Pujol et al. Arthritis Research & Therapy (2019) 21:148 https://doi.org/10.1186/s13075-019-1932-9 RESEARCH ARTICLE Open Access Testing the effects of gentle vibrotactile stimulation on symptom relief in fibromyalgia Jesus Pujol1,2*, Daniel Ramos-López1, Laura Blanco-Hinojo1,2, Guillem Pujol1, Héctor Ortiz3, Gerard Martínez-Vilavella1, Josep Blanch4, Jordi Monfort4 and Joan Deus1,5 Abstract Background: Sensory disturbances in fibromyalgia extend beyond nociception. It has been proposed that imbalance in the mutual competition between painful input and non-painful sensory activity may, to a significant extent, account for the augmented subjective perception of pain. In this context, non-nociceptive somatosensory stimulation could arguably attenuate fibromyalgia symptoms by restoring the sensory balance. We specifically tested the effect of vibrotactile stimulation on symptom relief in fibromyalgia patients with a randomized, double- blind, sham-controlled, crossover clinical trial. Methods: Seventy-seven female patients were randomized and data from 63 valid cases were analyzed. Active intervention involved extensive body stimulation with gentle mechanical vibrations administered during 3 h at night for 3 weeks, and the placebo effect was controlled using identical instruments to simulate an alternative treatment option. The primary outcome measure combined pain, fatigue, and complaints of poor cognition. Results: Vibrotactile stimulation was significantly superior to sham in alleviating fibromyalgia symptoms globally. However, univariate analyses showed that the effect was not universal. Benefits were perceived on unpleasant somatic sensations such as generalized pain and fatigue, but not on poor cognition, anxiety, and depression. Vibrotactile stimulation was notably well tolerated and sleep quality significantly improved despite the fact that vibrations were administered at night. Conclusions: Results thus provide new evidence that non-nociceptive somatosensory stimulation may favorably act upon altered somatosensory balance in fibromyalgia. From a clinical perspective, both the degree of improvement and the easy application of our proposal would seem to support a potential role for vibrotactile stimulation in the symptomatic treatment of fibromyalgia. Trial registration: ClinicalTrials.gov registration number NCT03227952. Registered 24 July, 2017. Keywords: Fibromyalgia, Somatosensory system, Pain, Vibrotactile stimulation, Sensory balance Background absence of an observable organic cause [1]. Nevertheless, Fibromyalgia has been a controversial disorder in some current methods for studying neural function offer new medical contexts due to the subjective nature of its opportunities to explore subtle clinical phenomena. One symptoms. Patients indeed complain of generalized pain, such tool that contributes to characterizing the patho- fatigue, unrefreshing sleep, and poor cognition in the physiology of fibromyalgia is functional MRI [2–4]. By means of functional connectivity and task activation ap- proaches, we have recently observed functional alter- * Correspondence: 21404jpn@comb.cat 1 MRI Research Unit, Department of Radiology, Hospital del Mar, Passeig ations beyond the nociceptive domain which would Marítim 25-29, 08003 Barcelona, Spain suggest a weak integration of other sensory inputs con- 2 Centro Investigación Biomédica en Red de Salud Mental, CIBERSAM G21, tributing to clinical pain in fibromyalgia [5, 6]. Other re- Barcelona, Spain Full list of author information is available at the end of the article search has provided data consistent with such a © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Pujol et al. Arthritis Research & Therapy (2019) 21:148 Page 2 of 11 downregulation of the non-nociceptive component of recommendations [26] and was registered at the US Na- somatosensory processing [7–12]. tional Institutes of Health (ClinicalTrials.gov), with iden- It has been proposed that imbalance in the mutual com- tifier number NCT03227952 and title “Study of the petition between painful inputs and non-painful sensory ac- effectiveness of vibrotactile sensory stimulation in fibro- tivity may, to a significant extent, account for the myalgia patients”. augmented subjective perception of pain and body discom- fort [6]. In this context, non-nociceptive somatosensory stimulation could arguably attenuate fibromyalgia symp- Participants toms by restoring the sensory balance, regardless of Patients were recruited from the Fibromyalgia Unit of whether non-painful sensory activity alteration is a primary Barcelona’s Hospital del Mar - Parc de Salut Mar be- phenomenon or a by-product of originally facilitated (or de- tween September 2017 and May 2018. A total of 166 pa- ficiently filtered) nociceptive signals. There are empirical tients clinically diagnosed with fibromyalgia were studies demonstrating the beneficial effects of treatments contacted through a consecutive order based on clinical based on physical/sensory stimulation that do not directly visit schedules (JB and JM). One hundred and seventeen aim at the nociceptive system. Tested procedures showing patients agreed to be screened for eligibility and were a range of success include, for instance, physical exercise fully briefed on the study and the corresponding inclu- and movement-based therapies, hydrotherapy, and periph- sion/exclusion criteria. Seventy-seven patients were eral nerve stimulation [11, 13–17]. Interestingly, the bene- eventually randomized following the exclusion of 23 pa- fits of non-nociceptive stimulation contrast with the tients who did not meet study criteria and 17 patients paradoxically poor efficacy of genuine