The effect of the EXOPULSE Mollii Suit on pain and fibromyalgia-related symptoms-A randomized sham-controlled crossover trial.

J.G.M., M.A.C, M.S., J.P.L., S.S.A: Design and conceptualization of the study. M.A.C. and J.P.L: Statistical analysis and visualization. J.G.M: Data collection. G.N.A.L: Data curation. S.S.A: Supervision and project administration. J.G.M., M.A.C, N.O., J.L.G., J.P.L, G.N.A.L, S.S.A: Data interpretation. J.G.M., M.A.C, S.S.A: Writing‐ Original draft preparation. J.G.M, M.A.C., N.O, M.S., J.L.G., J.P.L., G.N.A.L., S.S.A: Writing‐ Reviewing and Editing. All authors read, commented on, and approved the final version of the manuscript.

The Sponsor ( “promoteur” ) was the Institut de la Colonne Vertébrale et des Neurosciences de Paris (ICVNS‐Paris, France). This work received no external funding.

The study consisted of two phases (Figure  1 ). The first one is a randomized, sham‐controlled, double‐blind, cross‐over study, in which each participant was randomly assigned to receive daily 1 hour active, or sham, stimulation. Stimulation conditions lasted 2 weeks each and were separated by a two‐week washout interval. Randomization was performed by a research technician using a randomization list from the online tool Research Randomizer ( randomizer.org ). Patients were randomized 1:1 to one or the other intervention sequence, with no further stratification criteria. The randomization list was blinded from anyone involved in informing potential study participants. The treatment allocation codes (referring to the stimulation conditions) were then put into sealed envelopes which remained closed until the end of the study. A research technician, trained to use EXOPULSE Mollii Suit, and not involved in any other step of the protocol, programmed control units for active and sham stimulation, both being strictly identical in their aspect and use, allowing double‐blinding, both for investigators as well as participants. Study design. During active and sham intervention periods, patients had daily 1 h sessions with the EXOPULSE Mollii Suit at home. The second phase is an open‐label phase, that started at least 2 weeks following the end of the last stimulation session of phase 1, meaning that a washout interval of at least 2 weeks was respected between both phases. All participants received daily one‐hour active stimulation sessions for four consecutive weeks. The EXOPULSE Mollii Suit (EXONEURAL NETWORK AB, Danderyd, Sweden) is made of two combined CE‐marked medical devices: (i) body garments (jacket and pants, class I according to Regulation EU 2017/745), which contain embedded electrodes, conductive wires, and connectors, and (ii) a control unit (active class IIa), which is a battery‐powered electrical device, that delivers low energy electric pulses through connectors to the body garments (Figure  S1 ). The suit comes in 37 sizes (ranging from 104 cm up to 5XL) for women and men. The whole system provides non‐invasive electro‐stimulation, to key nerves and corresponding muscle groups, throughout the body (Figure  S2 ). As stated previously, during both two‐week intervention periods, the participants were instructed to use the medical device 1 h per day and to perform the stimulation at rest. During active sessions, the device delivered an electric current, with a constant voltage of 20 V and a frequency of 20 Hz. During the sham sessions, the device delivered the same electric current for 1 minute only, then switched off automatically, enabling cutaneous sensations, that mimicked the active stimulation and maintained blinding. Patients can only turn the device on and off, but they cannot change the stimulation parameters. Patients received a training session on EXOPULSE Mollii Suit usage, with instructions (paper and electronic format) provided as backup. Afterward, they were given the suit, with a control unit programmed for active or sham intervention, according to the sequence allocation, to perform home‐based daily sessions for two consecutive weeks. Patients were instructed to return the material (suit and control unit) at the end of the first stimulation period. Following a washout interval of at least 2 weeks, they were given another set of materials (suit and control unit), to be returned at the end of the second stimulation period. Finally, an EXOPULSE Mollii Suit, and an active control unit, were provided for the open‐label phase. The study took place at the Clinical Neurophysiology Department of Henri Mondor Hospital (Créteil, France). Eligible patients were adults (18–75 years old), with a definite diagnosis of fibromyalgia, according to the American College of Rheumatology (ACR) 2010 criteria (Wolfe et al.,  2010 ), for at least 1 month before inclusion, and suffered from pain, whose intensity was ≥4 on a visual analog scale (VAS pain ) over the last week before inclusion (Jensen et al.,  2003 ). Participants could not be included if they (i) had medical contraindications to wearing the suit (e.g., uncontrolled epilepsy, arrhythmias, cardiac stimulator, a ventriculoperitoneal shunt, intrathecal