel further through the skin than higher frequency vibrations with the same sound pressure [39].

T-VNS is a promising treatment, especially for patients that suffer from side-effects of medication or are not fit for invasive treatment due to comorbidities. Presumptively, t-VNS can be achieved by exposing the abdomen to low frequency vibrations, and combining it with music increases its analgesic potential [41-43]. This pilot study aimed to examine the safety and effect of music and low-frequency vibrations administered to the abdomen on CMP in elderly patients in addition to their current pain treatment(s). Both combinations of music and vibrations were expected to be safe and well tolerated by the elderly patients with CMP. It was also hypothesized that (I) the combination of vibrations administered to the abdomen and rhythmically aligned music would have a clinically meaningful analgesic effect in elderly with CMP and (II) the combination of music and vibrations with frequencies between 20–100 Hz would be more effective in relieving CMP in elderly than the combination of music and vibrations with higher frequencies (200–300 Hz), as in contrast to higher frequency vibrations, lower frequency vibrations could result in t-VNS.

Materials and methods

Study design

This study was an international multicenter, randomized, double-blind, pilot trial conducted between August 2019 and March 2020. The three centers that participated in this trial were Holbæk Hospital (Holbæk, Denmark), San Martino Hospital (Genova, Italy) and University Medical Center Groningen (Groningen, the Netherlands). The study duration per patient was 9 weeks. The medical ethics committee of UMC Groningen had approved the study (2019/227;NL69608.042.19), in accordance with the Helsinki declaration. The study protocol is available upon request. Written informed consent was obtained in advance from each patient. The trial was registered in the Netherlands Trial Register (NTR: NL7606) prior to the inclusion of the first patient.

Patients

Patients suffering from CMP were targeted by means of flyers hung in public places, media coverage of the trial, and by consultation of local general practitioners and in-clinic medical specialists. Interested patients were pre-screened during phone calls for suitability and if deemed suitable scheduled for examination to assess study eligibility definitely. Patients were included when at least 65 years, and suffering from symptoms of pain for at least 3 months, with a minimum intensity of Numeric (Pain) Rating Scale (NRS) of 4. Moreover, the pain had to be the result of a condition diagnosed as a musculoskeletal disease listed in the International Classification of Diseases-10 of 2014 under M00-M99 [44]. Patients participating in other (experimental) clinical studies, undergoing any kind of music therapy as a current treatment for their pain, suffering from active or untreated comorbidities (including moderate to severe depression), having pain related to malignancies, showing signs of the pain syndrome being exclusively or predominantly neurogenic during physical examination or being diagnosed with prolapsus disci intervertebralis with myelopathy/radiculopathies (listed as M50.0, M50.1, M51.0 and M51.1 in the International Classification of Diseases-10 of 2014) were excluded [44]. Patients were allowed to continue or even change both their pharmacological and non-pharmacological treatments (if applicable), but were requested to report any changes in treatment.

Procedures

Participants were randomized 1:1 per center to either regular ‘High Amplitude Low Frequency–Music Impulse Stimulation’ (HALF-MIS) treatment (i.e. the low-frequency group) or a variation of the HALF-MIS treatment, in which the low-frequency vibrations (20–100 Hz) of the regular HALF-MIS had been replaced by higher-frequency vibrations (200–300 Hz). This last treatment was considered to be an active control group for t-VNS, as higher frequency vibrations would not have the potential to result in vagal nerve stimulation. This would allow us to evaluate the second hypothesis, by purely examining the analgesic effect of t-VNS in addition to the effects music, rest, placebo and the mechanism of the gate control theory of pain on CMP in elderly. Before the inclusion of the first patient, a random sequence was created by T.E. and P.S. by matching study-ID’s to handpicked allocations. Then, for every future study-ID, a sealed opaque envelope containing the blinded group allocation (“A” or “B”) was distributed to the centers. The envelope was to be opened after a patient gave written informed consent and was found fit for inclusion. To ensure blinding of the patient, the equipment used for both treatment groups was also marked with “A” or “B”, but looked identical. Allocation was concealed for the patients until data collection was completed and for the primary researcher (RS) until the results were analyzed.

The San Martino Hospital suffered during the intervention period of the trial from both flooding and the inability to operate one of two treatment devices. These unfortunate circumstances led to a situation, in which protocol violations could not be averted. After the monitoring of all data by an independent monitor, it was concluded that the protocol violations were too severe for the data to be included in the analyses. That is the reason why the data of this center is not included in the CONSORT flow diagram (see Fig 1).

Fig 1

The CONSORT flow diagram.

Interventions

Treatments were administered by physicians T.E. and P.N., in the Netherlands and Denmark, respectively. During a treatment, patients sat or lied in the chair while being exposed to rhythmically aligned music administered through a headphone and synchronized vibrations, via a belt worn around the waist (see Fig 2). Both groups received 8 treatments of 24 minutes and 7 seconds (corresponding with the duration of the music) within three weeks, with at least one day without treatment between two consecutive treatments [21]. A follow-up visit was conducted 6 weeks after the last treatment. The equipment used for the HALF-MIS treatment was provided by PaciniMedico ApS (Copenhagen, Denmark). This consisted of a comfortable and adjustable chair, an elastic belt with a low frequency audio transducer (Guitammer, Westerville, OH), headphones, a central processing unit and a tablet (Apple Inc., Cupertino, CA) for the user interface on which both music volume and vibration intensity could be modified by the patient. The room in which the treatments took place, was standardized as much as possible throughout the intervention period to minimize exposure to external stimuli that could affect pain processing; windows were always closed, blinds were always down and the room was free of unnecessary material.

Fig 2

The equipment used for both the HALF-MIS treatment and the higher frequency variant in a clinical setting.

Reprinted from T.A.H. Eshuis et al. under a CC BY license, with permission from T.A.H. Eshuis et al., original copyright [2021].

Outcomes

The primary outcome to assess the safety of the intervention was the quantity and quality of reported (Serious) Adverse Events that were possibly related to the intervention([S]AE’s). The primary outcome to assess the efficacy of both treatments was patient perceived pain intensity score. The NRS score provides a valid and reliable measure for pain intensity for this study population, and measures pain on a 11-point scale ranging from 0 (no pain) to 10 (the worst pain imaginable) [45]. The effects of both treatments were assessed by a comparison of the pain intensity measured before the first treatment (baseline), after the last treatment and at follow-up.

The secondary outcomes were immediate pain relief, scores of the health-related quality of life (EQ-5D-3L), and pain disability (Pain disability Index; PDI). Immediate pain relief was quantified as the difference between the NRS score obtained immediately before and after each treatment, respectively and assessed for both groups. The health-related quality of life and pain disability questionnaires were filled out before the start of the first and eighth treatment and at follow-up, while screening for possible (S)AE’s took place at each visit. The EQ-5D-3L has been proven a valid tool for assessing quality of life in Danish, Dutch and Italian [46] and is found to be reliable when used in a Dutch elderly population [47]. The EQ-5D-3L is a self-report questionnaire that consists of 5 health dimensions (mobility, self-care, usual activities, pain/discomfort and anxiety/depression), which are all scored on a 3-point scale. It has a minimum score of -0.33 (no perceived quality of life) and a maximum score of 1 (highest quality of life possible perceived). Per country, the appropriate EQ-5D-3L valuation set was used for calculation of the total score, to account for sociocultural differences with regard to quality of life perspectives [48, 49]. The PDI has been proven a valid and reliable tool for measuring disability associated with pain in Dutch, and consists of 7 items which are scored on a 11-point scale [50]. The total score is calculated by the sum of scored item scores, divided by the number of scored items [51]. This yields a percentage ranging from 0 (no disability experienced) to 100 (experience of total disability). For this study, the PDI has been translated using a formal forward-backward method into Italian and Danish. To examine the course of quality of life and pain disability over time for each treatment group, within-group differences were compared for before the first treatment, after the last treatment and at follow-up.