analgesic drugs in re- who declined to participate. lieving fibromyalgia pain [13, 15, 18]. One way to selectively stimulate the non-nociceptive component of the somatosensory system is by using Eligibility criteria mechanical vibrations. Vibrotactile stimuli are captured Inclusion criteria were based on the following factors: fe- by a variety of widespread skin and musculoskeletal tis- male, aged 18 to 65 years, diagnosed by a specialist in sue receptors and transmitted via large-diameter myelin- fibromyalgia in accordance with the American College ated fibers separately from the nociceptive pathway [19]. of Rheumatology classification and diagnostic criteria [1, Experimental studies interestingly suggest that fibro- 27], the patient did not suffer from any other disorder myalgia pain can be effectively modulated by vibrotac- that might account for the pain, chronic use treatments tile stimuli [20]. Moreover, there are empirical studies in stable doses, a full understanding of the study, and an specifically testing the potential usefulness of vertical express commitment to compliance. Criteria to exclude oscillating platforms that have reported optimistic re- patients were generalized inflammatory articular or sults [21–25]. However, there have been no previous rheumatic disease; severe or non-stable medical, endo- studies controlling for the possible placebo effects of crine, or neurological disorder; psychotic disorder or vibration-based treatments applied to fibromyalgia drug abuse; evidence of poor compliance; and events patients. that could relevantly interfere with the trial. In our study, we tested the effect of vibrotactile stimula- Intervention effects were evaluated when added to on- tion on symptom relief in fibromyalgia patients with a ran- going treatments. Changes in the previously prescribed domized, double-blind, sham-controlled, crossover clinical treatments and procedural therapies (e.g., nerve blocks trial. The intervention involved extensive body stimulation or joint injections) were forbidden throughout the trial. with gentle mechanical vibrations administered during 3 h Only variations in the daily dose of the habitual analge- at night for 3 weeks, and the placebo effect was controlled sics were allowed. using identical instruments to simulate an alternative treatment option. The primary outcome was change in the key symptoms of fibromyalgia combining pain, fatigue, Study design and complaints of poor cognition. This randomized, double-blind, sham-controlled, two- period crossover clinical trial measured the effect of Materials and methods vibrotactile sensory stimulation versus sham treatment This study was conducted in accordance with the princi- on symptom relief in fibromyalgia patients. All patients ples expressed in the Declaration of Helsinki and was underwent a period of vibrotactile stimulation and a approved by the Ethical Committee of Clinical Research period of sham with complete counterbalancing, half of of the Parc de Salut Mar of Barcelona (reference no. them in that sequence and the other half first receiving 2016/6932/I). All patients provided written informed sham treatment. Both treatments lasted 3 weeks and consent. The trial was designed according to CONSORT were separated by a 2-week washout period. Pujol et al. Arthritis Research & Therapy (2019) 21:148 Page 3 of 11 Study interventions bedtime and 45% prior to getting up. Patients were The active treatment involved whole-body sensory instructed to turn on the system at bedtime and try to stimulation with mechanical stimuli of vibrotactile (pal- get to sleep naturally. The system was programmed to lesthetic) type at a relatively high rate, low intensity, and automatically stop 2 h later. The following morning, long duration administered at sleep time. To generate once the patients had naturally awoken, they were to set mechanical vibrotactile stimuli, 6 vibration motors (i.e., the device to run for a further hour and remain in their conventional electric motors with an eccentric mass beds. Patients were allowed to fall asleep during this fixed to the rotor) were fitted into a standard mattress period. (80 by 190 cm) symmetrically positioned to fully cover it, Intensity and frequencies of vibrations delivered were with the exception of the area in which the patient’s measured in real conditions using a tri-axial accelerom- head lay while sleeping. The 6 vibration motors gener- eter specifically designed to measure the exposure of ated a slowly varying spectrum of frequencies ranging workers to vibrations transmitted to the whole body from 2 to 90 Hz, which was mechanically transmitted to (CESVA AC033, CESVA Instruments, Barcelona) ac- the whole mattress. An electronic programmer was built cording to the International Organization for Standardi- to control stimulus delivery for duration and intensity zation’s norm ISO 2631-1. The accelerometer was (Fig. 1). The whole vibration system was manufactured calibrated using the CESVA CV211 multi-frequency vi- by LED SpA (Aprilia, Italy). bration calibrator. Vibration spectrum was determined The treatment involved the application of vibrotactile using VSHOOTER VBS1T vibration analysis camera stimulation with a total daily duration of 3 h distributed (Synergys Technologies, Altkirch, France). within a period of 2 h at bedtime and 1 h prior to getting The measurements obtained indicated that the nor- up. The intensity was set to 30% of power (see below) at malized vibration intensity applied to the body at 30% of Fig. 1 Instruments used to administer vibrotactile stimulation (above) and sham (bottom). An electronic engine controlled stimuli delivery. Mechanical