Open access Protocol Study protocol for a randomised, double-­blinded, placebo-­controlled phase III trial examining the add-­on efficacy, cost–utility and neurobiological effects of low-­dose naltrexone (LDN) in patients with fibromyalgia (INNOVA study) Ariadna Colomer-­Carbonell,1,2 Juan P Sanabria-­Mazo,1,2 Halbert Hernández-­Negrín,2,3 Xavier Borràs ‍ ‍,2 Carlos Suso-­Ribera,4,5 Azucena García-­Palacios,4,5 Jordi Muchart,6 Josep Munuera,6 Francesco D'Amico,7 Michael Maes,8 Jarred W Younger,9 Albert Feliu-­Soler,10 Antoni Rozadilla-­Sacanell,11 Juan V Luciano ‍ ‍1,10 To cite: Colomer-­Carbonell A, Sanabria-­Mazo JP, Hernández-­ Negrín H, et al. Study protocol for a randomised, double-­ ABSTRACT blinded, placebo-­controlled Strengths and limitations of this study Introduction There is evidence that low-­dose naltrexone phase III trial examining the (LDN; <5.0 mg/day) reduces pain and improves the add-­on efficacy, cost–utility ► This is thought to be the first randomised, double-­ quality of life of people with fibromyalgia syndrome (FMS). and neurobiological effects blinded, placebo-­controlled phase III trial to assess However, no randomised controlled trials with long-­term the efficacy, safety, cost–utility and neurobiological of low-­dose naltrexone (LDN) in patients with fibromyalgia follow-­ups have been carried out. The INNOVA study effects of low-­dose naltrexone (LDN) for reducing (INNOVA study). BMJ Open will evaluate the add-­on efficacy, safety, cost–utility and pain in patients with fibromyalgia syndrome. 2022;12:e055351. doi:10.1136/ neurobiological effects of LDN for reducing pain in patients ► The INNOVA protocol combines mobile-­technology-­ bmjopen-2021-055351 with FMS, with a 1-­year follow-­up. based ecological momentary assessment and as- Methods and analysis A single-­site, prospective, sessment with classical legacy measures to obtain ► Prepublication history for this paper is available online. randomised, double-­blinded, placebo-­controlled, parallel more precise information on the dynamics of the To view these files, please visit design phase III trial will be performed. Eligibility criteria assessed primary and secondary outcomes. the journal online (http://dx.doi.​ include being adult, having a diagnosis of FMS and ► This study will include immune and neuroimaging org/10.1136/bmjopen-2021-​ experiencing pain of 4 or higher on a 10-­point numerical biomarkers in order to explore the neurobiological 055351). rating scale. Participants will be randomised to a LDN underpinnings of LDN. intervention group (4.5 mg/day) or to a placebo control ► Challenges of the randomised controlled trial include AC-­C, JPS-­M and HH-­N are joint group. Clinical assessments will be performed at baseline the long follow-­up period (1 year) and potential mea- first authors. AF-­S, AR-­S and JVL are joint (T0), 3 months (T1), 6 months (T2) and 12 months (T3). The surement burden that takes the risk of high dropout senior authors. primary endpoint will be pain intensity. A sample size of 60 rate. patients per study arm (120 in total), as calculated prior to Received 09 July 2021 recruitment for sufficient power, will be monitored between Accepted 03 December 2021 January 2022 and August 2024. Assessment will also peer-­reviewed journals, conference presentations, social include daily ecological momentary evaluations of FMS-­ media and community engagement activities. related symptoms (eg, pain intensity, fatigue and sleep Trial registration number NCT04739995. © Author(s) (or their disturbance), and side effects via ecological momentary employer(s)) 2022. Re-­use assessment through the Pain Monitor app during the first INTRODUCTION permitted under CC BY-­NC. No 3 months. Costs and quality-­adjusted life years will be commercial re-­use. See rights Fibromyalgia: definition, prevalence and also calculated. Half of the participants in each arm will and permissions. Published by be scanned with MRI at T0 and T1 for changes in brain pharmaceutical indications BMJ. Fibromyalgia syndrome (FMS) is a chronic metabolites related to neuroinflammation and central For numbered affiliations see sensitisation. Inflammatory biomarkers in serum will also condition of unknown origin that is charac- end of article. terised by generalised musculoskeletal pain, be measured. Correspondence to Ethics and dissemination This study has been approved fatigue, stiffness, cognitive problems, sleep Dr Juan V Luciano; by the Ethics Committee of the Fundació Sant Joan de disturbances and malaise.1 2 This syndrome ​juanvicente.​luciano@​uab.​cat Déu. The results will be actively disseminated through is highly prevalent in the general population Colomer-­Carbonell A, et al. BMJ Open 2022;12:e055351. doi:10.1136/bmjopen-2021-055351 1 Open access (2.7% worldwide).3 Around 6% of adult patients who visit with FMS, which gives support to the role of neuroin- their general practitioner, and between 10% and 20% flammation in the aetiology of FMS. In addition, there is of those who visit rheumatology services, have FMS.2 In evidence that chronic stress facilitates the ‘priming’ and 2007, the U.S. Food and Drug Administration approved exaggerated activation of the microglia.28 pregabalin as the first drug indicated for the treatment Although FMS is not considered a classic inflamma- of FMS, and later approved duloxetine and milnacipran tory disease, there is extensive evidence that immune for this indication. However, the European regulatory pathways play a significant role in the pathogenesis and authorities rejected the indication of these three drugs maintenance of the syndrome. Cytokines play a key role in the treatment of FMS given the small effect sizes in in inflammatory response and in boosting the nocicep- various studies and the associated adverse effects.4–6 tive response due to their sensitisation actions, both on a peripheral and central level.29–32 Thus, there is evidence Pathogenesis of FMS that FMS involves an imbalance in pro-­inflammatory [eg, Although the aetiological factors of FMS are not known, interleukin (IL)-­1, IL-­6, IL-­17A and the tumour necrosis the primary hypothesis of the pathogenesis of this factor α (TNF-α)] and anti-­inflammatory (eg, IL-­4 and syndrome highlights the role of the central nervous IL-­10) levels of cytokines that could lead to a low-­intensity, system in the amplification of pain perception as well chronic state of inflammation. Bäckryd et al30 identified as in the development of comorbid symptoms such as both neuroinflammation and systemic inflammation by sleep-­related problems, fatigue, cognitive difficulties and evaluating levels of a broad panel of cytokines and chemo- emotional distress.1 7 8 Structural brain alterations have kines in cerebrospinal fluid and plasma. also been found in patients with FMS. For example, lower volumes of grey matter have been observed in areas asso- Low-dose naltrexone (LDN): a promising treatment for FMS ciated with the processing of stress (eg, parahippocampal Naltrexone is an opioid antagonist medication used to gyrus) and pain (eg, anterior cingulate cortex, insula, treat opioid and alcohol dependency. The drug blocks prefrontal cortex, and primary and secondary somato- mu-­opioid receptors, the delta-­opioid receptors and, to a sensory cortices).7 Functional MRI alterations have been lesser extent, the kappa-­opioid receptors. There is prom- associated with self-­reported pain intensity. Brain activity ising evidence to suggest that naltrexone administered in exhibited greater connectivity between