Use of analgesics over time during intervention period was considered a possible confounder and was therefore also recorded at the screening visit. Moreover, changes in analgesic use were kept track of at each visit. Analgesics intake was then quantified using the Analgesics Intake Scale [52]. This tool yields a score from 0 (no use of pain medication) to 8 (use of opioids in combination with benzodiazepines/antidepressant/anticonvulsants), and allows for comparing analgesic intake between patients by taking into account both type and quantity of medication used.

Sample size

Since this was the first trial investigating the effect of HALF-MIS on CMP in elderly, there was no published or unpublished pilot data available to base a sample size on. For this pilot study, it was decided to aim for inclusion of 60 patients in total (i.e. 30 patients per treatment group).

Statistical analysis

All analyses are performed according the intention-to-treat principle to avoid potential exclusion bias, meaning that patients who started but did not finish the intervention were included in the analyses [53]. Differences in age, sex distribution, weight, height and body mass index between groups were compared using tests depending on data distribution and level of measurement.

The effects of various possible predictors on pain intensity were analyzed by means of a multilevel linear model to examine the effect of both treatments over a longer period of time, both within and between treatment groups. Heterogeneous first-order autoregressive structure was used as repeated covariance type. The model was built by starting with a fixed intercept and the interaction effect of group by time, as this was the effect of interest to check the second hypothesis. The model was then expanded with factors (both fixed and random) that significantly improved the model fit (χ2 [1] > 3.84) [54]. As it was expected that the effect of time would not be linear, different ways of coding the three most meaningful moments in time (i.e. before the first treatment, after the last treatment and at follow-up, respectively) had been included. As multilevel models do not require completeness of data, missing data was not imputed.

To assess the immediate pain relief within treatment groups, the average NRS score per patient before the delivered treatments was compared to the average NRS score per patient after the delivered treatments for both groups using Wilcoxon signed-rank test. The average difference between pre- and post-treatment NRS per patient between treatment groups were compared by means of Mann-Whitney test. Exclusion bias and missing data bias have been reduced by first taking the mean of all pre-treatment and post-treatment NRS scores for the Wilcoxon signed-rank test and mean difference per patient between each pre- and post-treatment NRS scores for the Mann-Whitney test, before conducting the tests mentioned. Effect sizes for Wilcoxon signed-rank test and Mann-Whitney test were calculated as r [54]. A r-value < 0.3 represents a small effect, 0.3 ≤ r ≤ 0.5 represents a medium-size effect and r > 0.5 represents a large effect [54]. For the remaining secondary outcomes quality of life and pain disability, changes within groups were examined by a comparison of within-group differences at baseline, after the last treatment and at follow-up. This was done by means of Friedman’s two-way ANOVA’s as subgroups of both variables were not normally distributed judged by visual inspection. To control for familywise error in the comparison of within-group differences for the secondary outcomes, Bonferroni correction was used and the two-sided level of statistical significance was set to 0.0125. All other statistical analyses were conducted with a two-sided 0.05 level of statistical significance, using SPSS statistics (version 23.0; IBM Corp. Armonk, NY).

Results

A total of 46 patients have been included and randomized (see Fig 1 and Table 1). However, one patient had not started the intervention when the study was terminated due to the COVID-19 outbreak, and was therefore excluded from the analyses. Three patients have not completed the intervention period and missed a total of 16 treatments (and accessory pain intensity measures); these were 9 low-frequency and 7 high-frequency treatments. Consequently, the 45 randomized patients have received a total of 344 treatments, during which 686 pain scores were obtained. In each group, 1 pain score was missing. The total number of follow-up visits that were performed is 42; 21 for each group. Of these visits, no data was missing. One non-pharmacological treatment during the intervention period was reported; a patient visited her physiotherapist. Pain intensity scores along with analgesic use at baseline, after the last treatment and at follow-up are given in Table 2, while secondary outcomes are given in Table 3.

Table 1

Descriptive characteristics per group at baseline (Mean ± SD or Median [IQR], except for sex).

	Low-frequency group (n = 23)	High-frequency group (n = 22)	p-value1
Age (years) (Median [IQR])	72.0 [68.0–77.0]	73.0 [66.0–79.3]	0.65
Sex (M/F) (Ratio)	6/17	6/16	0.98
Weight (kg) (Median [IQR])	78.5 [68.8–95.3]	72.0 [61.0–78.0]	0.08
Height (cm) (Mean ± SD)	165.8 ± 9.9	164.7 ± 9.0	0.69
Body mass index (kg/m2) (Median [IQR])	28.9 [24.9–33.9]	25.9 [23.4–28.6]	0.08

1p-value of either Mann-Whitney test (age, body mass index and weight), χ2-test (sex) or independent T-test (height).

Table 2

Pain intensity and analgesic use per group and for total sample before the first treatment (T0), after the last treatment (T1) and at follow-up (T2) (Mean ± SD or Median [IQR]).

	T0	T1	T2
NRS (Mean ± SD)
Low-frequency group	5.4 ± 2.5	5.1 ± 2.3	5.0 ± 2.6
High-frequency group	5.4 ± 2.3	4.0 ± 2.7	5.4 ± 2.1
Total sample	5.4 ± 2.4	4.5 ± 2.5	5.2 ± 2.4
Analgesics Intake Scale (Median [IQR])
Low-frequency group	2 [0–7]	2 [1–7]	2 [1–7]
High-frequency group	2 [0.75–7]	1 [0–7]	1 [0–7]
Total sample	2 [0–7]	2 [0–7]	2 [0–7]

1p-value of Friedman’s ANOVA.

Table 3

Secondary outcomes per group and for total sample before the first treatment (T0), after the last treatment (T1) and at follow-up (T2) (Median [IQR]).

	T0	T1	T2	p-value1
EQ-5D-3L
Low-frequency group	0.72 [0.48–0.78]	0.78 [0.52–0.79]	0.65 [0.49–0.79]	p = 0.70
High-frequency group	0.65 [0.42–0.81]	0.70 [0.48–0.81]	0.67 [0.43–0.81]	p = 0.94
Total sample	0.69 [0.45–0.78]	0.76 [0.50–0.81]	0.65 [0.48–0.81]
Pain Disability Index
Low-frequency group	31.4 [10.0–40.0]	32.9 [11.5–45.4]	32.9 [21.7–54.2]	p = 0.09
High-frequency group	37.5 [27.0–44.3]	30.0 [12.1–48.0]	42.9 [20.7–56.7]	p = 0.07
Total sample	31.4 [13.3–41.4]	32.1 [11.6–45.2]	36.4 [21.6–54.2]

1p-value of Friedman’s ANOVA.