vibrotactile stimuli were generated using 6 vibration motors fitted into a standard mattress. Sham treatment was applied using identical instruments, with an electrical signal turning on a pilot light to indicate that the (simulated) treatment was operative Pujol et al. Arthritis Research & Therapy (2019) 21:148 Page 4 of 11 power in a 3-h exposure was 0.03 m/s2, and for 3 h at Outcome measures 45%, it was 0.04 m/s2. Such energy delivery is much The primary outcome was “change in fibromyalgia’s key lower than the daily occupational exposure permitted symptoms rated by 101-point numerical rating scales for whole-body vibrations at work (1.15 m/s2) according (NRS).” The effect on pain, fatigue, and “cognitive symp- to current European and Spanish regulations (Spanish toms” was jointly analyzed using a repeated measures law RD 1311/2005). And also below the limit from multivariate analysis of variance (MANOVA) including which adopting procedures for risk prevention at work separate NRS scores. In addition, participants rated is recommended (0.5 m/s2). When the stimulus was set intervention effects providing a single global NRS score at 30% of power, dominant frequencies were widely dis- for all three symptoms (pain, fatigue, and cognitive tributed with a peak at 22 Hz and secondary peaks from symptoms) considered together. This rating was given 4 to 90 Hz. When the stimulus was set at 45% of power, by participants as a direct estimation of percentage (%) the dominant frequency peaked at 30.6 Hz instead. improvement of their symptoms after the interventions. Sham treatment was applied using identical instru- The rationale for using such a combination of symp- ments and with power and duration programmed identi- toms is their relevance in characterizing fibromyalgia as cally. However, in this case, the output was not the a syndrome, in addition to “waking unrefreshed” [1]. signal activating the vibration motors, but rather an elec- “Waking unrefreshed” was not considered within the trical signal turning on an incorporated pilot light indi- primary outcome as the study’s intervention was applied cating that the (simulated) treatment was operating during the night and we had no a priori knowledge of its (Fig. 1). The patient was told that “the study involved effects on sleeping comfort. the comparison of 2 treatment options; vibration versus Pain, fatigue, and cognitive symptoms were individu- magnetic waves, both of which are potentially effective ally rated at study baseline as the overall symptom sever- treatments in fibromyalgia.” However, patients were ity during the last month. Upon their first visit, patients clearly informed that study interventions might or might were specifically instructed and trained to rate symp- not be effective in the context of a clinical trial. The par- toms using NRS, in which 0 was no symptoms and 100 ticipants were also informed that the interventions had no the worst possible. Due to the large dispersion of spon- known relevant health risks under the applied conditions. taneous pain and subjective symptom measures in fibro- Special attention was paid to ensuring blinding. One myalgia patients [18, 28, 29], treatment effects on both blind researcher collected all outcome measures. Also, vibrotactile stimulation and sham were evaluated at the the assessment interviews were entirely structured and end of the study using the initial scores as a reference. the interaction between the data collector and patients To assist symptom rating, patients were instructed to was strictly limited to the questions of interest. No other firstly evaluate the effects of the intervention as a per- conversations were permitted during the interview. Prior centage change (improvement or worsening). Percentage to the study, patients formally agreed not to interact ratings were then transformed into absolute NRS values. with the researcher collecting outcome measures. An- Patients evaluated each treatment effect globally as sus- other researcher was available throughout the study to tained (> 1 week) improvement (or worsening) attained be consulted by patients in the event of doubt, adverse during each 3-week treatment period. effects, or any kind of incident. Assessment and moni- Secondary outcomes included the following: (i) “Unre- toring were carried out in different centers with no freshed Sleep,” patients rated the quality of sleep by indi- interaction during the evolution of the study. cating the severity (i.e., how much of a problem) of awakening tired or unrefreshed using 101-point NRS. (ii) Randomization “Daily Activity Impact,” the interference of fibromyalgia Patients were randomly assigned to the intervention se- in their daily life activities and quality was similarly quence (vibrotactile stimulation first or sham first) with quantified using 101-point numerical ratings. (iii) “Emo- an equal allocation ratio (1:1). The randomization scheme tional Distress,” assessed on the Hospital Anxiety and was generated by means of the Web site Randomization. Depression Scale (HADS) [30, 31] with separate rating com http://www.randomization.com. The system gener- for anxiety and depression. (iv) The Fibromyalgia Impact ated a unique number for each randomized patient. The Questionnaire (FIQ) [32], total score. Secondary out- randomization schedule was stratified in blocks of 8. A come measures were obtained both before and after single independent researcher was aware of the each treatment period. randomization process and randomization information (LBH). This researcher was in charge of telephoning each Sample size patient on the day to start the interventions to communi- We estimated that a