baclofen pump, pregnancy, and/or body mass index above 35 kg/m 2 ), (ii) were suffering from other somatic diseases (i.e., other diseases causing osteoarticular and muscular pain) or psychiatric pathology (other than anxiety and depression), (iii) were included in another biomedical research protocol, (iv) were unable to submit to the medical follow‐up of the study for geographical or social reasons, (v) or had changed their pharmacological treatment over the last 3 months. Pregnant women were also not eligible for inclusion. Participants were screened for eligibility by one of the study investigators, who provided oral and written information about the study protocol. After a reflection period of at least 1 week, following the screening visit, patients were called by the investigator to confirm their willingness to participate in the study, then schedule baseline visits to finalize inclusion (i.e., verification of eligibility criteria, informed consent signature, baseline measures, and randomization). Apart from the screening visit, the protocol consisted of six visits: one at the beginning and one at the end of each of the three intervention periods (i.e., active and sham periods of phase 1 and the open‐label phase 2). Patients did not receive any compensation (financial or medical device) for their participation. They returned the medical device in question at the end of the study. Clinical and demographic data were collected from each participant, including age, sex, BMI, disease duration, pharmacological treatments, non‐pharmacological approaches, the Widespread Pain Index (WPI) which measures the number of painful body regions (range: 0–19) and the Symptom Severity (SS) scale which assesses the severity of three symptoms (fatigue, waking unrefreshed, and cognitive symptoms) from 0‐no problem to 3‐severe, and the extent of somatic symptoms in general from 0‐no symptoms to 3‐a great deal of symptoms. The SS scale score sums up these symptoms (range: 0 to 12). The study outcomes were assessed using questionnaires and scales that are validated in French, previously employed in French cohorts, and with good psychometric properties. The primary objective of this study was to evaluate the short‐term impact of EXOPULSE Mollii Suit on pain in adult patients with fibromyalgia. We hypothesized that using this device would reduce pain, as per VAS pain measured at the end of the intervention periods of the randomized phase. VAS pain consists of a 10 cm horizontal line over which the patient could place a mark between 0 (none) and 10 (unbearable) (Hayes & Patterson,  1921 ; Jensen et al.,  2003 ; Perrot et al.,  2010 ). Secondary objectives aim to assess the effects of EXOPULSE Mollii Suit on fatigue, anxiety, depression, and quality of life, assuming a clinical benefit. The Brief Pain Inventory (BPI) was used to evaluate pain severity (4 items) and pain interference (7 items) (Brasseur,  1997 ). Each BPI item is rated on a 10‐point scale ranging from 0 (no pain/does not interfere) to 10 (pain as bad as you can imagine/interferes completely) (Cleeland & Ryan,  1994 ; Sullivan et al.,  1995 ). The Pain Catastrophizing Scale (PCS) was employed to assess the presence and severity of feelings or thoughts that emerge when experiencing pain (13‐item scale, each rated on a 5‐point Likert scale ranging from 0—not at all to 4—all the time) (French et al.,  2005 ). VAS fatigue , a 10 cm horizontal line used like VAS pain , was adopted to evaluate the effects of this intervention on fatigue (Lee et al.,  1991 ; Perrot et al.,  2010 ). The Fibromyalgia Impact Questionnaire (FIQ) was also used. It is an 11‐item scale that assesses health status and functional disability, by exploring the impact of fibromyalgia on work, well‐being, fatigue, sleep, stiffness, anxiety, and depression (Burckhardt et al.,  1991 ; Perrot et al.,  2003 ). FIQ total score and subscores range from 0 to 100, with higher scores reflecting worse health status. The Hospital Anxiety and Depression Scale (HADS) was used to assess anxiety (7 items) and depression (7 items) (Bocéréan & Dupret,  2014 ). Scores range from 0 to 21 on each subscale, with higher scores indicating worse symptomatology. Quality of life was assessed via the Short Form 36 health survey (SF‐36), which provides scores for eight dimensions (physical functioning, role‐physical, role‐emotional, bodily pain, general health, vitality, social functioning, and mental health), and a remaining item on the perception of health change. Each score ranges from 0 to 100, with a higher score implying better health status (Perneger et al.,  1995 ; Ware & Gandek,  1998 ). In order to account for potential day‐to‐day pain fluctuation, patients daily filled in VAS pain over 1 week prior to each stimulation condition (average daily pain reporting at the end of the day over 1 week pre‐sham and pre‐active of phase 1 and 1 week pre‐open label phase) and throughout the stimulation conditions (days 1–14 of each condition in phase 1, days 1–28 of phase 2). This yielded an average VAS pain measure. In addition, to test the acute effects