different pain-­ low doses (ie, LDN; <5 mg/day) is effective in the manage- processing areas (eg, insula and secondary somatosensory ment of some pathologies which present with altered cortex), the default-­model network in persons with FMS, immune pathways, such as Crohn’s disease, multiple scle- as well as in the association between these areas and the rosis or FMS.33 34 The immune-­regulatory effect of LDN pain levels reported by patients and the right executive seems to be driven through the inhibition of the Toll-­like attention network.9 receptor 4 (TLR-­4) activity expressed in the membrane Altered functional connectivity has also been reported of various immune system cells (eg, microglia and macro- among various pain-­inhibiting areas.10 Some studies have phages).33 Moreover, due to a ‘rebound effect’, LDN could reported reduced levels of neurotransmitters involved exert an analgesic effect that strengthens the endogenous in the regulation of the descending analgesic response opioid pain inhibitory system. According to this hypoth- (serotonin and norepinephrine) and increased levels of esis, the low-­intensity and intermittent blockade of the glutamate (Glu) and substance P in people with FMS.11–14 opioid receptors generated by LDN induces a compensa- For example, high levels of Glu have been reported in the tory mechanism that facilitates an increase in the produc- posterior insula, posterior cingulate cortex and prefrontal tion of endogenous opioids and greater sensitivity of the ventrolateral cortex of patients with FMS when compared system to their effects.33 34 with healthy controls.15–19 To date, the effects of LDN in patients with FMS have only Activation of the microglia could be a contributing been evaluated through crossover pilot studies that have factor to the alteration of Glu neurotransmission in yielded preliminary results. In the first study conducted FMS.20 21 The microglia is normally found in a state of with LDN in FMS, significant reductions in pain, stress rest but it is activated by a wide range of stimuli such and fatigue levels were observed.35 In a subsequent study, as cell death, peripheral inflammation, chronic stress significant improvements in daily pain, satisfaction with and infections.22 Once activated, microglia release pro-­ life and mood were also observed.36 In another crossover inflammatory agents such as cytokines, excitatory amino investigation, the pre and post changes in the levels of acids and nitric oxide.23 These inflammatory factors plasma cytokines were evaluated over 8 weeks. Significant across multiple neural pathways can induce hyperal- reductions in a wide range of immune biomarkers were gesia, fatigue, depression and other symptoms which obtained (eg, IL-­1β, sIL-­1ra, IL-­4, IL-­6, IL-­10, IL-­17A and are known collectively as ‘cytokine induced sickness TNF-α), together with a reduction in the pain levels and behavior’.24 25 Microglia activation might trigger a series the severity of FMS symptoms.37 of actions that lead to an increase in Glu that ultimately While acknowledging the contribution of past studies results in synaptic dysfunction.26 A recent study using into the field, these have included small sample sizes positron emission tomography27 of translocator protein (n=8–31 participants) and crossover designs. Therefore, revealed a widespread cortical glial activation in patients a single-­site, prospective, randomised, double-­ blinded, 2 Colomer-­Carbonell A, et al. BMJ Open 2022;12:e055351. doi:10.1136/bmjopen-2021-055351 Open access placebo-­ controlled study (RCT) with a sufficiently the last 3 months; diagnosis of severe medical/psychiatric powered sample is presented here to conduct a meth- disorders (eg, cancer, haematological diseases, abnormal odologically robust investigation into the role of LDN hepatic/liver function, renal failure, suicide ideation, in FMS. Specifically, the main objective of the INNOVA psychotic disorder); pregnant (or planning to become study is to evaluate the efficacy, safety and cost–utility, and pregnant during the study period) or breast feeding; neurobiological effects of LDN to reduce pain in FMS. known allergy to naltrexone, naloxone or excipients; There is currently no gold standard pharmacological currently participating in other RCTs; ongoing litigation treatment for pain reduction in persons with FMS. There- related to FMS. fore, in the present study, placebo will be used instead of another drug in the control group. Additional selection criteria for the biomarkers and neuroimaging substudy (50% of patients in each study arm) All participants will meet the following inclusion criteria: METHODS AND ANALYSIS right-­handed (for the neuroimaging tests); and no Trial design comorbid rheumatologic conditions (eg, rheumatoid The randomised controlled trial (RCT) protocol has arthritis, lupus). Potential participants will be excluded been developed following the Standard Protocol Items: according to the following exclusion criteria: fever Recommendations for Interventional Trials.38 In addi- (>38°C); infection in the last 2 weeks; vaccination in the tion, the RCT was approved by European Union Drug last month; taking cortisone or anticytokine therapy; Regulating Authorities Clinical Trials (EudraCT; 2021-­ needle phobia or claustrophobia, metal implants or pace- 002534-­16). For reporting purposes, we will follow the makers; body mass index ≥36 kg/m2; smoking over 5 ciga- guidelines of the Consolidated Standards of Reporting rettes/day; presence of acute pain (eg, headache or back Trials39 and the Consolidated Health Economic Eval- pain) unrelated to FMS on the day of the scan. uation Reporting Standards statement.40 INNOVA is a 12-­month double-­blind RCT with two arms: LDN versus Recruitment strategy, procedure and randomisation placebo. LDN will be considered an add-­on treatment to Patients diagnosed with FMS with an appointment at the the usual care provided in the Spanish National Health Rheumatology Service of Parc Sanitari Sant Joan de Déu System for FMS. For transparency and analytical repro- (St. Boi de Llobregat, Spain) will be invited to participate ducibility purposes, the dataset and data coding will be in the study and will be asked to attend a screening eval- deposited in the Open Science Framework. uation with a research assistant and a clinician. Once the informed consent is obtained, the clinician will review Sample size the study selection criteria to confirm eligibility. The There are no previous RCTs about the efficacy of LDN week after, a face-­to-­face assessment (T0) including clin- for FMS; therefore, we estimated the sample size taking ical history and variables related to the use of services into account a previous LDN crossover study36 that had will be conducted with those patients meeting all the used self-­reported pain as main outcome (the effect size eligibility criteria. Only the participants that are included was d=0.99). Thus, with a sample of 60 participants per in the biomarkers substudy will require an additional arm, we aim to detect between-­group differences with a blood extraction and neuroimaging scan, which will be significance level of 5% and a power of 80%. Allowing for performed in the following 3–5 days. Participants will be a potential attrition rate of 20%, our final sample size is given a sealed envelope with an identifying code which they 60 participants per group. For the analysis of biomarkers will have to take to the pharmacy service. There, they will (involving 50% of the sample), an initial