total of 58 valid patients would be cate the device (real or sham) to be connected in each required to detect a difference between allocation groups phase. with a two-tailed α of 0.05, a “1-β” of 0.80, symptom Pujol et al. Arthritis Research & Therapy (2019) 21:148 Page 5 of 11 reduction standard deviation of 20 points (on 101-point rat- rating. To test intervention order effects, a between- ings), and symptom reduction difference of 15 points (rele- subject factor with two levels (sensory stimulation first or vant symptom reduction differences have been reported as sham first) was added to the repeated measures model. ranging from 10 to 20 in previous studies [33, 34]). If a po- Univariate analyses to compare intervention effects within tential case loss of 20% is assumed, our aim was to and between allocation groups were carried out using randomize a minimum of 73 patients (77 cases were finally paired t-test and two-sample t-test, respectively. randomized). Secondary outcomes were analyzed similarly. Partici- pants receiving at least 1 day of either study intervention Statistical analysis were included in adverse effect analyses. Proportions of The change regarding the corresponding baseline (pre participants with adverse effects were analyzed using minus post) was used as a measure of treatment re- Fisher’s exact test. All reported p values are two-tailed. sponse for pain, fatigue, and cognitive symptoms. Differ- All analyses were conducted using SPSS software, ver- ences between sensory stimulation and sham effects sion 20. were jointly tested for all three key symptoms using re- peated measures MANOVA. Within-subject effects were Results then estimated with the repeated measures model for Of the 77 patients randomized, 10 patients discontinued each individual (pain, fatigue, and cognitive symptoms) the study and 4 were excluded due to low treatment Fig. 2 CONSORT flow diagram Pujol et al. Arthritis Research & Therapy (2019) 21:148 Page 6 of 11 compliance. The final analyzed sample included 63 valid than for sham in pain and fatigue, but not in cognitive patients, of whom 30 belonged to the “sham first” group symptoms (Table 2, Fig. 3). and 33 to the “vibrotactile stimulation first” group (Fig. 2) Table 2 shows within and between comparisons for al- . Baseline demographic and clinical characteristics are location groups. A significant intervention order effect detailed in Table 1. was identified. That is, the difference between vibrotac- tile stimulation and sham for pain and fatigue was sig- Primary outcome nificantly greater in the “vibrotactile stimulation first” Repeated measures MANOVA including the variable group than in the “sham first” group (interaction “inter- change (pre minus post) in NRS ratings for pain, fatigue, vention by order” showing F = 8.8 and p = 0.004 for pain, and cognitive symptoms showed a significant within- and F = 10.7 and p = 0.002 for fatigue). subject overall effect (vibrotactile stimulation versus Figure 4 illustrates the intervention order effect for fa- sham) with F = 4.0 and p = 0.012. tigue ratings. There was a virtual absence of placebo ef- Significant differences between vibrotactile stimulation fect in the sham condition when administered in the and sham were also obtained for the percentage im- second period (n = 33; mean ± SD pre minus post NRS, provement rated for pain, fatigue, and cognitive symp- 3.5 ± 14.9; t = 1.3 and p = 0.190). Interestingly, the effect toms globally. Changes in the global symptom score for of sham when administered first was similar to the effect sham showed mean ± SD of 25.7% ± 30.7% and vibrotac- of vibrotactile stimulation when administered secondly tile stimulation 45.2% ± 33.9%, t = 3.4 and p = 0.001. (n = 30; NRS 12.0 ± 18.4 versus 13.0 ± 18.0, t = − 0.2 and p = 0.842). By contrast, the differences between both in- Fibromyalgia key symptoms terventions were notably large when vibrotactile stimula- Univariate contrasts for the three fibromyalgia key symp- tion was administered first (n = 33; NRS 27.6 ± 24.1 toms showed score change indicating a symptom im- versus 3.5 ± 14.9, t = 5.1 and p = 0.00001). provement significantly higher for vibrotactile stimulation Another carryover phenomenon was that the effect of vibrotactile stimulation, when administered first, did not Table 1 Clinical characteristics (n = 63) completely disappear at week 5 (i.e., 2 weeks after the Age, years, mean (SD) 53.7 (8.1) end of vibrotactile stimulation) (Table 2). For instance, Sex, females, n (%) 63 (100) NRS scores of fatigue at baseline showed a mean ± SD of Height, cm, mean (SD) 161 (6) 74.4 ± 12.5 and at week 5 a mean ± SD of 69.1 ± 16.3 Weight, kg, mean (SD) 69.2 (13.6) (n = 33; difference = 5.2; t = 3.0 and p = 0.005). By con- Hand-dominance, n (%) right-handers 56 (89) trast, differences in fatigue NRS ratings between baseline Education level, years, mean (SD) 14.1 (3.9) and week 5 were not significant in the “sham first” group (n = 30; baseline fatigue NRS scores 74.7 ± 14.3 versus Illness duration, mean (SD), years since diagnosis 7.7 (5.8) week 5 scores 73.1 ± 15.0; difference = 1.6; t = 1.8 and Fibromyalgia Impact Questionnaire*, mean (SD) 63.6 (15.1) p = 0.087). The interaction, however, was not significant General perception of health**, mean (SD) 33.1 (17.6) (F = 3.4 and p = 0.068). Hospital Anxiety and Depression Scale (HADS) Anxiety, mean (SD) 11.2 (4.1) Secondary outcomes Depression, mean (SD) 9.8 (4.4) The advantage of vibrotactile stimulation over sham was also significant for restoring sleep quality. That is, the Stable medication regime changes observed for the “Unrefreshed Sleep” variable NSAIDs, n (%) 29 (46) generally paralleled