of the intervention on VAS pain , the scale was completed before and after the first session of active and sham interventions. In both phases, all the remaining questionnaires (except VAS pain ) were completed before and after each intervention period (before the first stimulation session and at the end of each condition of phase 1, before the first session and at the end of phase 2). In addition, the Clinical Global Impression of Change (CGI‐C) was evaluated after each intervention period (Le Gal et al.,  2010 ). It consists of a 7‐point scale, ranging from 1‐“very much improved since the initiation of treatment” to 7‐“very much worse since the initiation of treatment” (Busner & Targum,  2007 ). Improvement was considered from 1 to 3, worsening from 5 to 7, and no change if 4. Participants were also asked to guess the type of stimulation received, active or sham at the end of each stimulation condition of phase 1, to assess blinding integrity, and were rated as follows: correct guess, wrong guess, and unable to guess. Considering the pilot nature of this work and the lack of previous studies applying EXOPULSE Mollii Suit in patients with fibromyalgia, sample size calculation did not rely on previous data. Sample size calculation was performed using G*Power Software (version 3.1.9.6, Faul et al.,  2007 ). When adopting the following parameters (medium effect size of 0.25, two‐tailed significant difference of α  = 0.05, estimated power of 80%) and taking into account the risk of dropouts, a sample size of 34 was considered. Data were collected using a case report form (investigator) and self‐questionnaires (patient), then entered in an electronic database (Excel file) for further statistical analysis performed using IBM SPSS Statistics for Windows (Version 29.0.2.0 Armonk, NY: IBM Corp). All endpoints collected during the randomized phase were compared, considering the intervention (active versus sham), and the time points (baseline versus end of each period). For the open‐label phase, long‐term effects were assessed, by comparing data at the start and end of the active intervention period. Since quantitative data did not follow a normal distribution (Shapiro–Wilk test), comparisons for the randomized phase were performed using Friedman's tests, using a covariate with four categories: active pre‐intervention, active post‐intervention, sham pre‐intervention, and sham post‐intervention. These were followed by post‐hoc Dunn's tests with a Bonferroni p‐value adjustment. In order to check if the patients who guessed the sham condition had different results from those who did not, the percentage of improvement for the primary outcome (VAS pain ) was calculated following active and following sham conditions, and the percentage of improvement of the sham condition was subtracted from the active condition as previously described in the literature (Lefaucheur et al.,  2011 ). % VASpain active − sham = 100 x VASpreactive − VASpostactive VASpreactive −( 100 x VASpresham − VASpostsham VASpresham ). % VASpain active − sham was compared between patients who guessed or not the sham condition. In a similar manner, % VASpain active − sham was compared between patients who received at least one treatment versus those who did not receive any medical treatment, as well as between those who received nonpharmacological interventions versus those who did not. The same analysis was repeated for each medication and nonpharmacological category using Mann–Whitney test when appropriate (i.e., analysis applied in the groups with at least 15% of the sample size under a specific medication or nonpharmacological therapy). Wilcoxon signed‐rank tests were used for the open‐label phase. Significance was set at 0.05. The estimation of effect size was based on Kendall's coefficient of concordance W (randomized phase) and r = Z/√N (open‐label phase). Categorial endpoints (CGI‐C, and blinding integrity) were compared in phase 1 (active versus sham conditions) with the Chi‐2 test. For all analyses, the effect size was classified as small (<0.3), moderate (≥0.3 and <0.5) or large (≥0.5). Quantitative variables are described with mean ± standard deviation. This clinical trial was prospectively registered on clinicaltrials.gov as ‘EXOPULSE Mollii Suit and Fibromyalgia (EXOFIB)’ ( NCT 05361577 ). The study was conducted in accordance with the ISO 14155 standard (Clinical investigation of medical devices for human subjects – Good clinical practice), and the European Union Regulation on medical devices (2017/745). The protocol was approved by an independent ethics committee (COMITÉ DE PROTECTION DES PERSONNES « EST IV ») on November 5, 2021. Following French legislation and regulatory requirements, the approval was sent for information to the French health authority ( Agence Nationale de sécurité du médicament et des produits de santé , Saint‐Denis, France) before the inclusion of the first participant. Written informed consent was obtained from all participants before enrollment into the study. Financial compensation was provided only for travel costs to study visits.