sample size of be given the dose of the corresponding drug (according 30 patients per arm is considered sufficient according to to the randomisation) for the first 3 months (90 tablets). previous studies.37 41 As shown in the patients’ flow chart (figure 1), further in-­person evaluations will be performed at 3 months (T1), Eligibility criteria 6 months (T2) and 12 months (T3). Neurobiological vari- General selection criteria ables will be obtained at T1 using the same protocol as in All participants will meet the following inclusion criteria: the baseline assessment (T0). women between 18 and 70 years; diagnoses of FMS Participants will be asked to abstain from taking any according to American College of Rheumatology (ACR) analgesic or anti-­inflammatory drug in the 72 hours prior 2016 criteria42 by a rheumatologist; pain intensity ranked to the blood extractions/scans. All patients, including ≥4 out of 10 on a 10-­point numerical rating scale in the those who do not participate in the biomarker substudy, past week; fluent in Spanish; provision of written informed shall be subjected to a blood test at baseline. Participants consent; stable pharmacological treatment in the last will return to the pharmacy service every 3 months and 2 months; and having a smartphone with android oper- will be given the assigned amount of LDN/placebo for the ating system for ecological momentary assessment (EMA). following 3 months (approximately 90 tablets). Uncon- Potential participants will be excluded according to the sumed tablets will be returned for treatment adherence following exclusion criteria: treatment with naltrexone, monitoring. The randomisation to conditions will be opioids, anticoagulants or central antihypertensives in conducted by a biostatistician from the Clinical Trials Colomer-­Carbonell A, et al. BMJ Open 2022;12:e055351. doi:10.1136/bmjopen-2021-055351 3 Open access Placebo The control group will take the placebo daily for 12 months (a film-­coated tablet identical to the LDN and filled with a lactose-­free filler). For the control arm, the same guidelines will be followed. In order to maintain the external validity of the study and for ethical reasons, the study participants’ active treat- ments will be unchanged by this clinical trial. In Spain, chronic pain management is mainly managed by general practitioners in regular consultations. These generally consist of face-­to-­face appointments with a duration of 5–10 min in which the clinicians monitor the physical and, ideally, the emotional status of the patient. General practitioners usually provide advice prescribe pharmaco- therapy (pain killers, hypnotics, antidepressants, etc) or refer patients to pain units in tertiary hospitals when more specialised pain management procedures are required. The frequency of consultations is based on the type of disease and its severity. In this study, usual care will be the same as in routine daily practice, without any modifi- cations. In addition, participants will be allowed to with- draw from this study for any reason at any time without detriment to the provision or quality of their usual care. If a severe adverse event (AE) occurs, unblinding will be possible and study participation will be discontinued. If the adverse effects are tolerable, the treatment will be administered until the end of the study. All these events will be recorded and reported at the end of the study. Figure 1 Flow chart of the INNOVA study based on the Study measures Consolidated Standards of Reporting Trials guidelines. FMS, fibromyalgia syndrome; ITT, intention-­to-­treat; LDN, low-­dose All participants will be assessed with a computer-­ naltrexone. administered battery of measures using the REDCap soft- ware (see table 1). Unit of Fundació Sant Joan de Déu who has no involve- Measures for sociodemographic characteristics, clinical ment in the eligibility screening, enrolment and treat- features and screening ment processes. The computer-­generated randomisation A sociodemographic questionnaire will be used to obtain will apply a permuted block design to ensure that the information about the following variables: gender, date study arms are balanced taking the biomarkers substudy of birth, marital status, living arrangements, educational eligibility criteria into account. As this is a double-­blind level, income level and employment status. RCT, neither the patient nor the evaluator or the clini- The Clinical data interview will be used to collect infor- cian will know to which treatment arm each patient has mation about history and duration of FMS symptoms, as been assigned. Only the clinical trial pharmacist who well as family history of medical/mental illness. Infor- stores and delivers capsules, but is not involved in patient mation regarding comorbidity with other diagnosed care, will know the allocation. physical-­psychiatric conditions and the type and dose of current drugs will be checked from medical records. Data management, central monitoring and audit The Fibromyalgia Survey Diagnostic Criteria43 44 is a 6-­item The clinical data entry, data management and central self-­report measure of the core FMS symptoms according monitoring will be performed with REDCap. An inde- to the latest revision of the ACR 2016 criteria.42 It includes pendent Clinical Research Organization (CRO) will be two subscales: the Widespread Pain Index, which is used responsible for overseeing the intrastudy data sharing and to identify the presence of pain in 19 body areas in the last storing process. Any modifications in the study protocol week, and the Symptom Severity Scale, in which the three will be communicated to the CRO. major FMS symptoms (fatigue, ‘fibrofog’ and waking up Treatments tired) are assessed along with three additional symptoms Low-dose naltrexone (pain in the lower stomach, depression and headache). A The intervention group will take one 4.5 mg naltrexone total score is obtained by adding the two subscales. This tablet (lactose-­free) daily for 12 months before going to total score ranges from 0 to 31, where higher values indi- sleep. cate greater FMS severity. 4 Colomer-­Carbonell A, et al. BMJ Open 2022;12:e055351. doi:10.1136/bmjopen-2021-055351 Open access Table 1 Time points for data collection Measures T0 (baseline) T1 (3-­m) T2 (6-­m) T3 (12-­m) Sociodemographic, clinical and screening measures  Sociodemographic data X        Clinical data (years of evolution, comorbidities, etc) X        FSDC screening and secondary outcome measure X X X X (fibromyalginess) Primary outcome measure  NRS (pain intensity) X X X X Secondary outcome measures  FIQR (functional impairment) X X X X  DASS-­21 (anxiety, depression, and stress) X X X X  MISCI (subjective cognitive impairment) X X X X  WHODAS 2.0 (disability) X X X X  GAD-­7 (general anxiety/worry) X X X X Other measures          EQ-­5D-­5L (quality of life) X     X  CSRI (medication consumption and service receipt) X     X  PGIC and PSIC (impression of change)   X X X  ACTTION checklist (adverse events throughout the trial) X X X X  Pain Monitor app (EMA) X X     Physiological variables  Immune biomarkers X X      Neuroimaging X X     ACTTION checklist, Analgesic, Anesthesic, and Addiction Clinical Trial Translations, Innovations, Opportunities, and Networks; CSRI, Client Service Receipt Inventory; DASS-­21, Depression Anxiety Stress Scales-­21; EMA, ecological momentary assessment; EQ-­5D-­5L, EuroQoL 5 Dimension 5 Level; FIQR, Fibromyalgia Impact Questionnaire Revised; FSDC, Fibromyalgia Survey Diagnostic Criteria; GAD-­7, Generalised