The treatments appeared well received and safe. No Serious Adverse Events that were possibly related to the intervention were reported. One non-serious Adverse Event was reported that was related to the treatments as one subject perceived low back pain caused by poor posture while treatment was administered. 13 other non-serious Adverse Events might have been related, of which short episodes of respectively dizziness/vertiginous (n = 5) and headache (n = 2) and tiredness (n = 2) have been reported more than once.

With regard to the effectiveness of both treatments, the parameter estimates of the multilevel linear model are shown in Table 4. Independent variables that were examined as possible predictors were age, sex, body mass index, Analgesic Intake scale score, group (0 = high-frequency group, 1 = low-frequency group), time (0 = baseline, 0.5 = after last treatment, 1 = at follow-up) and center (0 = Holbaek, 1 = Groningen). Only center, group, and time significantly contributed as fixed (interaction) effect to the model fit and were therefore included as parameters. The group by time effect did not significantly predict pain intensity, F(1, 45.93) = 0.002, p = 0.97. The use of random slopes did not result in any improvement of the model. However, the relationship between the included predictors and pain did show significant variance in intercepts across patients of both groups, Var(u0j) = 3.61, χ2(1) = 17.71, p < 0.01. Only pain intensity at baseline (i.e. the intercept of the model) and center were significant predictors of pain intensity at later stages, while an interaction of group by time by center was close to being a significant predictor.

Table 4

Parameter estimates of the fixed effects of the final model predicting pain intensity.

	b1	SEb	95% CI
Baseline NRS score	5.57**	0.47	4.62, 6.53
Center	-1.43*	0.64	-2.70, -0.15
Group x Time	-0.02	0.58	-1.20, 1.15
Group x Time x Center	-1.46	0.81	-3.09, 0.16

Independent variables: center (0 = Holbaek, 1 = Groningen), group (0 = high-frequency group, 1 = low-frequency group) and time (0 = baseline, 0.5 = after last treatment, 1 = at follow-up).

The relationship between group and pain showed significant variance in intercepts across patients, Var(μ0j) = 3.61, χ2(1) = 17.71, p < 0.01.

* = p < 0.05.

** = p < 0.01.

1b = unstandardized regression coefficient.

With regard to the immediate effects of both treatments, related-samples Wilcoxon Signed Rank Tests showed for both treatment groups that the average NRS score per patient after treatment was significantly lower than the average NRS per patient before treatment: for the low-frequency group, the average NRS score after treatment (Mdn = 3.9, IQR = 2.3–5.0) was lower than the average NRS score before treatment (Mdn = 4.8, IQR = 3.4–5.4), p < 0.01, r = -0.49, while for the high-frequency group, the average NRS score after treatment (Mdn = 3.6, IQR = 2.5–5.3) was lower than the average NRS score before treatment (Mdn = 4.3, IQR = 3.3–6.6), p < 0.01, r = -0.50. The average difference per patient between pre- and post-treatment NRS scores from the low-frequency group (Mdn = 0.8, IQR = 0.4–1.5) did not differ significantly from those of the high-frequency group (Mdn = 0.6, IQR = 0.1–1.3), U = 235.00, z = -0.41, p = 0.68, r = -0.06.

Neither for the low-frequency group, χ2(2) = 0.71, p = 0.70, nor for the high-frequency group, χ2(2) = 0.12, p = 0.94, the scores of the EQ-5D-3L changed over time. Similarly, neither treatment group showed significant changes over time in Pain Disability Index score: χ2(2) = 4.84, p = 0.09 for the low-frequency group and χ2(2) = 5.29, p = 0.07 for the high-frequency group, respectively.

Discussion

This pilot study aimed to examine the safety and effect of music and low-frequency vibrations administered to the abdomen on CMP in elderly patients in addition to their current pain treatment. It was hypothesized that the combination of music and vibrations with frequencies between 20–100 Hz would be more effective in relieving CMP in elderly than the combination of music and vibrations with higher frequencies (200–300 Hz). However, there was no statistically significant interaction effect of group by time in the multilevel linear model predicting pain intensity, meaning that the low-frequency treatment was not more effective than the high-frequency treatment when comparing pain intensity at baseline, after the last treatment and at follow-up, respectively. It was also expected that the combination of vibrations administered to the abdomen and rhythmically aligned music would have a clinically meaningful analgesic effect in elderly with CMP, but as no time effect was found either, it seems that both treatments do not decrease CMP in elderly in the long run. The expectation that the treatment would be safe and well tolerated by the patients was confirmed, judging by both the quantity and quality of reported (Serious) Adverse Events that were possibly related to the intervention.

Patients with a higher baseline pain intensity score and patients treated in Holbæk showed higher pain intensity at a given point in time then patients with a lower baseline pain intensity score or those being treated in Groningen, respectively. The latter can be explained by a difference in recruitment; while most Dutch patients showed interest in the study after reading about the study in a newspaper without being previously referred to the clinic, Danish patients were recruited after consultation of the in-clinic specialists. This second approach has yielded patients with more severe pain syndromes. The interaction effect of group by time by center was close to being a statistically significant predictor of pain intensity. In both Danish and Dutch low-frequency groups, mean pain intensity was decreased non-significantly when comparing the ‘after the last treatment’ to baseline values but non-significantly increased at follow-up compared to ‘after the last treatment’. While this trend is also found in the Dutch high-frequency group, the Danish high-frequency group showed no change in mean pain intensity over time at all.

However, both the low-frequency and the high-frequency treatments showed a statistically significant decrease in pain intensity when pain intensity scores before and after the treatment are compared. While the effect sizes are moderate (r = -0.49) for the low-frequency and large (r = -0.50) for the high-frequency group, these changes are not always significant from the patient’s perspective; as the median drop in mean pain intensity score per patient is 19% for the low-frequency group and 16% for the high-frequency group, most patients experience only partial temporary relief of their moderate to severe musculoskeletal pain immediately after treatment. Still, the immediate effects found in this study seem larger than immediate-term effects of oral and transdermal opioid analgesics on pain intensity in elderly with CMP found by a systematic review, which showed a standardized mean difference of 0.27 in a sample of 4998 patients by combining 16 studies [55].

The analgesic effect on musculoskeletal pain of both treatments that is observed immediately after the treatment, can hypothetically arise from the interaction of five components: rest, placebo, music [41-43], the mechanism of the gate control theory of pain [25-29] and VNS [23, 24]. It was hypothesized that only the low-frequency treatment could lead to VNS, as only the low-frequency vibrations were expected to have the ability to reach and stimulate vagal afferents in the mesentery. This would either mean that the low-frequency treatment did not result in VNS, or that VNS did not enhance the analgesic effect of the other component(s).

No changes in health-related quality of life were found over time for either group, while both groups showed statistically non-significant changes over time in pain disability. This suggests that neither treatment reduced the extent to which the chronic musculoskeletal pain interfered with the patients’ lives [46, 50]. It is possible that the outbreak of COVID-19 and its societal consequences might have played a confounding role, as part of the data was collected during the outbreak. For instance, it is found in Germany that people who refrain themselves from going to public places show significantly lower health-related quality of life [56].