Overall, 42 patients were screened initially. After the screening visits, 34 patients were recruited. The study took place between March 1, 2022, and July 31, 2023. The participation flow chart is presented in Figure  2 . Participation flow chart. Thirty‐three patients completed both study phases ( n  = 1 dropout in the sham condition). 31 (93.90%) were female, and the mean age was 51.33 ± 8.99 years. The mean BMI was 26.55 ± 5.42 Kg/m 2 . At the time of inclusion, disease duration was 8.94 ± 10.74 years. The WPI was 14.15 ± 3.36 and the SS scale score was 8.0 ± 2.38. All patients had diffuse pain, and only two reported predominant pain in specific body regions (the lower limbs ( n  = 1), left hemibody ( n  = 1)). No patients had pain‐related surgery. There was no missing data. The patients had the following controlled medical comorbidities: arterial hypertension ( n  = 8), migraine ( n  = 8), tension headache ( n  = 2), diabetes mellitus type 2 (n = 2), thyroid disease ( n  = 6), asthma ( n  = 2), polycystic ovary syndrome ( n  = 2), obstructive sleep apnea ( n  = 2), glaucoma ( n  = 1), atopic dermatitis ( n  = 1), fatty liver disease ( n  = 1), hepatitis B ( n  = 1) and endometriosis ( n  = 1). With regards to medication profile, 84.85% of patients received at least 1 pharmacological treatment and only 15.15% did not receive any medication. Patients were receiving antiepileptics (n = 6), antidepressants ( n  = 18), anxiolytics ( n  = 6), opioids analgesics ( n  = 5), combined opioids and acetaminophen medications ( n  = 10), anti‐inflammatory ( n  = 9), acetaminophen ( n  = 13), nefopam (n = 5), baclofen (n = 2), lidocaine transdermal patch ( n  = 4), and cannabinoids ( n  = 3). Physical therapy ( n  = 20) and other nonpharmacological methods [hypnosis ( n  = 4), physical exercise ( n  = 4), auriculotherapy ( n  = 2), yoga and meditation ( n  = 2), osteopathy ( n  = 2), acupuncture ( n  = 1), and music therapy ( n  = 1)] were also adopted. There were no missing days of treatment based on patients' reporting. Blinding integrity was preserved, as the rates of guessing the intervention type did not differ between active and sham conditions ( X 2  = 0.792; p =  0.673): wrong guess (18.20% vs. 12.10%; respectively), correct guess (48.50% vs. 45.50%, respectively), and unable to guess (33.33% vs. 42.4%, respectively). When adopting forced guessing for patients who were unable to guess, the differences remained nonsignificant ( X 2  = 0.580; p  = 0.447): wrong guess (42.40% in active vs. 33.33% in sham) and correct guess (57.60% vs. 66.67%, respectively). When comparing % VASpain active − sham between who guessed or not the sham condition (correct guess 45.59% vs. wrong guess 12.10% vs. unable to guess 42.40%), Kruskal‐Wallis test was nonsignificant ( p =  0.593). The same results were obtained when running Mann–Whitney test on data obtained with forced guessing (correct guess 66.67% vs. wrong guess 33.33%, p =  0.721). The CGI‐C distribution was significantly different between the active and sham intervention periods ( p  = 0.035): clinical worsening was reported for 0% versus 18.18% of participants ( p   0.05), and no change for 36.36% versus 33.33% of participants ( p  > 0.05) respectively, for active versus sham stimulation. Results from the randomized phase for efficacy endpoints are detailed in Table  1 . With regards to the primary endpoint (VAS pain ), Friedman's test of differences was statistically significant ( p =  0.014). The posthoc Dunn's test revealed a significant decrease in VAS pain after the active intervention period (pre‐active VAS pain : 6.85 ± 1.36; post‐active VAS pain : 5.91 ± 1.83, Dunn's p  = 0.029) (Figure  3 ). No significant difference was observed between pre‐and post‐VAS pain for the sham intervention, nor between the active and sham interventions for pre‐VAS pain (Dunn's p =  1.000). The results were also significant right after the first active session compared to sham (Friedman's test p  = 0.001; VAS pain before the first active session: 6.59 ± 2.13; VAS pain after the first active session: 4.91 ± 2.33, Dunn's p  = 0.003). Summary of efficacy endpoints before and after each two‐week intervention period (randomized phase). Note : Data are mean ± standard deviation. Abbreviations: BPI, brief pain inventory; FIQ, fibromyalgia impact questionnaire; HADS, hospital anxiety and depression scale; PCS, pain catastrophizing