the results of pain and fatigue Opioids (tramadol), n (%) 24 (38) (Table 3). As for disorder impact, the effect of vibrotac- Paracetamol, n (%) 34 (54) tile stimulation had a non-significant tendency to be Gabapentin, n (%) 25 (40) higher (compared with sham) for both measures (Daily Antidepressants Activity Impact and Fibromyalgia Impact Questionnaire) . Vibrotactile stimulation had no significant effect on SSRI, n (%) 18 (29) emotional distress assessed on the Hospital Anxiety and SNRI, n (%) 20 (32) Depression Scale. Others, n (%) 14 (22) Anxiolytics/hypnotics (benzodiazepines), n (%) 42 (67) Additional “intent-to-treat” analysis *Fibromyalgia Impact Questionnaire (FIQ), maximum score, 100. **According Although the aim of our study was to reflect actual to the 36-Item Short-Form Health Survey, maximum score, 100. Several treatment differences, an additional analysis was carried patients were taking more than one medication. NSAIDs non-steroidal anti- inflammatory drugs, SSRI selective serotonin reuptake inhibitor, SNRI serotonin- out including all patients with available outcome mea- norepinephrine reuptake inhibitor sures (n = 67). The results were almost identical for all Pujol et al. Arthritis Research & Therapy (2019) 21:148 Page 7 of 11 Table 2 Fibromyalgia key symptoms All patients (n = 63) Baseline (Bvs) Vibrotac. S. (VS) Baseline (Bsham) Sham (Bvs − VS) > (Bsham − Sham) Pain 71.3 ± 13.1 54.4 ± 23.5 69.9 ± 15.1 62.5 ± 21.0 F = 6.5 p = 0.014 Fatigue 73.8 ± 13.7 53.1 ± 26.4 71.8 ± 15.5 64.3 ± 22.9 F = 12.0 p = 0.001 Cognitive symptoms 65.5 ± 18.1 62.0 ± 21.0 65.8 ± 18.2 64.1 ± 19.4 F = 1.6 p = 0.206 Vibrotactile S first (n = 33) Baseline (B1) Vibrotac. S. (VS) Baseline (B2) Sham (B1 − VS) > (B2 − Sham) Pain 71.2 ± 10.4 48.0 ± 26.1 66.4 ± 16.3 62.8 ± 20.0 t = 4.0 p = 0.0003 Fatigue 74.4 ± 12.5 46.8 ± 27.8 69.1 ± 16.3 65.7 ± 21.9 t = 5.1 p = 0.00001 Cognitive symptoms 66.5 ± 18.0 61.9 ± 22.4 66.4 ± 18.1 63.9 ± 20.3 t = 0.9 p = 0.359 Sham first (n = 30) Baseline (B1) Sham Baseline (B2) Vibrotac. S. (VS) (B2 − VS) > (B1 − Sham) Pain 73.7 ± 12.9 62.2 ± 22.4 71.3 ± 15.7 61.3 ± 18.3 t = − 0.3 p = 0.783 Fatigue 74.7 ± 14.3 62.7 ± 24.2 73.1 ± 15.0 60.1 ± 23.4 t = 0.2 p = 0.842 Cognitive symptoms 65.2 ± 18.6 64.4 ± 18.8 64.4 ± 18.5 62.1 ± 19.7 t = 0.9 p = 0.382 (B1 − VS) > (B1 − Sham) (B2 − VS) > (B2 − Sham) Pain t = 2.2 p = 0.034 t = 1.5 p = 0.149 Fatigue t = 2.9 p = 0.005 t = 2.3 p = 0.025 Cognitive symptoms t = 1.9 p = 0.071 t = − 0.1 p = 0.911 variables with the exception of Daily Activity Impact and analysis, both variables showed only a tendency to statis- Fibromyalgia Impact Questionnaire. Differences between tical significance (Table 3). vibrotactile stimulation and sham for Daily Activity Im- pact rating in this analysis showed t = 2.1 and p = 0.041, Adverse effects and Fibromyalgia Impact Questionnaire score t = 2.0 and Both vibrotactile stimulation and sham were generally p = 0.049 (with data missing from one patient in the well tolerated. 101-point NRS for discomfort showed a Fibromyalgia Impact Questionnaire), both in the direc- mean ± SD of 11.9 ± 23.6 in patients receiving vibrotac- tion of symptom improvement being higher for vibrotac- tile stimulation and 6.1 ± 21.9 when receiving sham (n = tile stimulation. It is noteworthy that, in the primary 67; t = 1.7 and p = 0.099). Two additional patients were Fig. 3 Bar graphs showing NRS score reduction for the three fibromyalgia key symptoms Pujol et al. Arthritis Research & Therapy (2019) 21:148 Page 8 of 11 Fig. 4 Bar graphs showing NRS score reduction for fatigue separately for each allocation group to illustrate the intervention order effect initially excluded due to excessive discomfort; one who The extent of symptom improvement may be sufficiently failed to tolerate vibrotactile stimulation and the other relevant to suggest a potential role for vibrotactile stimula- failing to tolerate sham (Fig. 2). Six patients required the tion as symptomatic treatment in fibromyalgia. Moreover, intensity of the stimulus to be reduced to 66% of the the ease with which it can be administered during sleep, prescribed power due to discomfort; in 4 cases while be- with no other action required than turning on the system ing treated with vibrotactile stimulation and in 2 cases at bed time, may eventually facilitate long-term compliance, with sham (χ2 = 0.5 and p = 0.674). Therefore, night-time which is a relevant limiting factor for the success in treating vibrotactile stimulation did not generally affect sleep in chronic disorders [35]. patients, but it did significantly improve sleep quality in We have proposed that non-nociceptive somatosensory the whole study group (Table 3). stimulation, here administered in the form of extensive and gentle mechanical vibrations, may favor the recovery Discussion of sensory balance in fibromyalgia. Our empirical results Vibrotactile stimulation was significantly superior to sham are indeed consistent with this hypothesis. However, the in alleviating fibromyalgia symptoms globally. However, effect of vibrotactile stimulation on the sensory system univariate analyses showed that the effect was not universal. has not been directly tested in this study. The evidence of Benefits were perceived on unpleasant somatic sensations central nervous system sensory alteration has been ob- such as generalized pain and fatigue, but not on poor cog- tained from functional connectivity and task activation nition, anxiety, and depression. Vibrotactile stimulation was MRI [5–10]. Further neuroimaging research is thus neces- notably well tolerated and sleep quality significantly im- sary to specifically test the effects of vibrotactile