Anxiety Disorder 7-­Item Scale; MISCI, Multidimensional Inventory of Subjective Cognitive Impairment; NRS, Numerical Pain Rating Scale; PGIC and PSIC, Patient Global Impression of Change and Pain Specific Impression of Change; WHODAS 2.0, 12-­item interviewer administered version of the WHO Disability Assessment Schedule 2.0. Primary outcome measure impairment. The Spanish version of the FIQR has The Numeric Rating Scale45 is a unidimensional measure of obtained high internal consistency estimates (α=0.91– pain intensity mainly used for adults. The most frequently 0.95), adequate test–retest reliability indices (r=0.82) used version is an 11-­point numeric scale (a horizontal and good construct validity.47 bar or line) scored from 0 (‘no pain’) to 10 (‘worst pain 21 (DASS-­21)48 is The Depression Anxiety Stress Scales-­ imaginable’). Time frames vary between studies. In the a self-­ report scale developed to discriminate between present study, respondents will be asked to report average features of depression (anhedonia/low positive affect), pain intensity over the last week. anxiety (physical arousal) and stress (psychological tension/agitation) in clinical and non-­clinical samples. Secondary outcome measures The DASS has been validated in patients with FMS.49 The Revised Fibromyalgia Impact Questionnaire (FIQR)46 includes 21 items that are answered on a 0 to 10 numer- Responders are required to indicate the presence of a ical scale in which higher scores indicate greater func- symptom over the previous week. Each item is scored tional impairment. The questionnaire asks about the from 0 (‘did not apply to me at all over the last week’) previous 7 days. The items are distributed into three to 3 (‘applied to me very much or most of the time over domains: physical impairment, overall impact and the past week’). There are seven items on each of the severity of symptoms (ie, pain, energy, stiffness, sleep three subscales: depression, anxiety and stress. Therefore, quality, depression, memory issues, anxiety, pain to the total scores in each scale can range from 0 to 21. Higher touch, balance problems and increased sensitivity to scores indicate more severe levels of depression, anxiety noises, lights, smells or temperatures). A total score is and stress. The Spanish version showed adequate internal obtained by summing the three subscale scores. This consistency for depression (α=0.84), anxiety (α=0.70) can range from 0 to 100. Higher scores indicate greater and stress (α=0.82).50 Colomer-­Carbonell A, et al. BMJ Open 2022;12:e055351. doi:10.1136/bmjopen-2021-055351 5 Open access The Multidimensional Inventory of Subjective Cognitive patients are asked about the total appointments for acci- Impairment (MISCI)51 is a 10-­item self-­report measure of dent and emergency services, total number of general subjective cognitive dysfunction (ie, fibrofog) in FMS. inpatient hospital admissions, number of diagnostic tests Each item is scored from 1 (‘never’) to 5 (‘very often’) administered and total appointments with healthcare and the total score ranges from 10 to 50. Lower scores professionals for pain management (family physicians, indicate higher cognitive dysfunction. The MISCI showed nurses, social workers, psychologists, psychiatrists, group excellent internal reliability, low ceiling/floor effects and psychotherapy and other community healthcare profes- good convergent validity with a similar measure. The sionals). The CSRI will be administered on two occasions: Spanish version of the MISCI had sound psychometric at baseline and at 12-­month follow-­up, both referring to properties (α=0.91 and intraclass correlation coefficient, the previous 12 months. Medical records will be checked ICC=0.88).52 to verify the accuracy of the collected data. The WHO Disability Assessment Schedule 2.0 (WHODAS The Patient Global Impression of Change measures mean- 2.0)53 is a 12-­item self-­report measure of the level of diffi- ingful change in overall status and the Pain Specific culty experienced taking into consideration how they Impression of Change60 measures the perception of pain usually do the activity. This includes the use of any devices improvement. The most frequently used scale is a 7-­point to assist them and/or the help of a person. In each item, numerical scale scored from 1= ‘much better’) to 7= individuals estimate the magnitude of the disability during ‘much worse’). the previous 30 days using a 5-­point scale scored from 1 (none) to 5 (extreme/cannot do). The total score ranges Ecological momentary assessment from 0 to 100. Higher scores reflect greater disability. The Pain intensity and other pain-­ related variables (eg, 12-­item WHODAS 2.0 has sound psychometric properties depressive-­ anxious symptoms and activity level) can in patients with FMS.54 fluctuate during the day and across days depending on The Generalized Anxiety Disorder 7-­Item Scale (GAD-­7)55 is personal and environmental factors. Collecting self-­ a 7-­item self-­report measure of pathological worry. Each reported data prospectively and closer in time to its item is scored from 0 (‘not at all’) to 3 (‘nearly every day’). occurrence substantially improves the accuracy, reliability The total score ranges from 0 to 21, where higher scores and quality of data. EMA has been successfully performed reflecting greater anxiety symptoms. The GAD-­ 7 has in patients with a variety of physical and mental prob- sound psychometric properties (α=0.92 and ICC=0.83) in lems.61 62 There is growing evidence indicating that patients with FMS in previous studies.56 well-­designed smartphone apps can be easy to use and well-­tolerated even in relatively old pain populations, with Other measures compliance rates as high as 85%.61 In this RCT, we will The ACTTION AE is a reporting checklist used to measure use the Pain Monitor (Monitor de Dolor, by its Spanish safety and benefit-­risk of a clinical trial.57 The Safety and name) app63 to assess a wide range of variables (see items Benefit-­Risk Reporting and Evaluation Working Group of in table 2) twice a day (once in the morning and once in the Analgesic, Anaesthetic, and Addiction Clinical Trial Translations, Innovations, Opportunities, and Networks (ACTTION; http://www.acttion.org) public–private part- Table 2 List of items administered via Pain Monitor app nership with the FDA developed an AE reporting check- Items Morning Evening list that will be used in the present study. The EuroQoL (version EQ-­5D-­5L)58 is a health-­related Pain intensity X X quality of life questionnaire that consists of two parts. Fatigue X X In the first one, the individual’s difficulties concerning Perceived control over pain X X mobility, self-­care, pain/discomfort and anxiety/depres- Depression X X sion are evaluated. In the second part, the perceived Anxiety X X health is assessed by means of a Visual Analogue Scale Stress X X ranging from 0 to 100. The EQ-­5D-­5L scores will be used to calculate the quality-­adjusted life years (QALYs) during Sleep disturbance X   the follow-­up period by adjusting the duration of time Activity level   X affected by the health outcome by the value of the utility. Interference with leisure activities   X The Client Service Receipt Inventory (CSRI)59 is a self-­ Interference with work-­related activities   X report tool used to collect retrospective data on medi- Adverse effects   X cation consumption and service receipt. Patients are asked to bring their daily medication prescriptions and Rescue medications   X information about