scale; SF‐36, short‐form 36 health survey; VAS, visual analog scale. p ‐value from Friedman's test comparing all four interventions (active/sham)*time(pre−/post‐) values. Kendall's W. Bolded values represent significant p ‐values (<0.05) for both Friedman's and post‐hoc Dunn's tests. Some of Friedman's test p values were significant, but post hoc analysis did not reveal significant differences between pre‐and post‐active intervention. # Baseline scores comparison with post‐hoc Bonferroni correction: p  = 0.052. Pain scores before and after active and sham periods of stimulation (randomized phase). VAS, Visual Analog Scale (10 cm scale). * p  = 0.014 for Friedman's test (global test) and 0.029 for the post‐hoc Dunn's test (pre‐active versus post‐active values). No statistically significant difference in the percentage of improvement ( % VASpain active − sham ) between the patients who were receiveing or not pharmacological treatments (treated vs. untreated). Similarly, the percentage of improvement in VAS pain did not significantly differ between patients who were receiving alternative interventions compared to those who did not. In addition, a significant reduction in pain was also observed following the active intervention (pre‐ versus post‐values) when assessed via the BPI, FIQ, and SF‐36 (BPI pain interference , FIQ pain SF‐36 bodily pain , respectively). However, no change was found in PCS total , or in its subscales, which assess cognition related to pain (helplessness, rumination, and magnification). Fatigue also significantly decreased following the active, but not the sham, intervention periods, as per FIQ fatigue , FIQ rested , and SF‐36 vitality , but not VAS fatigue . Similarly, a statistically significant (or almost significant) difference after active intervention periods only was found with regards to some anxiety‐related endpoints (i.e., FIQ anxiety , but not HADS anxiety ) and to disease impact (FIQ total , FIQ physical impairment , FIQ stiffness ). Conversely, depression‐related endpoints (HADS depression , FIQ depression ) were not significantly modified by any intervention. Lastly, no significant effects were observed on the remaining quality of life dimensions (SF‐36 scores for physical functioning, role‐physical, role‐emotional, mental health, social functioning, general health, and health change). For all endpoints that showed a benefit of the active intervention, the effect size was small, ranging from 0.08 to 0.23. The most important effect sizes were observed for SF‐36 pain scores (0.23) and FIQ total (0.20). The whole results are represented in Figure  S3 . Results for efficacy endpoints from the open‐label phase are detailed in Table  2 . Improvement was reported for 78.8% of patients at the end of the 4‐week active intervention period, as per the CGI‐C. Following 1 month of intervention, the Wilcoxon signed rank tests yielded statistically significant effects for most of the study outcomes: VAS pain , VAS fatigue , BPI total and both BPI subscales (pain severity and interference), PCS total and all PCS subscales (rumination, magnification, and helplessness), HADS anxiety , HADS depression , FIQ total and most FIQ subscales (physical impairment, feel good, pain, fatigue, rested, stiffness, and anxiety), and most of SF‐36 scores (physical functioning, role‐physical, role‐emotional, vitality, mental health, social functioning, bodily pain, and health change). Conversely, SF‐36 general health and FIQ depression did not change significantly following intervention. The effect sizes ranged from 0.25 (small) to 0.54 (large), with the highest reported for VAS pain . Figure  S4 provides a graphical representation of the whole results. Summary of efficacy endpoints before and after the 4‐week active intervention period (open phase). Note : Data are mean ± standard deviation. Abbreviations: BPI, brief pain inventory; FIQ, fibromyalgia impact questionnaire; HADS, hospital anxiety and depression scale; PCS, pain catastrophizing scale; SF‐36, short‐form 36 health survey; VAS, visual analog scale. p‐value from the Wilcoxon signed rank test comparing pre−/post‐intervention values. (Z/√N). Bolded values represent endpoints that are statistically significantly different ( p  < 0.05) prior to and after the active intervention. All the stimulation sessions were well tolerated, and no serious adverse events were reported at any time.