Although there is consensus that VNS affects pain processing, it remains unclear how this happens exactly [15, 16]. In the last decades, traditional VNS is gradually being replaced by safer t-VNS, but t-VNS is predominantly applied in the treatment of headache, epilepsy and depression disorders [17, 57]. Literature about the effect of electrical t-VNS on nociception is conflicting: while one study found an increase in pressure pain threshold and a decrease in reported pain ratings after an hour of transcutaneous stimulation of the auricular branch of the vagal nerve, another found that only most participants showed an increase in pain threshold after 30 minutes of similar stimulation [23, 24]. Moreover, some participants of the latter study even showed a decrease in pain threshold, indicating that t-VNS can result in both pro- and anti-nociceptive effect, depending on the individual. These latter results might explain why the low-frequency treatment did not outperform the high-frequency treatment in both short and long term, and why no long term effect was found for either treatment. Moreover, the result that the low-frequency treatment was a beneficial add-on in the treatment of depression [21], is in line with the idea that the low-frequency treatment leads to VNS.

Yet, it is more likely that the low-frequency treatment did not result in VNS. Although we have not measured (a proxy of) vagal activity, vibration-sensitive vagal afferents in the mesentery have never been found, and the indications that suggest that t-VNS might also be induced by vibratory stimulation are weak. This lack of evidence that HALF-MIS treatment actually leads to t-VNS is considered an important limitation of this study. Another limitation is that the anticipated number of 60 patients was not achieved: in March 2020, the study had to be terminated prematurely due to the COVID-19 outbreak. However, the data does not suggest that the hypotheses of this pilot study would have been confirmed if statistical power was higher. Feasibility was not examined in this pilot, which is retrospectively considered to be a limitation as well. A final limitation is that the COVID-19 outbreak and its societal consequences could have played a confounding role in this study. The isolating measures affect musculoskeletal pain as physical activity is generally drastically reduced, and perceived pain is often increased by anxiety and stress for instance, as pain can comprise cognitive, social and emotional components [58-60]. Since elderly are known to be at increased risk for an unfavorable course of illness [61], they are likely to be more affected by these consequences than any other group.

To our knowledge, this is the first study relying on the ability to excite abdominal vagal afferents transcutaneously with low-frequency vibrations in order to achieve pain reduction. However, changes in vagal activity have not been examined and it has not been proven that the mesenterial Pacinian corpuscles are innervated by the vagal nerve. The treatment proved to be safe, but both the low-frequency and the high-frequency treatments did not decrease CMP in elderly over time. A statistically significant but clinically insignificant immediate decrease in pain intensity was found for both treatments when comparing pre-treatment and post-treatment scores. Moreover, no effects were found on quality of life, pain disability or pain intensity over a longer period of time for both groups. The hypotheses that (I) the combination of vibrations administered to the abdomen and rhythmically aligned music would have a clinically meaningful analgesic effect in elderly with CMP and (II) the administration of low-frequency vibrations and music would be more effective than the same treatment, but with higher-frequency vibrations as a treatment for CMP in elderly are rejected. More research is needed in the treatment of CMP in elderly, as this population often has reduced treatment options due to comorbidities and poly-pharmacy. The treatment options that are left, often fail to achieve satisfactory pain relief. VNS has analgesic potential, but it remains unclear if the low-frequency treatment is able to induce VNS. Future research should focus on the relationship between VNS and pain processing and other possibilities to achieve transcutaneous VNS such as via the auricular branch of the vagal nerve.

CONSORT 2010 checklist of information to include when reporting a randomised trial*.

(DOC)

Click here for additional data file.

Research protocol.

(PDF)

Click here for additional data file.

Dataset.

(SAV)

Click here for additional data file.

11 Aug 2021

PONE-D-21-20572

The effect of High Amplitude Low Frequency – Music Impulse Stimulation on chronic musculoskeletal pain in elderly: a multicenter randomized controlled pilot study.

PLOS ONE

Dear Dr. Eshuis,

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.

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. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Walid Kamal Abdelbasset, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for stating the following in the Competing Interests section:

“I have read the journal's policy and the authors of this manuscript have the following competing interests:

PMN is registered inventor of the HALF-MIS technology. As a consequence, PMN did not participate in statistics processing or discussion of the results. For the remaining authors none were declared.”

Please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials, by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests).  If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf.

3. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

4. We note that Figure 2 in your submission contain copyrighted images. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright.

We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

a. You may seek permission from the original copyright holder of Figure 2 to publish the content specifically under the CC BY 4.0 license.

We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text:

“I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). Please be aware that this license allows unrestricted use and distribution, even commercially, by third parties. Please reply and provide explicit written permission to publish XXX under a CC BY license and complete the attached form.”

Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission.

In the figure caption of the copyrighted figure, please include the following text: “Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year].”

b. If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder’s requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.

[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: Partly

Reviewer #2: No

Reviewer #3: Yes

**********

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

Reviewer #1: No

Reviewer #2: I Don't Know

Reviewer #3: I Don't Know

**********

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: No

Reviewer #2: No

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: No

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: This is a very interesting randomised pilot study examining the effect of music and low frequency vibrations on chronic musculoskeletal pain in elderly patients. Well done to team for carrying out such important research.

The major comment is, this is a pilot study looking at efficacy and safety of an intervention. Which is misleading as generally speaking pilot studies are requisites to the initial step in exploring the feasibility of pre-specified approaches (e.g recruitment rate, retentions, adherence etc) intended for the larger study. Also the sample size does not suggest or reflect what the effect size is. Thus the making conclusions about clinically meaningful results misleading as we don’t know what effect size the study was designed to detect.

They are some comments worth mentioning for the authors attention.

1) Sample size section should come before statistical analysis section,

2) Line 220 “All analyses are performed according the intention-to-treat principle” can the authors explicitly define this. as CONSORT and Table 1 don't agree.

3) As this is a pilot randomised controlled trial, it’s not recommended to test baseline characteristics as any differences that may occur are simply due to chance and also due to randomisation you expect a balance between groups. i.e remove p-values from Table 1. i.e If analysis was based on ITT, why does Table 1 include =45 instead of the n=46 randomised individuals.

4) Table 1 – usually a table showing baseline characteristics by randomised group. Results stratified by centre should be omitted from this table. Also please indicate what summary statistic has been presented. i.e. Age(years) ?median [?IQR] do the same for other variables.

5) The authors have used non-parametric methods on parametric estimates.. If Wilcoxon has been used meaning data is believed to be skewed therefore this should be used to test if there is a difference in medians? Similar comment for use of Mann-Whitney tests.

6) Authors should make reference to Cohen’d effect size and not just state ‘effect’ size

7) In the statistical analysis, authors should mention, how missing data would be handled.

8) Was there an analysis plan and finalised prior to final analysis?

9) Within group differences have been presented in Table 2, has multiple testing been accounted for? And not sure this is worth presenting as a linear mixed effect model has been fitted.

10) Where the variables chosen to be included in the model decided as priori?

11) Table 3 should present the unadjusted and adjusted group estimates i.e. for T1 and T2 – together with 95%CIs and p-values

Reviewer #2: Title

Kindly frame title such that it is accurate, informative, descriptive, succinct, simple and specific.

Introduction

Totally need editing and rewrite and it is too long

Methods:

Kindly focus on three basic elements of methods section.

How the study was designed?

How the study was carried out?

How the data were analyzed?