pain-­related drugs (analgesics, anti-­ The Pain Monitor app automatically informs patients when to inflammatories, opioids, muscle relaxants, antidepres- respond (by default, at 11:00 and 19:00) using a push notification sants, etc) is recorded. This includes the name of the system, but patients can respond with a margin of 2 hours from drug, the dosage, total number of prescription days and given times. Collected data are stored on a secure server at the Jaume I University, Spain. daily dosage consumed. Concerning service receipt, 6 Colomer-­Carbonell A, et al. BMJ Open 2022;12:e055351. doi:10.1136/bmjopen-2021-055351 Open access the evening, at convenient times along the week) during algebra (DARTEL). We will use both unmodulated and 120 days. The app and the data will be stored on different modulated grey matter images to convey complementary servers with different domain names and connected volumetric information. We will use FreeSurfer ENIGMA locally only (the server containing the data does not have pipelines to perform the VBM. In addition, we will quan- Internet access). According to a recent meta-­analysis,64 tify metabolites concentrations using LCModel (V.6.3-­1J). EMA-­completion rates are higher among elderly patients We will only include high-­ quality spectra, defined as compared with younger patients. Considering that the signal-­to-­noise ratio >15, Cramer-­Rao lower bounds< 15% majority of FMS patients in our study are not expected to and full width at half-­ maximum of metabolites <0.07. be young and that the EMA item battery does not require The spectroscopy analysis will account for the effects of a long response time (<1 min), it is expected not to place cerebrospinal fluid and grey matter within the voxel, and an excessive burden on participants. interindividual differences in cortical grey matter. Inflammatory biomarkers Statistical analysis After obtaining the blood sample, it will be allowed to The main analysis will compare the effect of LDN versus coagulate for a minimum of 30 min at room tempera- placebo on the primary outcome (pain intensity at T1). ture. It will then be centrifuged for 10 min at 1000g. The Data analyses will be performed following an intention-­to-­ resulting serum will be stored at −80°C during the same treat plan. Then, we will compute analysis of the primary morning of extraction until it is ready to be analysed. All outcome (at T2 and T3) and analysis of the secondary samples (at T0 and T1) will be analysed in a single analyt- outcomes at T1, T2 and T3. Linear mixed models will be ical batch to reduce interassay variability (approximately created using the restricted maximum likelihood method 15%). The serum levels of IL-­1β, sIL-­1ra, IL-­4, IL-­6, sIL-­6r, for the estimation of parameters. The effect sizes will be sgp130, CXCL-­8, IL-­10, IL-­17, TNF-α and high-­sensitivity calculated according to Cohen’s d. An interim analysis is (hs) C reactive protein will be evaluated.29 For the quan- planned at T1 once 50% of the total sample has been eval- tification of the cytokines, the Milliplex reagents from uated. A 5% significance level will be used in all two-­tailed the company MerckMillipore will be used and analysed tests, applying the Benjamini-­ Hochberg correction for using a Luminex platform. The high sensitivity multiplex multiple comparisons. Additionally, to make the findings kit will be used: Human High Sensitivity T Cell, cata- from our study clinically meaningful, the number needed logue number: HSTCMAG28SPMX11, adapted to the to treat will be reported. For this analysis, we will dichot- aforementioned cytokines. The hs-­PCR will be quantified omise participants into responders or non-­ responders using turbidimetry in an Olympus AU5400 auto-­analyser. using two different cut-­off criteria in compliance with the These biomarkers will only be evaluated at baseline (T0) Initiative on Methods, Measurement, and Pain Assess- and 3 months (T1) for the following reasons: (1) there is ment in Clinical Trials (IMMPACT) recommendations45: evidence of significant inflammatory changes at 8 weeks at least 50% pain relief over baseline (substantial benefit) with LDN37; (2) this results in lower risk of dropout (vs or 30% or more pain relief (moderate benefit). For these evaluating them at 6 or 12 months); (3) conducting at analyses, we will use SPSS V.26 (IBM Corp). least two measures allows to use the change between Regarding EMA, a recent recommended approach is baseline and 3 months as a mediator of long-­term clinical ‘network analysis’. There has been burgeoning interest in changes; and (4) budget constraints. conceptualising chronic pain as a network of interacting symptoms and psychobiological processes.65 Network Neuroimaging analysis will offer us a good chance to quantify and visu- The scans (protocol duration: approximately 30 min) will alise relationships between pain intensity and pain-­related be performed in a Phillips Ingenia 3T MRI scanner with a variables (eg, depression, anxiety, fatigue, sleep distur- 32-­channel head coil at Hospital Sant Joan de Déu (Esplu- bance). We will estimate temporal networks by means gues de Llobregat, Spain). To examine cingulate, insular, of vector autoregression techniques66; these ‘temporal amygdalar, occipital, angular, parahippocampal and networks’ would indicate potential causality with one or prefrontal grey matter volume, we will use voxel-­based more variables preceding one or more variables in time. morphometry (VBM). We will also use surface-­ based Network analysis will be performed with the free statis- morphometry (with FreeSurfer calculation of cortical tical software JASP.67 thickness, surface area and local gyrification index) for In economic evaluation, it is important to calculate the examining cortical abnormalities. Additionally, Glu, relationship between the costs of each treatment and its glutamine, myo-­inositol, N-­acetylaspartate, choline and consequences in the form of QALYs, an index measure creatine (and creatinine ratios) levels will be analysed designed to assess both quantity of life (years) and health-­ using magnetic resonance spectroscopy. Specifically, we related quality of life. A year lived with the maximum quality will conduct the following processing for the regions of of life would be transformed into 1 QALY; a year lived with interest according to the corresponding hypotheses. For half the maximum quality of life would be transformed VBM, we will apply a bias field correction, tissue segmen- into 1/2 QALY. This relative value is called the incremental tation with SPM12 plus normalisation with Diffeomor- cost–utility ratio (ICUR) and it expresses the relationship phic Anatomic Registration Through Exponentiated Lie between the costs and the effects of one option compared Colomer-­Carbonell A, et al. BMJ Open 2022;12:e055351. doi:10.1136/bmjopen-2021-055351 7 Open access with another. The QALYs obtained in the 12 months after underpinnings of LDN in patients with FMS. This manu- the treatment onset will be calculated by the area under the script presents the design and rationale of a randomised, curve. The direct costs will be calculated by adding together double-­blinded, placebo-­controlled phase III study, which the costs derived from the medication and the use of the is a powerful design to assess the efficacy of LDN. We have health services. The cost of medications will be calculated by decided to administer 4.5 mg/day of LDN in this RCT multiplying the price per milligram by the total daily dose because this dose seems to provide an optimal balance consumed (in milligrams) and the number of days that the between significant analgesic efficacy and minimal side treatment is received. The cost arising from the use of the effects (nausea, sleep disturbance, nightmares, etc) health services (primary care, specialist and accident and according to a recent study.68 emergency consultations, and hospital admissions) will be Our findings using this design in conjunction with those obtained from the clinical electronic records (http://www.