The present study is the first randomized controlled trial carried out to investigate the efficacy of EXOPULSE Mollii Suit in patients with fibromyalgia. It is also the first study to explore mid‐ and long‐term treatment, made of repeated daily sessions (i.e., 2 weeks during the randomized phase, and 4 weeks in the open‐label phase). This design suggests analgesic effects of the therapy, as reflected by the improvement in the primary outcome (VAS pain ) after 2 weeks of active intervention compared to sham. The present results are in line with previous data from studies on the acute effects of a single EXOPULSE Mollii Suit session in fibromyalgia (Riachi et al.,  2023 ; Rubio‐Zarapuz et al.,  2023 , 2024 ). The results of both phases converge to confirm the analgesic potential of EXOPULSE Mollii Suit, which is potent after two weeks and appears strengthened when the treatment lasts a longer time (i.e., four weeks). The mean VAS pain decreased by almost 1 cm (on a 10 cm scale) after two weeks, and by 1.7 cm after four weeks of daily active sessions. Although we did not find in the literature any assessment of the minimal clinically important difference for this criterion in the context of fibromyalgia, the second change appears clinically relevant, by analogy with chronic musculoskeletal pain, measured on a numerical rating scale (Fleagle et al.,  2024 ; Salaffi et al.,  2004 ). VAS has good psychometric properties for assessing pain (Campbell & Lewis,  1990 ; Kahl & Cleland,  2005 ) a high sensibility and sensitivity, and a good discriminative power in patients with fibromyalgia (80%, 80% and 0.864, respectively) (Marques et al.,  2008 ). Investigating the acute effects of EXOPULSE Mollii Suit on 50 patients with fibromyalgia, Riachi et al. found a change in VAS pain immediately after a single one‐hour session, and the change was still significant 24 hours later (Riachi et al.,  2023 ). Furthermore, our results suggest that the multisite stimulation delivered by EXOPULSE Mollii Suit also improved fatigue during the controlled phase, and more during the open‐label phase. Analgesic and antifatigue effects were associated with an improvement in fibromyalgia‐related stiffness and anxiety and a better perception of health change. Depression scores, some pain‐induced behaviours, and several quality‐of‐life domains did not significantly improve during the randomized phase. Still, most of them did after the open‐label phase of the study. Comparing the higher number of significantly improved endpointsobserved during the four‐week intervention to those from the two‐week intervention, it can be inferred that applying a long‐term treatment might help increase the clinical improvement. It can also be hypothesized that longer treatment periods would durably improve mood disorders and quality of life. However, it is worth stating that, in the absence of a control condition, the results of the open‐label phase should be interpreted with caution as it is not possible to rule out sham effects. A multimodal treatment approach for fibromyalgia is needed since pathophysiological mechanisms interact with the psychological ability of individuals to cope with this chronic disease in a long‐term perspective, and since no isolated strategy has so far proven its effectiveness (Sarzi‐Puttini et al.,  2020 ). Non‐invasive electrotherapy techniques provide promising results in alleviating pain while causing negligible side effects (Coskun Benlidayi,  2020 ). Further research on EXOPULSE Mollii Suit protocols is required to optimize the benefits of treatment, together with the acceptability and adherence of patients: duration and frequency of sessions, overall duration, daily schedule, association with pharmacological or alternative strategies, adjustments based on individual's response, etc. (Perpetuini et al.,  2023 ). The protocol chosen in this trial—one‐hour daily sessions—was very well tolerated. Future studies on larger samples and with longer‐term treatment should be planned to guide the therapeutic use of EXOPULSE Mollii Suit in fibromyalgia. The mechanisms involved in the analgesic efficacy of this method remain to be elucidated. To start, transcutaneous electrical stimulation could exert effects at the peripheral (impulse blockade), segmental (spinal), and extra‐segmental (supraspinal descending inhibition) levels (Johnson,  2021 ). The rationale for using transcutaneous electrical stimulation for pain relief primarily refers to gate control theory, according to which, stimulation of large proprioceptive fibres would inhibit the nociceptive information transmitted by small fibres (Melzack & Wall,  1965 ). In this work, EXOPULSE Mollii Suit effects could have resulted from this mechanism. In addition, its impact may have arisen from other peripheral mechanisms (e.g., involving muscular, vascular, and immune factors). For instance, a case report and a randomized controlled trial performed on fibromyalgia found that a unique one‐hour session of this device triggered a drastic increase in muscle oxygenation documented using near‐infrared spectroscopy (a device that measures muscle oxygen saturation (SmO2), total haemoglobin (tHb), oxygenated haemoglobin (O2Hb) and deoxygenated haemoglobin (HHb)) (Rubio‐Zarapuz et al.,  2023 , 2024 ). Indeed, the impairment of muscle oxygen utilization could lead to muscular fatigue, and reduced exercise tolerance, as observed