Eligibility criteria for participants not clear and Provide sufficient details of interventions of each group to allow replication

Statistics

Completely Not clear –please rewrite again and discuss which test you used and why also all tables need rearrange

Outcomes and estimation need to be explained well

Discussion:

needs to be as per well defined objectives

Describe sources of potential bias and imprecision

It has to be framed in such a way that readers are able to have good understanding of the current evidences and rationale of the paper

Reviewer #3: Dear Authors,

Kindly consider the following amendments or respond to them

Abstract:

Conclusion: line 53 the phrase “assumed” does not reflect a conclusive summary… recommend to change to “we believe”

Line 55: trial registration should be removed to the design and methodology section

Introduction

Line 61: you mentioned LBP, neck pain and OA and just gave 2 prevalence (4th and 13th)

Line 108: you need to rely on more than 1 reference (38) to support the efficacy of low frequency vibrations

The rationale or research gaps still needs to be more clear

Materials and methods:

Design:

Line 125: As the study was conducted between 125 August 2019 and March 2020, are there any reasons to state publishing your results after 2 years?

Line 126: although it is stated that San Martino Hospital (Genova, Italy) participated in the study settings but not of the author affiliations represent it!

Patients:

Line 132: what about rheumatic arthritic conditions? Were they excluded?

How was the analgesic effect of medication controlled? Especially if patients were using different medications?

Procedure:

Line 152: full term before acronyms in the first statement

Kindly list who provided the therapy and his/her expertise

Line 180-182: reference for the time and frequency of treatment

Outcomes:

list validity and reliability of the outcome measures used

line 209: has the translation been validated? How?

Statistical analysis:

Line 220: why was the intent to treat used since it’s a pilot study?

Results:

Table 1: kindly add an astrex “*” to the significant value “ weight”

Table 2: p values missing for NRS

Line 304: full term of “AE”

Discussion:

Lines 345-349: are there no other justifications for the non statistical difference rather than the Covid?

The discussion is well written but still lacks consolidated justification of the findings

**********

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: No

Reviewer #2: No

Reviewer #3: Yes: Tamer M. Shousha

[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.

Submitted filename: reviewer recommendations -1.docx

Click here for additional data file.

11 Oct 2021

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: Partly

Reviewer #2: No

Reviewer #3: Yes

Answer: Throughout the conduct of the study, we did face some obstacles that led to certain limitations (e.g. the COVID pandemic and the exclusion of the Italian site). Despite these limitations, we are convinced to have produced a technically sound manuscript with conclusions that are supported by the data.

________________________________________

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

Reviewer #1: No

Reviewer #2: I Don't Know

Reviewer #3: I Don’t Know

Answer: We have improved the readability and clearity of our “Statistical analysis” – paragraph, and hope this allows the reviewers to judge our analyses more easily.

________________________________________

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: No

Reviewer #2: No

Reviewer #3: No

Answer: We have now added the minimal data set that was used for this study.

________________________________________

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: No

Reviewer #3: Yes

Answer: We have thoroughly inspected the language of our manuscript, and are pleased to see that no specific grammar, spelling or fashion errors are specified by the reviewers.

________________________________________

5. Review Comments to the Author

Reviewer #1: This is a very interesting randomised pilot study examining the effect of music and low frequency vibrations on chronic musculoskeletal pain in elderly patients. Well done to team for carrying out such important research.

The major comment is, this is a pilot study looking at efficacy and safety of an intervention. Which is misleading as generally speaking pilot studies are requisites to the initial step in exploring the feasibility of pre-specified approaches (e.g recruitment rate, retentions, adherence etc) intended for the larger study. Also the sample size does not suggest or reflect what the effect size is. Thus the making conclusions about clinically meaningful results misleading as we don’t know what effect size the study was designed to detect.

Answer: We thank the reviewer for the kind words on the conduct of our study. We understand that not elaborately examining the feasibility is a limitation, which we now addressed in the limitation section in lines 448-449: “Feasibility was not examined in this pilot, which is retrospectively considered to be a limitation as well”. Yet, we think it is correct to refer to this study as a pilot study, since we could not make an estimate on the effect size before the start of the trial, as no comparable data was available, as mentioned in lines 237-238: “Since this was the first trial investigating the effect of HALF-MIS on CMP in elderly, there was no published or unpublished pilot data available to base a sample size on.”.

Regarding the strength of our drawn conclusions, we agree that we might have been too strong when stating that the achieved pain relief was generally not clinically meaningful, and that we should be more tentative in our conclusions. Please take note of the following adaptions that have been made to be more tentative in our conclusion and avoid “misleading conclusions” (to cite Reviewer #1):

- In the abstract, we specified that the effect of either treatment over time was clinically not meaningful (instead of initially stating that the treatment was generally clinically not meaningful (see lines 55-57: “However, over time, neither the low-frequency treatment group nor the high-frequency treatment group provided clinically meaningful pain relief.”)

- In the discussion a more tentative conclusion is drawn (see lines 383-387: “It was also expected that the combination of vibrations administered to the abdomen and rhythmically aligned music would have a clinically meaningful analgesic effect in elderly with CMP, but as no time effect was found either, it seems that both treatments do not decrease CMP in elderly in the long run.”)

- In the discussion, we used different wording when interpreting our results (see lines 404-409: “While the effect sizes are moderate (r = -0.49) for the low-frequency and large (r = -0.50) for the high-frequency group, these changes are not always significant from the patient’s perspective; as the median drop in mean pain intensity score per patient is 19% for the low-frequency group and 16% for the high-frequency group, most patients experience only partial temporary relief of their moderate to severe musculoskeletal pain immediately after treatment.”)

- In the discussion, the word “irrelevant” was replaced by “insignificant” (see lines 460-462: “A statistically significant but clinically insignificant immediate decrease in pain intensity was found for both treatments when comparing pre-treatment and post-treatment scores.” )

- In general, the results regarding the safety of the treatment have become more pronounced (see for instance lines 46-47: “Primary outcomes were safety (Serious Adverse Events) and pain intensity, measured at baseline, after the last treatment and at six weeks follow-up.”, lines 49-51: “. After 344 treatments, 1 Adverse Event was found related to the intervention, while 13 Adverse Events were possibly related.”, lines 54-55: “The combination of music and abdominally administered vibrations was found to be safe and well tolerated by the elderly patients.”, lines 320-325: “The treatments appeared well received and safe. No Serious Adverse Events that were possibly related to the intervention were reported. One non-serious Adverse Event was reported that was related to the treatments as one subject perceived low back pain caused by poor posture while treatment was administered. 13 other non-serious Adverse Events might have been related, of which short episodes of respectively dizziness/vertiginous (n = 5) and headache (n = 2) and tiredness (n = 2) have been reported more than once.”, and lines 387-389: “The expectation that the treatment would be safe and well tolerated by the patients was confirmed, judging by both the quantity and quality of reported (Serious) Adverse Events that were possible related to the intervention.”)

They are some comments worth mentioning for the authors attention.

1) Sample size section should come before statistical analysis section,

Answer: Thank you for bringing this to our attention. We have changed this accordingly.

2) Line 220 “All analyses are performed according the intention-to-treat principle” can the authors explicitly define this. as CONSORT and Table 1 don’t agree.