​ that will be obtained in another ongoing RCT that is being oblikue.com/en/esalud.html). The indirect costs will be carried out in Denmark (The FINAL study)69 may facilitate calculated based on the days off work, which will be multi- the approval of the first drug indicated for the treatment plied by the official minimum wage during the study period. of FMS in Europe. The FINAL study is an ongoing single-­ The effect of the treatments will be estimated using ordi- centre, randomised, double-­blinded, placebo-­controlled nary least squares multivariate regression, adjusting for the trial that is being carried out in Odense (Denmark). A baseline differences between groups. In order to manage total of 100 women between 18 and 64 years old with FMS uncertainty in the sampling distribution of the ICUR, non-­ will take either LDN or placebo for 3 months. Besides self-­ parametric bootstrapping will be applied, with 1000 repli- report measures, Danish researchers will also examine cations in each comparison. Cost–utility analyses will be the levels of pro-­ inflammatory and anti-­ inflammatory conducted with STATA V.16.0 (StataCorp). cytokines. If our respective findings strongly differ in effi- cacy or safety, we might analyse which factors can account Patient and public involvement for the divergence and plan a multicountry confirma- Patients and the public will not be involved in the design, tory trial with an agreed design and methodology. As conduct, reporting or dissemination of our research. pointed out by Kim and Fishman,70 a common problem with a generic, compounded medication is the lack of commercial support for research. To begin studies such ETHICS AND DISSEMINATION as INNOVA or FINAL, it is crucial to have the synergistic All procedures performed in this study will be in accor- support from public funding bodies, private entities and dance with the 1964 Helsinki declaration and its last commercial companies. This has been the case in the amendments (7th revision, adopted by the 64th World present study, with different public and private organisa- Medical Association General Assembly, Fortaleza, Brazil). tions providing economic and logistic support. Signed informed consent will be obtained from all The inclusion of brain and blood immune biomarkers patients once they have been informed of the study proce- will allow us to determine whether LDN modulates dures, potential risks, and their right to withdraw at any neuroinflammatory processes involving inflammatory time from the RCT. The FSJD Ethics Committee Board cells such as glial cells. These markers will also allow us evaluated and approved the study protocol in June 2021 to explore the ‘hormetic’ effects of the drug, that is, if (PIC-­178-­19). Only the principal investigators (AR-­S and a low dose of an antagonist (naltrexone) may paradoxi- JVL) will have full access to the final trial dataset. Modifi- cally act as an agonist of the endogenous opiate system. As cations in the study protocol will be reported to the FSJD explained above, it is posited that LDN mainly acts as an Ethics Committee Board as well as the independent CRO. immunomodulatory drug via blockade of TLR-­4, which Once the RCT is completed, we will publish our results provides a therapeutic pathway to reduce activation of in international peer-­ reviewed biomedical journals and the inflammatory cascade and the nociceptive system.71 present them at national and international conferences. In a recent pilot study,72 patients with opioid induced Authorship will be assigned in accordance with the Interna- hyperalgesia and patients with FMS were treated with tional Committee of Medical Journal Editors guidance. In LDN for 3 months. Via different mechanisms of action, addition, we will send participating patients a short report of LDN improved pain tolerance (measured with the cold our findings. A copy of the report will also be sent to Insti- pressor test) in both groups of patients, being the effect tute of Health Carlos III (main funding body). The principal even stronger in those participants with opioid induced investigators will organise an end-­of-­study seminar. The main hyperalgesia. According to the authors, the mechanism objective of this activity will be to share the study findings of action that would explain the beneficial effects of LDN with stakeholders to discuss how to maximise uptake of the for FMS may be transient blockade of the opioid growth findings in patient treatment and clinical practice, and to factor receptor. determine future research directions. Obtaining empirical evidence for cost–utility of treat- ments or interventions is required by the Ministry of DISCUSSION Health in Spain for reimbursement. In Spain, a threshold As far as we know, no RCT has been published about of €22 000–25 000 per QALY gained is found to be consis- the efficacy, safety, cost–utility and neurobiological tent with decisions of adopting new technologies by the 8 Colomer-­Carbonell A, et al. BMJ Open 2022;12:e055351. doi:10.1136/bmjopen-2021-055351 Open access National Health Service.73 To our knowledge, there is an Patient consent for publication Not applicable. absence of economic evaluations for LDN; therefore, an Provenance and peer review Not commissioned; externally peer reviewed. important feature of the present study is the cost–utility Open access This is an open access article distributed in accordance with the assessment of the drug. Creative Commons Attribution Non Commercial (CC BY-­NC 4.0) license, which FMS remains a chronic, debilitating and difficult to permits others to distribute, remix, adapt, build upon this work non-­commercially, and license their derivative works on different terms, provided the original work is manage condition for many individuals around the world. properly cited, appropriate credit is given, any changes made indicated, and the use After three decades of intensive research, the clinical is non-­commercial. See: http://creativecommons.org/licenses/by-nc/4.0/. benefits of pharmacological treatments remain unclear ORCID iDs and limited. This study will evaluate the analgesic effi- Xavier Borràs http://orcid.org/0000-0003-3972-1385 cacy, safety and cost–utility of LDN using a rigorous and Juan V Luciano http://orcid.org/0000-0003-0750-1599 powered design. If efficacious and cost-­ effective, LDN might be the first drug approved for FMS in Europe. Trial status REFERENCES This study is currently in the recruitment phase. The first 1 Häuser W, Ablin J, Fitzcharles M-­A, et al. Fibromyalgia. Nat Rev Dis Primers 2015;1:15022. patient will be enrolled in January 2022, and the study is 2 Fietta P, Fietta P, Manganelli P. Fibromyalgia and psychiatric expected to end in August 2024. disorders. Acta Biomed 2007;78:88–95. 