stimula- proved despite the vibrations being administered at night. tion on the sensory balance. Nonetheless, it is important Pujol et al. Arthritis Research & Therapy (2019) 21:148 Page 9 of 11 Table 3 Secondary outcome measures All patients (n = 63) Baseline (Bvs) Vibrotac. S. (VS) Baseline (Bsham) Sham (Bvs − VS) > (Bsham − Sham) Unrefreshed Sleep 69.6 ± 17.8 55.1 ± 22.5 67.1 ± 18.5 60.3 ± 23.4 t = 2.2 p = 0.033 Daily Activity Impact 70.6 ± 17.9 67.9 ± 16.9 69.7 ± 17.8 69.8 ± 18.6 t = 1.7 p = 0.098 Anxiety (HADS) 10.9 ± 4.4 10.1 ± 3.9 11.0 ± 3.9 10.9 ± 4.3 t = 1.4 p = 0.164 Depression (HADS) 9.3 ± 4.7 9.3 ± 4.7 9.3 ± 4.4 9.3 ± 4.6 t = − 0.6 p = 0.955 FIQ 61.9 ± 16.2 54.8 ± 17.3 60.7 ± 17.9 58.1 ± 18.0 t = 1.9 p = 0.069 Vibrotactile S. first (n = 33) Baseline (B1) Vibrotac. S. (VS) Baseline (B2) Sham (B1 − VS) > (B2 − Sham) Unrefreshed Sleep 70.8 ± 16.8 52.7 ± 23.8 61.9 ± 20.6 61.9 ± 24.2 t = 4.4 p = 0.0001 Daily Activity Impact 68.9 ± 19.4 64.8 ± 18.6 66.6 ± 19.0 67.7 ± 20.3 t = 2.4 p = 0.020 Anxiety (HADS) 10.7 ± 4.7 10.0 ± 3.8 10.5 ± 4.2 10.7 ± 4.4 t = 1.1 p = 0.261 Depression (HADS) 9.8 ± 4.8 9.4 ± 4.5 9.0 ± 4.8 9.6 ± 4.7 t = 1.5 p = 0.148 FIQ 63.1 ± 14.0 54.1 ± 18.2 57.5 ± 19.0 58.3 ± 18.7 t = 3.4 p = 0.002 Sham first (n = 30) Baseline (B1) Sham Baseline (B2) Vibrotac. S. (VS) (B2 − VS) > (B1 − Sham) Unrefreshed Sleep 72.8 ± 14.1 58.6 ± 22.7 68.4 ± 19.1 57.6 ± 21.0 t = − 0.6 p = 0.527 Daily Activity Impact 73.2 ± 15.9 71.1 ± 16.7 72.4 ± 16.2 71.3 ± 14.3 t = − 0.7 p = 0.502 Anxiety (HADS) 11.6 ± 3.4 11.2 ± 4.2 11.2 ± 4.2 10.2 ± 4.0 t = 0.8 p = 0.427 Depression (HADS) 9.8 ± 3.9 8.9 ± 4.6 8.8 ± 4.6 9.1 ± 4.9 t = − 1.4 p = 0.171 FIQ 64.2 ± 16.3 57.8 ± 17.5 60.7 ± 18.4 55.6 ± 16.5 t = − 0.4 p = 0.753 (B1 − VS) > (B1 − Sham) (B2 − VS) > (B2 − Sham) Unrefreshed Sleep t = 1.0 p = 0.336 t = 2.2 p = 0.029 Daily Activity Impact t = 1.4 p = 0.162 t = 1.3 p = 0.198 Anxiety (HADS) t = 0.3 p = 0.759 t = 1.8 p = 0.082 Depression (HADS) t = −0.6 p = 0.549 t = 0.5 p = 0.595 FIQ t = 0.7 p = 0.462 t = 1.6 p = 0.111 FIQ Fibromyalgia Impact Questionnaire, HADS Hospital Anxiety and Depression Scale to emphasize that symptom improvement in our study study adopted the strategy of using the same device to ad- was perceived at day time with no vibrations and that their minister both types of intervention, presented to patients as effects persisted 2 weeks after the treatment was termi- two potentially effective options. The paraphernalia sur- nated. The fact that the effects endured beyond stimula- rounding the interventions, implicating the home installa- tion may well indicate some functional rearrangement as tion of a motorized mattress controlled by a computerized opposed to a transient interference with pain signals at engine, was identical for both vibrotactile stimulation and the entry gate [36]. sham. The placebo effect in these circumstances was indeed Symptoms in the fibromyalgia syndrome are not inde- large and significant (e.g., first period fatigue reduction in pendent clinical expressions but are highly interrelated. the group sham first showed t = 3.6 and p = 0.001). How- For instance, chronic pain, as a stressful situation, may ever, the placebo effect was virtually nonexistent when favor fatigue and, conversely, fatigue may augment pain sham was administered in the second period. This is rele- perception. Moreover, unrefreshing sleep may potentiate vant in that it may indicate that the control of treatment ef- the feeling of both pain and fatigue. Sleep quality signifi- fects with our sham approach was not complete in this cantly improved in our study (Table 3). Therefore, there period, which is admittedly a limitation in our study. On is a possibility that sleep improvement was to some ex- the other hand, the data may also inform on the magnitude tent a primary driver of the improved pain and fatigue. of placebo response and its dynamics in such an intricate In turn, a significant degree of pain relief may well con- chronic pain disorder as fibromyalgia, and contributes to tribute to improving sleep quality. Future research may current efforts to improve the characterization of pla- interestingly be addressed to disentangle fibromyalgia cebo effects on pain [37, 38]. It is also noteworthy symptom interactions. that, in studies reporting the effect of nonpharmaco- Important methodological aspects in our study in- logical treatment for chronic pain, the most common cluded the control of therapeutic effects with strict sham comparison has been against “usual care” with no and the rigorous observation of blind conditions. Our control of the placebo effect [17]. Pujol et al. Arthritis Research & Therapy (2019) 21:148 Page 10 of 11 A limitation in our study, however, precisely concerns Ethics approval and consent to participate the carryover effects. Although a complete counterbalan- This study was approved by the Ethical Committee of Clinical Research of the Parc de Salut Mar of Barcelona (reference no. 2016/6932/I). All patients cing in our study prevents inflation effects of either provided written informed consent. treatment option, carryover did significantly influence the magnitude of change during the second period in Consent for publication two ways. Firstly, symptom change was attenuated in the Not applicable. second period, with the previously described lack of sig- nificant placebo effects, and, secondly, the effect of Competing interests The authors declare that they have no competing interests. vibrotactile stimulation persisted at least until week 5. Although this would interestingly suggest that the effect Author details 1 of vibrotactile stimulation in fibromyalgia is long-lasting, MRI Research Unit, Department of Radiology, Hospital del Mar, Passeig Marítim 25-29, 08003 Barcelona, Spain. 