Answer: There was indeed a mistake in the CONSORT Flow diagram, which we have corrected. Moreover, we have now explicitly defined how all analyses have been performed regarding the intention-to-treat analysis (see lines 242-244: “All analyses are performed according the intention-to-treat principle to avoid potential exclusion bias, meaning that patients who started but did not finish the intervention were included in the analyses (53).”). To clarify the exact number of patients that have been included in the analyses; 46 patients were randomized in the trial, but 1 of these was not seen for baseline measurement or had any treatment. We have explained this now in lines 280-282: “A total of 46 patients have been included and randomized (see Fig 1 and Table 1). However, one patient had not started the intervention when the study was terminated due to the COVID-19 outbreak, and was therefore excluded from the analyses.”.

3) As this is a pilot randomised controlled trial, it’s not recommended to test baseline characteristics as any differences that may occur are simply due to chance and also due to randomisation you expect a balance between groups. i.e remove p-values from Table 1. i.e If analysis was based on ITT, why does Table 1 include =45 instead of the n=46 randomised individuals.

Answer: We agree that any differences between groups are not expected. However, any differences that simply occur by chance, could potentially influence the effectiveness of the treatment, and lead to biased results. For the sake of transparency, we have kept the p-values in Table 1. For the ITT, please see our answer under comment #2, above.

4) Table 1 – usually a table showing baseline characteristics by randomised group. Results stratified by centre should be omitted from this table. Also please indicate what summary statistic has been presented. i.e. Age(years) ?median [?IQR] do the same for other variables.

Answer: We agree with the reviewer, and removed the centre groups from the Table. Additionally, we indicated which summary statistic was used per variable.

5) The authors have used non-parametric methods on parametric estimates.. If Wilcoxon has been used meaning data is believed to be skewed therefore this should be used to test if there is a difference in medians? Similar comment for use of Mann-Whitney tests.

Answer: We believe this is a misunderstanding, probably caused by our wording, as both Wilcoxon and Mann-Whitney tests have been used on non-parametric estimates (i.e. medians, not means). What we have done is calculate e.g. the mean NRS score per patient before each treatment, and also calculate the mean NRS score per patient after each treatment, so that you get two variables which are scored per patient, called “mean NRS score pre-treatment” and “mean NRS score post-treatment”. Then, we compared these by means of a Wilcoxon test (using the medians of the [mean NRS score per patient]), since these two variables were not meeting the assumptions required for parametric testing etc.

We hope to prevent future misunderstandings, by replacing the word “mean” by “average” in lines 257-261 (“To assess the immediate pain relief within treatment groups, the average NRS score per patient before the delivered treatments was compared to the average NRS score per patient after the delivered treatments for both groups using Wilcoxon signed-rank test. The average difference between pre- and post-treatment NRS per patient between treatment groups were compared by means of Mann-Whitney test.”) and lines 356-365 (“With regard to the immediate effects of both treatments, related-samples Wilcoxon Signed Rank Tests showed for both treatment groups that the average NRS score per patient after treatment was significantly lower than the average NRS per patient before treatment: for the low-frequency group, the average NRS score after treatment (Mdn = 3.9, IQR = 2.3-5.0) was lower than the average NRS score before treatment (Mdn = 4.8, IQR = 3.4-5.4), p < 0.01, r = -0.49, while for the high-frequency group, the average NRS score after treatment (Mdn = 3.6, IQR = 2.5-5.3) was lower than the average NRS score before treatment (Mdn = 4.3, IQR = 3.3-6.6), p < 0.01, r = -0.50. The average difference per patient between pre- and post-treatment NRS scores from the low-frequency group (Mdn = 0.8, IQR = 0.4–1.5) did not differ significantly from those of the high-frequency group (Mdn = 0.6, IQR = 0.1–1.3), U = 235.00, z = -0.41, p = 0.68, r = -0.06.”), to distinguish between the meaning related to its summarizing nature (to which we want to refer) and the meaning related to a statistical method.

6) Authors should make reference to Cohen’d effect size and not just state ‘effect’ size

Answer: We did not use Cohen’s d, but r as an effect size. We believe that r is a more appropriate than Cohen’s d in case of non-parametric testing, as the latter assumes normal distribution and is based on means, and should therefore only be used as a parametric measure of effect size.

7) In the statistical analysis, authors should mention, how missing data would be handled.

Answer: We agree, and have added information to our methods section (see lines 255-256: “As multilevel models do not require completeness of data, missing data was not imputed.” and lines 261-265: “Exclusion bias and missing data bias have been reduced by first taking the mean of all pre-treatment and post-treatment NRS scores for the Wilcoxon signed-rank test and mean difference per patient between each pre- and post-treatment NRS scores for the Mann-Whitney test, before conducting the tests mentioned.”). We did not impute missing data.

8) Was there an analysis plan and finalised prior to final analysis?

Answer: Due to the exploratory nature of this study, we did not a priori know which exact choices in statistics would be made. There was a global analysis plan, which was stated in the study protocol and approved by the METC.

9) Within group differences have been presented in Table 2, has multiple testing been accounted for? And not sure this is worth presenting as a linear mixed effect model has been fitted.

Answer: We added to the paragraph “Statistical analysis” (lines 271-273, “To control for familywise error in the comparison of within-group differences for the secondary outcomes, Bonferroni correction was used and the two-sided level of statistical significance was set to 0.0125.”). The multilevel linear model was only used for the prediction of pain intensity scores, and our secondary outcome measures have not been included in the model. Therefore, we believe presenting the within-group differences of our secondary outcomes is of added value.

10) Where the variables chosen to be included in the model decided as priori?

Answer: Group and Time and Center effects were included a priori in the study protocol.

11) Table 3 should present the unadjusted and adjusted group estimates i.e. for T1 and T2 – together with 95%CIs and p-values

Answer: Also in line with the feedback from Reviewer#2, we have rearranged the tables. Table 2 now describes pain intensity and analgesic use for T0, T1 and T2. As both outcomes have been included and tested in the multilevel linear model, no test statistics are given in Table 2. In Table 3, the secondary outcomes are both described and tested (with p-values provided), as these are not included in the model. In Table 4, the parameter estimates of the model are given, with 95% CI’s.

Reviewer #2: Title

Kindly frame title such that it is accurate, informative, descriptive, succinct, simple and specific.

Answer: We have changed the title to “Music and low-frequency vibrations for treatment of chronic musculoskeletal pain in elderly: a pilot study”.

Introduction

Totally need editing and rewrite and it is too long

Answer: We have critically looked at our introduction, and omitted any unnecessary detailed information. As stated in the discussion, the indications that suggest that t-VNS might also be induced by vibratory stimulation are weak. We aimed to provide an comprehensive background and rationale for this study, which is also intelligible for readers that are no expert in the fields of pain medicine, vagal nerve stimulation or vibro-acoustic therapy.

Methods:

Kindly focus on three basic elements of methods section.

How the study was designed?

How the study was carried out?

How the data were analyzed?

Answer: We adhered to the CONSORT 2010 checklist, which is required by PLOS ONE when reporting a clinical trial. We believe that we have reported the methods of this study concisely.