3 Queiroz LP. Worldwide epidemiology of fibromyalgia. Curr Pain Headache Rep 2013;17:356. Confidentially 4 Briley M. Drugs to treat fibromyalgia - the transatlantic difference. Curr Opin Investig Drugs 2010;11:16–18. Personal data will be stored in accordance with the 5 Nüesch E, Häuser W, Bernardy K, et al. Comparative efficacy Spanish regulation guidelines for clinical research. of pharmacological and non-­pharmacological interventions in Participants will be allocated a unique identification fibromyalgia syndrome: network meta-­analysis. Ann Rheum Dis 2013;72:955–62. (ID) number at entry. The master list linking participant 6 Thieme K, Mathys M, Turk DC. Evidenced-­based guidelines on the personal information and ID number will be maintained treatment of fibromyalgia patients: are they consistent and if not, why not? have effective psychological treatments been overlooked? in a password-­protected hard drive at the Parc Sanitari J Pain 2017;18:747–56. Sant Joan de Déu (PSSJD; St. Boi de Llobregat, Spain). 7 Cagnie B, Coppieters I, Denecker S, et al. Central sensitization in Data will be stored for 10 years after study completion. fibromyalgia? A systematic review on structural and functional brain MRI. Semin Arthritis Rheum 2014;44:68–75. 8 Woolf CJ. Central sensitization: implications for the diagnosis and Author affiliations treatment of pain. Pain 2011;152:S2–15. 1 Teaching, Research & Innovation Unit, Parc Sanitari Sant Joan de Déu, St. Boi de 9 Napadow V, LaCount L, Park K, et al. Intrinsic brain connectivity in Llobregat, Spain fibromyalgia is associated with chronic pain intensity. Arthritis Rheum 2 2010;62:2545–55. Department of Basics, Developmental and Educational Psychology, Autonomous 10 Jensen KB, Berna C, Loggia ML, et al. The use of functional University of Barcelona, Cerdanyola del Vallès, Spain neuroimaging to evaluate psychological and other non-­ 3 Universidad de Ciencias Médicas de Villa Clara, Santa Clara, Cuba pharmacological treatments for clinical pain. Neurosci Lett 4 Department of Basic and Clinical Psychology and Psychobiology, Universitat Jaume 2012;520:156–64. I, Castello de la Plana, Spain 11 Zunhammer M, Schweizer LM, Witte V, et al. Combined glutamate 5 Biomedical Research Centre in Physiopathology of Obesity and Nutrition and glutamine levels in pain-­processing brain regions are associated with individual pain sensitivity. Pain 2016;157:2248–56. (CIBERobn), Madrid, Spain 12 Pyke TL, Osmotherly PG, Baines S. Measuring glutamate levels in 6 Diagnostic Imaging Department, Hospital Sant Joan de Déu, Esplugues de the brains of fibromyalgia patients and a potential role for glutamate Llobregat, Spain in the pathophysiology of fibromyalgia symptoms: a systematic 7 review. Clin J Pain 2017;33:944–54. Personal Social Services Research Unit, London School of Economics and Political Science, London, UK 13 Napadow V, Harris RE. What has functional connectivity and 8 chemical neuroimaging in fibromyalgia taught us about the Department of Psychiatry, Chulalongkorn University, Bangkok, Thailand 9 mechanisms and management of 'centralized' pain? Arthritis Res University of Alabama at Birmingham, Birmingham, Alabama, USA Ther 2014;16:425. 10 Department of Clinical and Health Psychology, Autonomous University of 14 Fayed N, Andres E, Rojas G, et al. Brain dysfunction in Barcelona, Cerdanyola del Vallès, Spain fibromyalgia and somatization disorder using proton magnetic 11 Department of Rheumatology, Parc Sanitari Sant Joan de Déu, St. Boi de resonance spectroscopy: a controlled study. Acta Psychiatr Scand Llobregat, Spain 2012;126:115–25. 15 Fayed N, Garcia-­Campayo J, Magallón R, et al. Localized 1H-­NMR spectroscopy in patients with fibromyalgia: a controlled study of Contributors JVL, AR-­S and AF-­S conceived the study and revised the manuscript. changes in cerebral glutamate/glutamine, inositol, choline, and N-­ JVL drafted the manuscript. AC-­C, JPS-­M, HH-­N, XB, CS-­R, AG-­P, JorM, JosM, FD, acetylaspartate. Arthritis Res Ther 2010;12:R134. MM, JWY, AF-­S and AR-­S provided feedback on the manuscript, and all authors 16 Harris RE, Sundgren PC, Craig AD, et al. Elevated insular glutamate reviewed and approved the final version of the manuscript. in fibromyalgia is associated with experimental pain. Arthritis Rheum 2009;60:3146–52. Funding This study has been funded by the Institute of Health Carlos III (ISCIII; 17 Feraco P, Bacci A, Pedrabissi F, et al. Metabolic abnormalities in ICI20/00080; CPII19/00003) and has been co-­financed with European Union ERDF pain-­processing regions of patients with fibromyalgia: a 3T MR funds. JPS-­M has a PFIS predoctoral contract from the ISCIII (FI20/00034). AC-­C spectroscopy study. AJNR Am J Neuroradiol 2011;32:1585–90. has a FI predoctoral contract from AGAUR (FI_B/00216). AF-­S acknowledges the 18 Harris RE, Napadow V, Huggins JP, et al. Pregabalin rectifies aberrant funding from the Serra Húnter program (UAB-­LE-­8015). The ISCIII did not have any brain chemistry, connectivity, and functional response in chronic pain patients. Anesthesiology 2013;119:1453–64. role in the analysis and interpretation of data, in the writing of the manuscript, or in 19 Murga I, Guillen V, Lafuente J-­V. Cerebral magnetic resonance the decision to submit the paper for publication. changes associated with fibromyalgia syndrome. Med Clin Competing interests None declared. 2017;148:511–6. 20 Woolf CJ. Pain: moving from symptom control toward Patient and public involvement Patients and/or the public were not involved in mechanism-­specific pharmacologic management. Ann Intern Med the design, or conduct, or reporting, or dissemination plans of this research. 2004;140:441–51. Colomer-­Carbonell A, et al. BMJ Open 2022;12:e055351. doi:10.1136/bmjopen-2021-055351 9 Open access 21 Ji R-­R, Berta T, Nedergaard M. Glia and pain: is chronic pain a questionnaire in two Spanish samples. Arthritis Care Res gliopathy? Pain 2013;154:S10–28. 2013;65:1682–9. 22 Lynch MA. The multifaceted profile of activated microglia. Mol 48 Lovibond PF, Lovibond SH. The structure of negative emotional Neurobiol 2009;40:139–56. states: comparison of the depression anxiety stress scales (DASS) 23 Watkins LR, Hutchinson MR, Ledeboer A, et al. Norman Cousins with the Beck depression and anxiety inventories. Behav Res Ther Lecture. Glia as the "bad guys": implications for improving clinical 1995;33:335–43. pain control and the clinical utility of opioids. Brain Behav Immun 49 Alok R, Das SK, Agarwal GG, et al. Relationship of severity of 2007;21:131–46. depression, anxiety and stress with severity of fibromyalgia. Clin Exp 24 Kettenmann H, Hanisch U-­K, Noda M, et al. Physiology of microglia. Rheumatol 2011;29:S70–2. Physiol Rev 2011;91:461–553. 50 Bados A, Solanas A, Andrés R. Psychometric properties of the 25 Dantzer R. Cytokine, sickness behavior, and depression. Immunol Spanish version of depression, anxiety and stress scales (DASS). Allergy Clin North Am 2009;29:247–64. Psicothema 2005;17:679–83. 26 Haroon E, Miller AH, Sanacora G. Inflammation, glutamate, and 51 Kratz AL, Schilling SG, Goesling J, et al. Development and initial glia: a trio of trouble in mood disorders. Neuropsychopharmacology validation of a brief self-­report measure of cognitive dysfunction in 2017;42:193–215. fibromyalgia. J Pain 2015;16:527–36. 27 Albrecht DS, Forsberg A, Sandström A, et al. Brain glial activation 52 Feliu-­Soler A, Pérez-­Aranda A, Andrés-­Rodríguez L, et al. Digging in fibromyalgia - A multi-­site positron emission tomography into the construct of fibrofog: psychometric properties of the Spanish investigation. Brain Behav Immun 2019;75:72–83. version of the multidimensional inventory of subjective cognitive 28 Niraula A, Sheridan JF, Godbout JP. Microglia priming with aging and impairment in patients with fibromyalgia. J Appl Biobehav Res stress. Neuropsychopharmacology 2017;42:318–33. 