2Centro Investigación Biomédica en washout time should be better adjusted in future studies. Red de Salud Mental, CIBERSAM G21, Barcelona, Spain. 3Department of Another study limitation involves to the general prob- Project and Construction Engineering, Universitat Politècnica de Catalunya lems inherent to subjective symptom measurements. It (UPC), Barcelona, Spain. 4Rheumatology Department, Hospital del Mar, Barcelona, Spain. 5Department of Clinical and Health Psychology, is very difficult for people, particularly for patients with Autonomous University of Barcelona, Barcelona, Spain. chronic pain, to reliably rate the amount of perceived pain, or feelings in general [28, 29, 39]. We opted to as- Received: 19 December 2018 Accepted: 6 June 2019 sess outcome measures at the end of the study to facili- tate the comparison of treatment effects using baseline References scores as reference. 1. Wolfe F, Clauw DJ, Fitzcharles MA, Goldenberg DL, Katz RS, Mease P, Russell AS, Russell IJ, Winfield JB, Yunus MB. The American College of Conclusions Rheumatology preliminary diagnostic criteria for fibromyalgia and measurement of symptom severity. Arthritis Care Res (Hoboken). 2010;62: The effect of gentle vibrotactile stimulation of the body 600–10. on symptom relief was tested in the framework of a con- 2. Gracely RH, Ambrose KR. Neuroimaging of fibromyalgia. Best Pract Res Clin trolled clinical trial in fibromyalgia patients. Results Rheumatol. 2011;25:271–84. 3. López-Solà M, Woo CW, Pujol J, Deus J, Harrison BJ, Monfort J, Wager TD. showed significant reduction of pain and fatigue, and Towards a neurophysiological signature for fibromyalgia. Pain. 2017;158:34–47. sleep quality improved despite the fact that stimulation 4. Napadow V, Harris RE. What has functional connectivity and chemical was applied during sleep time. The degree of improve- neuroimaging in fibromyalgia taught us about the mechanisms and management of ‘centralized’ pain? Arthritis Res Ther. 2014;16:425. ment and the easy application of our proposal would 5. López-Solà M, Pujol J, Wager TD, Garcia-Fontanals A, Blanco-Hinojo L, seem to be sufficiently relevant to suggest a potential Garcia-Blanco S, Poca-Dias V, Harrison BJ, Contreras-Rodríguez O, Monfort J, role for vibrotactile stimulation in the treatment of Garcia-Fructuoso F, Deus J. Altered functional magnetic resonance imaging responses to nonpainful sensory stimulation in fibromyalgia patients. fibromyalgia symptoms. Arthritis Rheumatol. 2014;66:3200–9. 6. Pujol J, Macià D, Garcia-Fontanals A, Blanco-Hinojo L, López-Solà M, Garcia- Abbreviations Blanco S, Poca-Dias V, Harrison BJ, Contreras-Rodríguez O, Monfort J, Garcia- FIQ: Fibromyalgia Impact Questionnaire; HADS: Hospital Anxiety and Fructuoso F, Deus J. The contribution of sensory system functional Depression Scale; MANOVA: Multivariate analysis of variance; NRS: Numerical connectivity reduction to clinical pain in fibromyalgia. Pain. 2014;155:1492– rating scales 503. 7. Cifre I, Sitges C, Fraiman D, Muñoz MÁ, Balenzuela P, González-Roldán A, Acknowledgements Martínez-Jauand M, Birbaumer N, Chialvo DR, Montoya P. Disrupted We thank the Agency of University and Research Funding Management of functional connectivity of the pain network in fibromyalgia. Psychosom the Catalonia Government for their participation in the context of Research Med. 2012;74:55–62. Groups SGR2017-0134 and SGR2017-1198. Dr. Monfort is a member of the 8. Craggs JG, Staud R, Robinson ME, Perlstein WM, Price DD. Effective Scientific Advisory Board of Pfizer. connectivity among brain regions associated with slow temporal summation of C-fiber-evoked pain in fibromyalgia patients and healthy Authors’ contributions controls. J Pain. 2012;13:390–400. JB, JM, and JD carried out patient recruitment and selection. GP and HO 9. Flodin P, Martinsen S, Mannerkorpi K, Löfgren M, Bileviciute-Ljungar I, Kosek developed the stimulation system prototype and performed vibration E, Fransson P. Normalization of aberrant resting state functional connectivity exposure measurements. DRL carried out the acquisition of outcome in fibromyalgia patients following a three month physical exercise therapy. measures. GMV contributed to data acquisition and study coordination. LBH Neuroimage Clin. 2015;9:134–9. performed the randomization process and treatment order allocation. JP 10. Kim J, Loggia ML, Cahalan CM, Harris RE, Nat BFDP, Garcia RG, Kim H, Wasan analyzed and interpreted the data and primarily drafted the manuscript. All AD, Edwards RR, Napadow V. The somatosensory link in fibromyalgia: authors contributed the study design and drafting and read and approved functional connectivity of the primary somatosensory cortex is altered by the final manuscript. sustained pain and is associated with clinical/autonomic dysfunction. Arthritis Rheumatol. 2015;67:1395–405. Funding 11. Soriano-Maldonado A, Ruiz JR, Aparicio VA, Estévez-López F, Segura-Jiménez This work was supported in part by the Ministry of Economy, Industry and V, Álvarez-Gallardo IC, Carbonell-Baeza A, Delgado-Fernández M, Ortega FB. Competitiveness of Spain (Grant PSI2017-83777-P). The funder has no role in Association of physical fitness with pain in women with fibromyalgia: the al- this study. Ándalus project. Arthritis Care Res (Hoboken). 2015;67:1561–70. 12. Stranden M, Solvin H, Fors EA, Getz L, Helvik AS. Are persons with Availability of data and materials fibromyalgia or other musculoskeletal pain more likely to report hearing Not applicable. loss? A HUNT study. BMC Musculoskelet Disord. 