Eligibility criteria for participants not clear

Answer: We have improved the readability of the eligibility criteria (see lines 141-151: “Patients were included when at least 65 years, were eligible if suffering from the symptoms of pain for at least≥ 3 months, with a minimum intensity of Numeric (Pain) Rating Scale (NRS) of 4. Moreover, the pain had to be the result of a condition diagnosed as a musculoskeletal disease listed in the International Classification of Diseases-10 of 2014 under M00-M99 (44). Patients participating in other (experimental) clinical studies, undergoing any kind of music therapy as a current treatment for their pain, suffering from active or untreated comorbidities (including moderate to severe depression), having pain related to malignancies, showing signs of the pain syndrome being exclusively or predominantly neurogenic during physical examination or being diagnosed with prolapsus disci intervertebralis with myelopathy/radiculopathies (listed as M50.0, M50.1, M51.0 and M51.1 in the International Classification of Diseases-10 of 2014) were excluded (44).”).

and Provide sufficient details of interventions of each group to allow replication

Answer: We agree on the reviewer and we inserted a paragraph entitled “Interventions” (starting from line 177). This should now enable replication of this study.

Statistics

Completely Not clear –please rewrite again and discuss which test you used and why also all tables need rearrange

Answer: We have elaborated the description of our statistics (please see lines 242-244: “All analyses are performed according the intention-to-treat principle to avoid potential exclusion bias, meaning that patients who started but did not finish the intervention were included in the analyses (53).”, lines 261-265: “Exclusion bias and missing data bias have been reduced by first taking the mean of all pre-treatment and post-treatment NRS scores for the Wilcoxon signed-rank test and mean difference per patient between each pre- and post-treatment NRS scores for the Mann-Whitney test, before conducting the tests mentioned.” and lines 271-274: “To control for familywise error in the comparison of within-group differences for the secondary outcomes, Bonferroni correction was used and the two-sided level of statistical significance was set to 0.0125. All other statistical analyses were conducted with a two-sided 0.05 level of statistical significance, using SPSS statistics (version 23.0; IBM Corp. Armonk, NY).”. Also in line with the feedback from Reviewer#1 comment #11, we have rearranged the tables. Table 2 now describes pain intensity and analgesic use for T0, T1 and T2. As both outcomes have been included and tested in the multilevel linear model, no test statistics are given in Table 2. In Table 3, the secondary outcomes are both described and tested (with p-values provided), as these are not included in the model. In Table 4, the parameter estimates of the model are given, with 95% CI’s.

Outcomes and estimation need to be explained well

Answer: We shifted the focus slightly to safety (besides the efficacy, see the last point of our answer to the general comment of Reviewer#1 above). Moreover, we have also added comments about the reliability of our outcome measures, in line with the feedback of Reviewer#3 (see for instance lines 205-207: “The NRS score provides a valid and reliable measure for pain intensity for this study population, and measures pain on a 11-point scale ranging from 0 (no pain) to 10 (the worst pain imaginable) (45).”, lines 215-216 “The EQ-5D-3L has been proven a valid tool for assessing quality of life in Danish, Dutch and Italian (46) and .is found to be reliable when used in a Dutch elderly population (47).”and lines 222-223 “The PDI has been proven a valid and reliable tool for measuring disability associated with pain in Dutch, and consists of 7 items which are scored on a 11-point scale (50).”).

Discussion:

needs to be as per well defined objectives

Answer: We have edited our discussion; in the discussion, we now refer to the hypotheses that were composed in the introduction (see lines 378-389: “It was hypothesized that the combination of music and vibrations with frequencies between 20-100 Hz would be more effective in relieving CMP in elderly than the combination of music and vibrations with higher frequencies (200-300 Hz). However, there was no statistically significant interaction effect of group by time in the multilevel linear model predicting pain intensity, meaning that the low-frequency treatment was not more effective than the high-frequency treatment when comparing pain intensity at baseline, after the last treatment and at follow-up, respectively. It was also expected that the combination of vibrations administered to the abdomen and rhythmically aligned music would have a clinically meaningful analgesic effect in elderly with CMP, but, as no time effect was found either, it seems that both treatments do not decrease CMP in elderly in the long run.. The expectation that the treatment would be safe and well tolerated by the patients was confirmed, judging by both the quantity and quality of reported (Serious) Adverse Events that were possibly related to the intervention.”), and concisely reflect upon these by interpreting our results (e.g. lines 402-409: “However, both the low-frequency and the high-frequency treatments showed a statistically significant decrease in pain intensity when pain intensity scores before and after the treatment are compared. While the effect sizes are moderate (r = -0.49) for the low-frequency and large (r = -0.50) for the high-frequency group, these changes are not always significant from the patient’s perspective; as the median drop in mean pain intensity score per patient is 19% for the low-frequency group and 16% for the high-frequency group, most patients experience only partial temporary relief of their moderate to severe musculoskeletal pain immediately after treatment.” and comparing these with existing literature (see e.g. lines 409-412: “Still, the immediate effects found in this study are larger than immediate-term effects of oral and transdermal opioid analgesics on pain intensity in elderly with CMP found by a systematic review, which showed a standardized mean difference of 0.27 in a sample of 4998 patients by combining 16 studies (55).”).

Describe sources of potential bias and imprecision. It has to be framed in such a way that readers are able to have good understanding of the current evidences and rationale of the paper

Answer: we described potential biases and limitations in line 441-455: “Yet, it is more likely that the low-frequency treatment did not result in VNS. Although we have not measured (a proxy of) vagal activity, vibration-sensitive vagal afferents in mesentery have never been found, and the indications that suggest that t-VNS might also be induced by vibratory stimulation are weak. This lack of evidence that HALF-MIS treatment actually leads to t-VNS is considered an important limitation of this study. Another limitation is that the anticipated number of 60 patients was not achieved: in March 2020, the study had to be terminated prematurely due to the COVID-19 outbreak. However, the data does not suggest that the hypotheses of this pilot study would have been confirmed if statistical power was higher. Feasibility was not examined in this pilot, which is retrospectively considered to be a limitation as well. A final limitation is that it is likely that the COVID-19 outbreak and its societal consequences could have played a confounding role in this study. The isolating measures affect musculoskeletal pain as physical activity is generally drastically reduced, and perceived pain is often increased by anxiety and stress for instance, as pain can comprise cognitive, social and emotional components (58, 59, 60). Since elderly are known to be at increased risk for an unfavorable course of illness (61), they are likely to be more affected by these consequences than any other group.”. Please note that this includes the limitation of us not addressing the feasibility of this study.

Reviewer #3: Dear Authors,

Kindly consider the following amendments or respond to them

Abstract:

Conclusion: line 53 the phrase “assumed” does not reflect a conclusive summary… recommend to change to “we believe”

Answer: We fully agree that “assumed” does not reflect a conclusive summary, and changed this to: “There is no evidence that the low-frequency treatment elicited vagal nerve stimulation” (lines 57-58). In our opinion, this is better than “we believe”, as we feel this way of expression still lacks conclusive properties.

Line 55: trial registration should be removed to the design and methodology section

Answer: Although we fully agree that trial registration does indeed belong to the design and methodology section, PLOS ONE also requires authors to state the trial registration in the abstract, in case of clinical trials. Therefore, we did not edit this part of the abstract.

Introduction

Line 61: you mentioned LBP, neck pain and OA and just gave 2 prevalence (4th and 13th)

Answer: We provided three causes of years lived with disability, namely: Leading (I) 4th (II) and 13th (III).