2018;23:e12134. 29 Andrés-­Rodríguez L, Borràs X, Feliu-­Soler A, et al. Peripheral 53 Vázquez-­Barquero JL, Vázquez Bourgón E, Herrera Castanedo immune aberrations in fibromyalgia: a systematic review, meta-­ S, et al. [Spanish version of the new World Health Organization analysis and meta-­regression. Brain Behav Immun 2020;87:881–9. Disability Assessment Schedule II (WHO-­DAS-­II): initial phase of 30 Bäckryd E, Tanum L, Lind A-­L, et al. Evidence of both systemic development and pilot study. Cantabria disability work group]. Actas inflammation and neuroinflammation in fibromyalgia patients, as Esp Psiquiatr 2000;28:77–87. assessed by a multiplex protein panel applied to the cerebrospinal 54 Smedema SM, Yaghmaian RA, Ruiz D. Psychometric validation fluid and to plasma. J Pain Res 2017;10:515–25. of the world health organization disability assessment schedule 31 Kiguchi N, Kobayashi Y, Kishioka S. Chemokines and cytokines in 2.0-­12-­item version in persons with fibromyalgia syndrome. J Rehabil neuroinflammation leading to neuropathic pain. Curr Opin Pharmacol 2016;82:28–35. 2012;12:55–61. 55 Spitzer RL, Kroenke K, Williams JBW, et al. A brief measure for 32 Miller AH, Haroon E, Raison CL, et al. Cytokine targets in the brain: assessing generalized anxiety disorder: the GAD-­7. Arch Intern Med impact on neurotransmitters and neurocircuits. Depress Anxiety 2006;166:1092–7. 2013;30:297–306. 56 Robinson RL, Kroenke K, Williams DA, et al. Longitudinal observation 33 Younger J, Parkitny L, McLain D. The use of low-­dose naltrexone of treatment patterns and outcomes for patients with fibromyalgia: (LDN) as a novel anti-­inflammatory treatment for chronic pain. Clin 12-­month findings from the REFLECTIONS study. Pain Med Rheumatol 2014;33:451–9. 2013;14:1400–15. 34 Patten DK, Schultz BG, Berlau DJ. The safety and efficacy of low-­ 57 Smith SM, Wang AT, Katz NP, et al. Adverse event assessment, dose naltrexone in the management of chronic pain and inflammation analysis, and reporting in recent published analgesic clinical in multiple sclerosis, fibromyalgia, Crohn's disease, and other chronic trials: ACTTION systematic review and recommendations. Pain pain disorders. Pharmacotherapy 2018;38:382–9. 2013;154:997–1008. 35 Younger J, Mackey S. Fibromyalgia symptoms are reduced by low-­ 58 Herdman M, Gudex C, Lloyd A, et al. Development and preliminary dose naltrexone: a pilot study. Pain Med 2009;10:663–72. testing of the new five-­level version of EQ-­5D (EQ-­5D-­5L). Qual Life 36 Younger J, Noor N, McCue R, et al. Low-­dose naltrexone for the Res 2011;20:1727–36. treatment of fibromyalgia: findings of a small, randomized, double-­ 59 Vázquez-­Barquero JL, Gaite L, Cuesta MJ. Spanish version of the blind, placebo-­controlled, counterbalanced, crossover trial assessing CSRI: a mental health cost evaluation interview. Arch Neurobiol daily pain levels. Arthritis Rheum 2013;65:529–38. 1997;60:171–84. 37 Parkitny L, Younger J. Reduced pro-­inflammatory cytokines after 60 Scott W, McCracken LM. Patients’ impression of change following eight weeks of low-­dose naltrexone for fibromyalgia. Biomedicines treatment for chronic pain: global, specific, a single dimension, or 2017;5:16. many? J Pain 2015;16:518–26. 38 Chan A-­W, Tetzlaff JM, Altman DG, et al. SPIRIT 2013 statement: 61 May M, Junghaenel DU, Ono M, et al. Ecological momentary defining standard protocol items for clinical trials. Ann Intern Med assessment methodology in chronic pain research: a systematic 2013;158:200–7. review. J Pain 2018;19:699–716. 39 Schulz KF, Altman DG, Moher D, et al. CONSORT 2010 statement: 62 Colombo D, Fernández-Álvarez J, Patané A, et al. Current state updated guidelines for reporting parallel group randomised trials. and future directions of technology-­based ecological momentary BMJ 2010;340:c332. assessment and intervention for major depressive disorder: a 40 Husereau D, Drummond M, Petrou S, et al. Consolidated health systematic review. J Clin Med 2019;8:465. economic evaluation reporting standards (CHEERS) statement. BMC 63 Suso-­Ribera C, Castilla D, Zaragozá I, et al. Validity, reliability, Med 2013;11:80. feasibility, and usefulness of pain monitor: a multidimensional 41 Fayed N, Olivan-­Blázquez B, Herrera-­Mercadal P, et al. Changes smartphone APP for daily monitoring of adults with heterogenous in metabolites after treatment with memantine in fibromyalgia. A chronic pain. Clin J Pain 2018;34:900–8. double-­blind randomized controlled trial with magnetic resonance 64 Ono M, Schneider S, Junghaenel DU, et al. What affects the spectroscopy with a 6-­month follow-­up. CNS Neurosci Ther completion of ecological momentary assessments in chronic pain 2014;20:999–1007. research? An individual patient data meta-­analysis. J Med Internet 42 Galvez-­Sánchez CM, Reyes Del Paso GA. Diagnostic criteria for Res 2019;21:e11398. fibromyalgia: critical review and future perspectives. J Clin Med 65 Gómez Penedo JM, Rubel JA, Blättler L, et al. The complex interplay 2020;9:1219. of pain, depression, and anxiety symptoms in patients with chronic 43 Häuser W, Jung E, Erbslöh-­Möller B, et al. Validation of the pain: a network approach. Clin J Pain 2020;36:249–59. fibromyalgia survey questionnaire within a cross-­sectional survey. 66 Bringmann LF, Vissers N, Wichers M, et al. A network approach to PLoS One 2012;7:e37504. psychopathology: new insights into clinical longitudinal data. PLoS 44 Carrillo-­de-­la-­Peña MT, Triñanes Y, González-­Villar A, et al. One 2013;8:e60188. Convergence between the 1990 and 2010 ACR diagnostic criteria 67 Love J, Selker R, Marsman M, et al. JASP: Graphical statistical and validation of the Spanish version of the fibromyalgia survey software for common statistical designs. J Stat Softw questionnaire (FSQ). Rheumatol Int 2015;35:141–51. 2019;88:1–17. 45 Dworkin RH, Turk DC, Wyrwich KW, et al. Interpreting the clinical 68 Bruun-­Plesner K, Blichfeldt-­Eckhardt MR, Vaegter HB, et al. Low-­ importance of treatment outcomes in chronic pain clinical trials: dose naltrexone for the treatment of fibromyalgia: investigation of IMMPACT recommendations. J Pain 2008;9:105–21. dose-­response relationships. Pain Med 2020;21:2253–61. 46 Bennett RM, Friend R, Jones KD, et al. The revised fibromyalgia 69 Bruun KD, Amris K, Vaegter HB, et al. Low dose naltrexone for the impact questionnaire (FIQR): validation and psychometric properties. treatment of fibromyalgia: protocol for a double-­blind, randomized, Arthritis Res Ther 2009;11:R120. placebo-­controlled trial. Trials 2021;22:804. 47 Luciano JV, Aguado J, Serrano-­Blanco A, et al. Dimensionality, 70 Kim PS, Fishman MA. Low-­dose naltrexone for chronic pain: update reliability, and validity of the revised fibromyalgia impact and systemic review. Curr Pain Headache Rep 2020;24:1–8. 10 Colomer-­Carbonell A, et al. BMJ Open 2022;12:e055351. doi:10.1136/bmjopen-2021-055351 Open access 71 Hatfield E, Phillips K, Swidan S, et al. Use of low-­dose naltrexone in 73 Vallejo-­Torres L, García-­Lorenzo B, Serrano-­Aguilar P. Estimating the management of chronic pain conditions: a systematic review. a cost-­effectiveness threshold for the Spanish NHS. Health Econ J Am Dent Assoc 2020;151:891–902. 2018;27:746–61. 72 Jackson D, Singh S, Zhang-­James Y, et al. The effects of low dose naltrexone on opioid induced hyperalgesia and fibromyalgia. Front Psychiatry 2021;12:593842. Colomer-­Carbonell A, et al. BMJ Open 2022;12:e055351. doi:10.1136/bmjopen-2021-055351 11