in fibromyalgia (Shang et al.,  2012 ), while pain could result from muscle ischemia and local vasoconstriction (Katz et al.,  2007 ). Also, while oxidative stress has been incriminated in fibromyalgia pathophysiology, antioxidative strategies might have some benefit in this condition (Assavarittirong et al.,  2022 ). Therefore, EXOPULSE Mollii Suit might improve fibromyalgia pain and fatigue by reversing the abnormalities mentioned above. Moreover, previous works have suggested a reduction in the level of pro‐inflammatory cytokines (i.e., interleukin‐6) in patients with pain, which might also apply in the case of EXOPULSE Mollii Suit and fibromyalgia (Johnson,  2021 ). Besides peripheral mechanisms, EXOPULSE Mollii Suit might have effects at the central level, as found in previous studies evaluating transcutaneous electrical nerve stimulation. For instance, in some works, transcutaneous electrical nerve stimulation resulted in changes in several metabolites in the cerebrospinal fluid (glutamate, aspartate, enkephalins, and endorphins) and cortical electrical waves (Ong Sio et al.,  2023 ). Finally, the effects of EXOPULSE Mollii Suit, on other fibromyalgia symptoms and impact, could be discussed and explained in the light of the biopsychosocial model of fibromyalgia, which consists of sensory, immune, emotional, cognitive, and social factors, among others (Clauw et al.,  2024 ; Popkirov et al.,  2020 ). The model implies bidirectional interaction among these factors involved in pain processing, perception, psychobehavioral, and social factors. The effects of EXOPULSE Mollii Suit might be exerted on one or several components of this model. For instance, analgesic effects per se (as per pain scales) might subsequently result in, or be associated with, less fatigue, less anxiety/stress, more ability to engage in social activities, and less disease impact. This is in line with the positive results observed in the first phase of the study, and the additional improvement noted following a longer treatment period, as shown in the open‐label phase of the trial. When discussing pain, it is important to mention the sensory (nociceptive), cognitive, and emotional components of its so‐called matrix (Bushnell et al.,  2013 ; Garcia‐Larrea & Peyron,  2013 ). While the nociceptive component has improved following phase 1, pain‐related cognitive processes (i.e., catastrophizing) and emotional experience (i.e., namely depression) might require more time to reach statistically significant improvement, as observed at the end of phase 2. Here, it is noteworthy that some authors have applied functional brain neuroimaging and raised the hypothesis supporting the influence of catastrophizing on pain perception, via the modulation of cognitive and emotional responses to pain (Gracely et al.,  2004 ). This was supported by an association between pain catastrophizing and increased activity in cerebral areas, involved in pain attention and anticipation (frontal, cerebellar, cingular regions), and emotional aspect of pain (claustrum) (Gracely et al.,  2004 ). In this context, EXOPULSE Mollii Suit might have yielded delayed effects on catastrophizing and depressive symptoms, which in turn, might have further increased the analgesic response and quality of life. This could be reflected by the additional clinical improvement reported at the end of phase 2 in terms of pain scores, clinical global impression of change, and quality of life. While this study was the first randomized controlled trial carried out on the EXOPULSE Mollii Suit, some limitations can be highlighted. For instance, the trial was conducted in a single center, so the global care of patients cannot be extrapolated to the general population. In addition, the study is performed in a relatively small sample and over only two weeks per stimulation condition and four weeks in the open‐label phase. Therefore, the current results merit to be replicated in larger cohorts and a longer follow‐up duration. In addition, as previously stated, the one cannot rule out a potential placebo effects in the open‐label phase. Also, only 6.1% of males were included in the cohort, which corresponds to the rate of males referred with fibromyalgia in clinical practice. Still, it is far lower than data from unbiased studies (>40%), which suggests a dramatic underdiagnosis of men (Wolfe et al.,  2018 ). Moreover, although the blinding test yielded nonsignificant results between active and sham, some patients were able to guess the type of stimulation. Furthermore, there was no tracking system to monitor how long the suit was being used per day, or whether the stimulation duration was limited to once per day. Therefore, future studies would benefit from replicating the results with a trackable and controllable version of the medical device. However, despite the mentioned limitations, it is important also to note that the recruited cohort consisted of patients with chronic fibromyalgia (mean and median disease duration: 8.9 and 5 years, respectively) and who were already receiving pharmacological and non‐pharmacological interventions (i.e., analgesics, antidepressants, anxiolytics, physical therapy, and other alternative interventions). This could emphasize the potential beneficial effects of EXOPULSE Mollii Suit in patients with chronic fibromyalgia, suffering from insufficient response to the available management options.