Line 108: you need to rely on more than 1 reference (38) to support the efficacy of low frequency vibrations

Answer: We have added a reference to a paper which supports the efficacy of low-frequency vibrations (line 112). We hope this suffices.

The rationale or research gaps still needs to be more clear

Answer: We completely agree that there are considerable research gaps in this field of research. Indeed, there has not been any anatomical, nor physiological studies directly showing connections between sensors in the abdomen and the vagal nerve. In the introduction, we aimed to fully expose these research gaps, while remaining concise and avoiding being exhaustive (with regard to Reviewer#2, the comment about the introduction being too long). Moreover, we address these research gaps as important limitations (see lines 441-445: “Yet, it is more likely that the low-frequency treatment did not result in VNS. Although we have not measured (a proxy of) vagal activity, vibration-sensitive vagal afferents in the mesentery have never been found, and the indications that suggest that t-VNS might also be induced by vibratory stimulation are weak. This lack of evidence that HALF-MIS treatment actually leads to t-VNS is considered an important limitation of this study.”). In order to progress in this field, we conducted this pilot study.

Materials and methods:

Design:

Line 125: As the study was conducted between 125 August 2019 and March 2020, are there any reasons to state publishing your results after 2 years?`

Answer: When the study was prematurely terminated in March 2020, we initially awaited how the COVID pandemic would develop, as we were hoping for the trial to be continued. After some time and analyses, the study was terminated definitively. Yet, only one year has passed since the first submission in the meantime, and it has been submitted to two different journals previously.

Line 126: although it is stated that San Martino Hospital (Genova, Italy) participated in the study settings but not of the author affiliations represent it!

Answer: As the authors were not defined a priori, and after excluding the Italian partners, none of them met the ICMJE criteria for authors.

Patients:

Line 132: what about rheumatic arthritic conditions? Were they excluded?

Answer: Since rheumatic arthritic conditions are listed under M05 and M06 in the ICD-10, these patients were not excluded a priori, but eligible if meeting the other inclusion criteria.

How was the analgesic effect of medication controlled? Especially if patients were using different medications?

Answer: Since the intervention was in addition to care as usual, patients were allowed to continue, stop or start any pain medication. We controlled for analgesic use by means of keeping track of patients’ analgesic use throughout the trial, which was quantified using the Analgesic Intake Scale (lines 230-235: “Use of analgesics over time during intervention period was considered a possible confounder and was therefore also recorded at the screening visit. Moreover, changes in analgesic use were kept track of at each visit. Analgesics intake was then quantified using the Analgesics Intake Scale (52). This tool yields a score from 0 (no use of pain medication) to 8 (use of opioids in combination with benzodiazepines/antidepressant/anticonvulsants), and allows for comparing analgesic intake between patients by taking into account both type and quantity of medication used.”). Based on this, we concluded that there were no significant changes in medication intake during the trial. Moreover, the Analgesic Intake Scale was used in the multilevel linear model, but did not show to be a predictor of pain intensity.

Procedure:

Line 152: full term before acronyms in the first statement

Answer: We have made sure the full term is listed first, before using acronyms for referral.

Kindly list who provided the therapy and his/her expertise

Answer: We have added that the treatments were provided by physicians (and [co-]authors) Eshuis and Nielsen (see lines 178-179: “Treatments were administered by physicians T.E. and P.N., in the Netherlands and Denmark, respectively.”).

Line 180-182: reference for the time and frequency of treatment

Answer: We have added a reference for the time and frequency of the treatment (see line 183).

Outcomes:

list validity and reliability of the outcome measures used

Answer: We have added the reliability of the outcome measures, as validity was already given (see for instance lines 205-207: “The NRS score provides a valid and reliable measure for pain intensity for this study population, and measures pain on a 11-point scale ranging from 0 (no pain) to 10 (the worst pain imaginable) (45).”, lines 215-216 “The EQ-5D-3L has been proven a valid tool for assessing quality of life in Danish, Dutch and Italian (46) and .is found to be reliable when used in a Dutch elderly population (47).”and lines 222-223 “The PDI has been proven a valid and reliable tool for measuring disability associated with pain in Dutch, and consists of 7 items which are scored on a 11-point scale (50).”.

line 209: has the translation been validated? How?

Answer: The PDI has been validated in other languages including Dutch. We did not perform a formal validation for Danish or Italian, which is a limitation. Yet, we believe that within the scope of this study and its results (in which the PDI is only a secondary outcome), this limitation is not worth mentioning when compared to other limitations with bigger impact (e.g. the thin indications that suggest that t-VNS might also be induced by vibratory stimulation).

Statistical analysis:

Line 220: why was the intent to treat used since it’s a pilot study?

Answer: We believed that it would be valuable to apply the intention-to-treat principle and reduce potential exclusion bias, as applying the intention-to-treat principle yields the most generalizable results and expectations based on realistic adherence, for when the treatment was to be implemented in usual care. In the hypothetical case that adherence to the treatment was low, we would not have overestimated the effect of the treatment, which would happen when applying the per-protocol principle instead.

Results:

Table 1: kindly add an astrex “*” to the significant value “ weight”

Answer: We agree that adding an asterix would have been appropriate to indicate the statistically significant difference in weight between both treatment groups. However, Reviewer#1 requested to omit the comparison of demographic characteristics between the two centers. Therefore, we have omitted these comparisons in Table 1.

Table 2: p values missing for NRS

Answer: In Table 2, only pain intensity and analgesic intake are described at T0, T1 and T2, which are tested in the multilevel linear model. Hence, there are no p-values in that table, as that would result double testing.

Line 304: full term of “AE”

Answer: We have replaced all AE abbreviations with the full term ‘adverse event’.

Discussion:

Lines 345-349: are there no other justifications for the non statistical difference rather than the Covid?

The discussion is well written but still lacks consolidated justification of the findings

Answer: We think the main reason for finding non-significant differences is because the low-frequency treatment might not elicit vagal nerve stimulation (see lines 441-445, “Yet, it is more likely that the low-frequency treatment did not result in VNS. Although we have not measured (a proxy of) vagal activity, vibration-sensitive vagal afferents in mesentery have never been found, and the indications that suggest that t-VNS might also be induced by vibratory stimulation are weak. This lack of evidence that HALF-MIS treatment actually leads to t-VNS is considered an important limitation of this study.”).

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

19 Oct 2021

Music and low-frequency vibrations for the treatment of chronic musculoskeletal pain in elderly: a pilot study.

PONE-D-21-20572R1

Dear Dr. Eshuis,

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,

Walid Kamal Abdelbasset, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

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: (No Response)

Reviewer #2: Yes

Reviewer #3: Partly

**********

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

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: I Don't Know

**********

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: No

Reviewer #2: Yes

Reviewer #3: Yes

**********

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: Yes

Reviewer #3: Yes

**********

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: thanks alot or your response

Reviewer #3: Most of the comments have been addressed.. Thank you for the 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: No

Reviewer #2: No

Reviewer #3: No

22 Oct 2021

PONE-D-21-20572R1

Music and low-frequency vibrations for the treatment of chronic musculoskeletal pain in elderly: a pilot study.

Dear Dr. Eshuis:

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

Dr. Walid Kamal Abdelbasset

Academic Editor

PLOS ONE