2016;17:477. Pujol et al. Arthritis Research & Therapy (2019) 21:148 Page 11 of 11 13. Clauw DJ. Fibromyalgia and related conditions. Mayo Clin Proc. 2015;90: 35. Vermeire E, Hearnshaw H, Van Royen P, Denekens J. Patient adherence to 680–92. treatment: three decades of research. A comprehensive review. J Clin Pharm 14. Dailey DL, Rakel BA, Vance CG, Liebano RE, Amrit AS, Bush HM, Lee KS, Lee Ther. 2001;26:331–42. JE, Sluka KA. Transcutaneous electrical nerve stimulation reduces pain, 36. Melzack R, Wall PD. Pain mechanisms: a new theory. Science. 1965;150:971–9. fatigue and hyperalgesia while restoring central inhibition in primary 37. Wager TD, Atlas LY. The neuroscience of placebo effects: connecting fibromyalgia. Pain. 2013;154:2554–62. context, learning and health. Nat Rev Neurosci. 2015l;16:403–18. 15. Macfarlane GJ, Kronisch C, Dean LE, Atzeni F, Häuser W, Fluß E, Choy E, 38. Zunhammer M, Bingel U, Wager TD. Placebo imaging consortium. Placebo Kosek E, Amris K, Branco J, Dincer F, Leino-Arjas P, Longley K, McCarthy GM, effects on the neurologic pain signature: a meta-analysis of individual Makri S, Perrot S, Sarzi-Puttini P, Taylor A, Jones GT. EULAR revised participant functional magnetic resonance imaging data. JAMA Neurol. recommendations for the management of fibromyalgia. Ann Rheum Dis. 2018. https://doi.org/10.1001/jamaneurol.2018.2017. 2017;76:318–28. 39. Bartley EJ, Robinson ME, Staud R. Pain and fatigue variability patterns 16. Plazier M, Ost J, Stassijns G, De Ridder D, Vanneste S. C2 nerve field distinguish subgroups of fibromyalgia patients. J Pain. 2018;19:372–81. stimulation for the treatment of fibromyalgia: a prospective, double-blind, randomized, controlled cross-over study. Brain Stimul. 2015;8:751–7. 17. Skelly AC, Chou R, Dettori JR, Turner JA, Friedly JL, Rundell SD, Fu R, Brodt Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in ED, Wasson N, Winter C, Ferguson AJR. Noninvasive nonpharmacological published maps and institutional affiliations. treatment for chronic pain: a systematic review. Rockville: Agency for Healthcare Research and Quality (US); 2018. Report No.: 18-EHC013-EF 18. Derry S, Wiffen PJ, Häuser W, Mücke M, Tölle TR, Bell RF, Moore RA. Oral nonsteroidal anti-inflammatory drugs for fibromyalgia in adults. Cochrane Database Syst Rev. 2017;3:CD012332. 19. Kandel ER, Schwartz JH, Jessell TM, Siegelbaum SA, Hudspeth AJ, editors. Principles of neural science. 5th ed. New York: McGraw-Hill; 2012. 20. Staud R, Robinson ME, Goldman CT, Price DD. Attenuation of experimental pain by vibro-tactile stimulation in patients with chronic local or widespread musculoskeletal pain. Eur J Pain. 2011;15:836–42. 21. Alentorn-Geli E, Padilla J, Moras G, Lázaro Haro C, Fernández-Solà J. Six weeks of whole-body vibration exercise improves pain and fatigue in women with fibromyalgia. J Altern Complement Med. 2008;14:975–81. 22. Alev A, Mihriban A, Bilge E, Ayça E, Merve K, Şeyma C, Uğur E, Adnan B, Zeynel K, Mahmut GS. Effects of whole body vibration therapy in pain, function and depression of the patients with fibromyalgia. Complement Ther Clin Pract. 2017;28:200–3. 23. Bidonde J, Busch AJ, van der Spuy I, Tupper S, Kim SY, Boden C. Whole body vibration exercise training for fibromyalgia. Cochrane Database Syst Rev. 2017;9:CD011755. 24. Olivares PR, Gusi N, Parraca JA, Adsuar JC, Del Pozo-Cruz B. Tilting whole body vibration improves quality of life in women with fibromyalgia: a randomized controlled trial. J Altern Complement Med. 2011;17:723–8. 25. Sañudo B, de Hoyo M, Carrasco L, McVeigh JG, Corral J, Cabeza R, Rodríguez C, Oliva A. The effect of 6-week exercise programme and whole body vibration on strength and quality of life in women with fibromyalgia: a randomised study. Clin Exp Rheumatol. 2010;28(6 Suppl 63):S40–5. 26. Altman DG, Schulz KF, Moher D, Egger M, Davidoff F, Elbourne D, Gøtzsche PC, Lang T. CONSORT GROUP (consolidated standards of reporting trials). The revised CONSORT statement for reporting randomized trials: explanation and elaboration. Ann Intern Med. 2001;134:663–94. 27. Wolfe F, Smythe HA, Yunus MB, Bennett RM, Bombardier C, Goldenberg DL, Tugwell P, Campbell SM, Abeles M, Clark P. The American College of Rheumatology 1990 Criteria for the Classification of Fibromyalgia. Report of the Multicenter Criteria Committee. Arthritis Rheum. 1990;33:160–72. 28. Harris RE, Williams DA, McLean SA, Sen A, Hufford M, Gendreau RM, Gracely RH, Clauw DJ. Characterization and consequences of pain variability in individuals with fibromyalgia. Arthritis Rheum. 2005;52:3670–4. 29. Stussman BJ, Nahin RL, Čeko M. Fibromyalgia patients and healthy volunteers express difficulties and variability in rating experimental pain: a qualitative study. Scand J Pain. 2018;18:657–66. 30. Quintana JM, Padierna A, Esteban C, Arostegui I, Bilbao A, Ruiz I. Evaluation of the psychometric characteristics of the Spanish version of the hospital anxiety and depression scale. Acta Psychiatr Scand. 2003;107:216–21. 31. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand. 1983;67:361–70. 32. Burckhardt CS, Clark SR, Bennett RM. The fibromyalgia impact questionnaire: development and validation. J Rheumatol. 1991;18:728–33. 33. Dworkin RH, Turk DC, McDermott MP, Peirce-Sandner S, Burke LB, Cowan P, Farrar JT, Hertz S, Raja SN, Rappaport BA, Rauschkolb C, Sampaio C. Interpreting the clinical importance of group differences in chronic pain clinical trials: IMMPACT recommendations. Pain. 2009;146:238–44. 34. Dworkin RH, Turk DC, Wyrwich KW, Beaton D, Cleeland CS, Farrar JT, et al. Interpreting the clinical importance of treatment outcomes in chronic pain clinical trials: IMMPACT recommendations. J Pain. 2008;9:105–21.