In conclusion, we observed the benefit of daily one‐hour sessions of EXOPULSE Mollii Suit to alleviate pain and related‐symptoms, in adult patients with fibromyalgia, after 2 weeks of intervention. This strategy appears promising, in the context of debilitating and difficult‐to‐manage diseases, such as fibromyalgia. Its potential utility in the management of fibromyalgia symptoms merits further exploration.

Fibromyalgia is a multifactorial and complex syndrome characterized by chronic and widespread pain affecting the musculoskeletal system, pressure sensitivity, and a low threshold to noxious stimuli (Fillingim et al.,  2014 ; Treede et al.,  2015 ). Besides pain, patients could suffer from several symptoms, including fatigue, sleep disturbance, as well as somatic, affective, and cognitive symptoms, and an altered quality of life (Burwinkle et al.,  2005 ; Wolfe et al.,  2014 , 2015 ). The aetiology of fibromyalgia is far from being uncovered. The interplay between some factors might be incriminated such as genetic predisposition, immune components, stressful life events, oxidative stress, neurotransmission, central sensitization, as well as emotional and cognitive mechanisms (Clauw et al.,  2024 ; Pinto et al., 2023 ). Several central mechanisms have been proposed: a hyperactive nociceptive system, and/or a defective antinociceptive pain inhibitory system, possibly due to a GABAergic/glutamatergic imbalance (Schmidt‐Wilcke & Clauw,  2011 ; Schmidt‐Wilcke & Diers,  2017 ), and/or implication of the serotoninergic and dopaminergic systems (Okifuji & Hare,  2013 ; Wolfe et al.,  2014 ). Peripheral mechanisms have also been suggested, such as reduced blood flow, muscle hypoxia, and metabolic defects at the muscular level (Okifuji & Hare,  2013 ). Despite the available diagnostic criteria (Macfarlane et al.,  2017 ; Wolfe et al.,  2010 ), setting the diagnosis remains a clinical challenge (Häuser et al.,  2019 ). Its prevalence range from 0.2 to 6.6% and seems to increase with female sex, increasing age, comorbidities, and a family history of this condition (Heidari et al.,  2017 ; Marques et al.,  2017 ; Queiroz,  2013 ; Wolfe et al.,  2013 ). Fibromyalgia ranks in the third position among the most frequent musculoskeletal conditions (Sarzi‐Puttini et al.,  2020 ), rendering its management a public health concern. More than 40 molecules have been tested and resulted in different benefits and side‐effects profiles (Schmidt‐Wilcke & Clauw,  2011 ; Schmidt‐Wilcke & Diers,  2017 ). Current approved pharmacological treatments consist of antidepressants and antiepileptic drugs, but monotherapy rarely yields satisfactory management, with only 10%–25% of patients exhibiting meaningful pain reduction (Johnson et al.,  2017 ; Taylor et al., 2019 ). In this context, a wide range of non‐pharmacological methods have also been proposed. Here, transcutaneous Electrical Nerve Stimulation (TENS) yielded promising results according to recent reviews and meta‐analyses (Coskun Benlidayi,  2020 ; García‐López et al.,  2024 ). However, ‘classical’ TENS devices can only target a limited number of body sites (two to four sites), which does not address the widespread pain encountered in fibromyalgia. To overcome this issue, a multisite device might be of help in this context. EXOPULSE Mollii Suit, a full‐body garment (jacket and pants) with 58 integrated electrodes, can transcutaneously stimulate 40 body sites, using a low‐intensity electric current. EXOPULSE Mollii Suit appears to have promising effects when applied over a single one‐hour session, in a few preliminary reports (Riachi et al.,  2023 ; Rubio‐Zarapuz et al.,  2023 , 2024 ). However, randomized controlled trials, with repeated sessions, are needed, to draw formal conclusions on the specific effects of this device, on pain and other symptoms, in patients suffering from this disease, as well as to assess the long‐term effects of this intervention. This was the aim of the current study.

SA declares having received compensation from Sanofi Aventis, France; Novartis, France; Exoneural Network AB, Sweden and Ottobock, France. MC declares having received compensation from Janssen Global Services LLC, Exoneural Network AB, Sweden, and Ottobock, France. The remaining authors declare no conflicts of interest.

Data S1.